Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 Future Research Needs In the brief period since the first descriptions of HIV and its unambig- uous identification as the cause of AIDS, a tremendous amount has been learned about the genetic structure and transmission of the virus. Much less is known, however, about how it initiates infection, how it maintains infection, and what determines the progression and diversity of the resulting illness. Research has been very effective in discovering the routes of viral transmission, enabling public health and education programs to be designed that incorporate increasingly accurate and specific information. Research has also been particularly effective in elucidating the complete genomic structure of the virus, allowing definition of many, if not all, of the virus's genes. Such insights, however impressive, are only the beginning of what promises to be a long and difficult path toward effective therapeutic interventions to minimize or eliminate the debilitating effects of HIV infection and toward limiting the spread of the virus by means of safe and effective vaccines. This chapter summarizes some of the opportunities and obstacles that will be encountered along that path. In many areas, predictions of progress are difficult to make. It is easier, however, to specify the mechanisms that will facilitate that progress. Successful development of vaccines or drugs to modify the prevalence or consequences of HIV infection will be greatly aided by a substantially improved basic under- standing of the virus, of the functioning of the healthy and impaired 177
178 CONFRONTING AIDS human immune system, and of their interaction in the progression from infection to disease. The progress achieved to date in identifying and characterizing the causative agent of AIDS would not have been possible without the scientific and medical knowledge achieved over the years through the pursuit of basic biomedical research. In that pursuit, the investigator is rarely certain of when or if research findings will be applicable to a disease. The instance of AIDS exemplifies the value of basic research, however, in that the current understanding of AIDS is based on knowl- edge derived largely from studies carried out before AIDS was even recognized as a disease. The unusual speed with which the etiologic agent of AIDS was isolated and remarkably well characterized was heavily dependent upon 20 years of investment in molecular biology and virology. Continued support of the medical and scientific communities in their pursuit of basic knowledge related to HIV infection and AIDS will provide an essential adjunct to the necessary applied studies. These basic and applied studies can be expected to prove mutually beneficial as increased under- standing of the mechanisms and consequences of HIV infection is translated into effective interventions to limit the virus's impact. THE STRUCTURE AND REPLICATION OF HIV Retroviral Structure The molecular cloning and nucleotide analysis of a number of indepen- dent isolates of HIV have completely defined a retroviral genomic structure (see Figure 6-1) of unprecedented complexity and marked diversity (Meusing et al., 1985; Ratner et al., 1985; Sanchez-Pescador et al., 1985; Wain-Hobson et al., 19851. While HIV shares some genetic and structural elements with other known retroviruses, it possesses distinc- tive features that have not been observed previously. The replication cycles of all previously known retroviruses depend on the functions of the protein products encoded by three viral genes termed gag, pol, and env (Weiss et al., 19851. These genes specify the structural and enzymatic functions required for viral infection and transmission and are situated in a common left-to-right (5' to 3') configuration in the retroviral genome. The gag gene encodes the proteins that constitute the internal core of the virion particle. The pol gene specifies the viral enzyme known as reverse transcriptase, which is responsible for synthesizing a DNA copy of the retroviral RNA genome early after infection. The gag and pot proteins are first synthesized as a large precursor, which is then cleaved by a virus-encoded protease to give the final proteins. The ens gene codes for the surface envelope proteins of the retrovirus, which mediate the process of
FUTURE RESEARCH NEEDS 179 sor ~-tat-3 . 3'-orf E] art -- ~ gag pol ,; ] env . - - genes for vinon proteins ~3 ~ ~ - - - genes for regulator proteins row ~ FIGURE 6-1 HIV genome. Source: Courtesy of Howard Temin, University of Wisconsin School of Medicine, Madison. virus binding to the surface membranes of host target cells. The termini of the DNA form of the retroviral genome are provided by repetitive sequences known as long terminal repeats (LTRs), which contain the essential genetic regulatory elements controlling viral expression and integration. With these three genes and the additional regulatory sequences in their LTRs, many animal retroviruses are fully competent to replicate in an appropriate host target cell. HIV, however, contains a minimum of four additional genes (see Figure 6-1), at least two of which are also function- ally required in its replication cycle. Because they have been indepen- dently described in a number of laboratories, these novel genes carry a multitude of designations. A gene known as tat-III serves a necessary function in HIV replication by controlling, in a trans-acting fashion (i.e., at a distance by means of a diffusible product), the level of expression of the other viral genes (Arya et al., 1985; Sodroski et al., 19841. The most recently discovered HIV gene, variously known as art or trs, is thought to control, in a trans-acting manner, the differential expression of viral structural and regulatory functions; it is also necessary for viral replica- tion (Feinberg et al., in press; Sodroski et al., 1986b). Two other viral genes, named sor (also known as orf-l, P', or Q) and 3'-orf (also known as orf-2, E', or F), serve unknown functions in the life-cycle of HIV, although they are known to be expressed. They are apparently not needed for replication in tissue culture cells (Allen et al., 1985; Fisher et al., 1986b; Kan et al., 1986; Lee et al., 1985; Sodroski et al., 1986a). Retroviral Replication Retroviral infection (see Figure 6-2) is initiated by the binding of a virus particle to a specific receptor molecule expressed on the surface of an
~ 80 CONFRONTING AIDS \ ENTRANCE ~REVERSE RCUlARIZATI INTEGRATION ~ TRANSCRIPTION ~) \ ~ N TRANSCRIPTION /& k ~ I TRANSPORT to tat ENCAPSI DAT10~ L_/ BUDDING Wit gag-pol TRANSLATION & PROCESSING . . env @) FIGURE 6-2 Life-cycle of HIV. Source: Courtesy of Howard Temin, University of Wisconsin School of Medicine, Madison. appropriate target cell (Weiss et al., 19851. After binding takes place, the retrovirus enters the cell and uncoats in the host cell's cytoplasm. The retroviral genetic information contained in its single-stranded RNA genome is then transferred to a full-length linear duplex DNA intermedi- ate by the synthetic activities of the reverse transcriptase enzyme, which accompanied it in the virion particle. This linear DNA intermediate is transported to the nucleus, where it is circularized before becoming stably integrated into the DNA of the host cell. The process of integration is thought to involve the specific interaction between retroviral sequences at the edges of the LTRs and an additional enzymatic function known as the integrase encoded by the pol gene. Once integrated into the host chromosome, the retroviral genome is termed a provirus. There it serves as a template for RNA transcription, the primary product of which is a full-length viral RNA molecule. The gag and pot products are translated from the full-length transcript, while a portion of the viral transcript undergoes RNA splicing to yield an
FUTURE RESEARCH NEEDS i~ ~ envelope-coding mRNA from which the gag and pol sequences have been deleted. The mRNAs encoding the sor, tat-III, art, and 3'-orf proteins of HIV are derived from the genomic RNA transcript via complex splicing events (Arya et al., 1985; Muesing et al., 1985; Rabson et al., 19851. The virus's structural and regulatory proteins are then synthesized in the cytoplasm. Following secondary processing, the constituents of the virions (gag, pot, and envelope proteins) proceed to assemble in the proximity of the cellular plasma membrane into virus particles that have incorporated the full-length viral RNA genome. The retroviral membrane is derived directly from the cellular plasma membrane as the virion is released from the cell by a process known as budding. The development of specific antiviral therapies for HIV infection will depend upon identifying and interfering with critical stages of the retro- viral life-cycle. Although the replicative mechanisms of HIV have not yet been studied in great detail, many of its essential processes may be understood through analogy with other, more thoroughly analyzed exam- ples of retroviruses. It should not be assumed, however, that HIV follows a pattern of replication identical to those previously elucidated in other retroviral systems. Indeed, some significant differences have already been identified. As discussed below, these may provide additional targets for future antiviral strategies. Definition of the Structural and Functional Constituents of HIV The protein products from all seven of the genes so far identified in HIV isolates have been recognized by antibodies from persons infected with HIV. This has permitted the initial identification of these proteins and has demonstrated their expression in viva. Some of these protein products have also been expressed in bacteria, yeast, or mammalian cells in vitro. The production of large quantities of all of the virus's genes in a biologically active form will provide necessary substrates for structural and functional analyses. In addition to the study of such recombinant DNA products, it will be important to directly study the native proteins as they exist in the virion and in infected cells. Their directly determined amino acid sequences (as opposed to computer translations from the DNA sequence), and the nature and location of posttranslational modifi cations (e.g., glycosylation, phosphorylation, myristylation), should be ascertained. The nature of the proteolytic cleavages and other post- translational processing involved in the synthesis of the virus's structural or functional components may thus be established, perhaps indicating new approaches for antiviral interventions.
I 82 CONFRONTING AIDS _ m0 , , Aged .' ENVELOPE ) L INTERNAL PROTEINS RNA GENOME FIGURE 6-3 Structure of the HIV virion. Source: Courtesy of Howard Temin, University of Wisconsin School of Medicine, Madison. Determination of the Structure of the HIV Virion Figure 6-3 shows a model of the structure of the HIV virion. Electron microscopic examinations of HIV particles show that it possesses an envelope with protruding spikes surrounding a central electron-dense core. The virion's spikes are formed by collections of the viral envelope glycoprotein, gpl20, which is anchored in the cell-derived plasma mem- brane through an attachment to the HIV transmembrane protein, gp41. The retroviral core is composed of collections of the gag proteins in association with the genomic RNA molecules, the reverse transcriptase, and other enzymes. Although there are generally accepted models for the structure of the interior of retrovirus virions, the models are conceptual in character and lack experimental validation (Weiss et al., 1982, 19851. Thus, the detailed structure of the potentially analogous virion of HIV is not clearly understood. Specifically, the locations and amounts of the various inter- nal proteins and the nature of their interactions are not known. Knowl- edge of these locations and interactions will be of great utility in drug design. For example, drugs that inhibit the formation of any of these complexes could block HIV replication. High-resolution structural determinations of the viral proteins, sepa- rately and in complexes, need to be performed; important candidates include the external portion of gpl20, the gpl20-gp41 complex, and the gpl20-gp41-internal protein complex. Similar structural studies have been performed with the hemagglutinin molecule of influenza virus and with the virions of rhinovirus 14 and poliovirus (Hoyle et al., 1985; Rossman et al., 1985; Wilson et al., 1981) and have provided an important foundation for understanding the interaction between infecting viruses and host cells and between viruses and neutralizing antibodies.
FUTURE RESEARCH NEEDS I 83 Interrupting Infection by HIV The progressive stages in the life-cycle of HIV present a number of opportunities for specific interruption. Much basic knowledge must be attained, however, before a rational approach to the development of prophylactic and therapeutic measures for HIV infection and AIDS will be feasible. A critical and early event in HIV infection involves the virus's attachment, via its envelope glycoprotein, to a receptor on the surface of a susceptible cell. The primary, if not exclusive, cellular receptor for HIV appears to be provided by the CD4 molecule (Dalgleish et al., 1984; Klatzmann et al., 19841. The CD4 molecule is expressed by helper/inducer T lymphocytes and by certain types of cells of the macrophage/monocyte lineage, a distribution that parallels the target cells for HIV infection. The expression of the CD4 molecule may completely explain the tissue tropism of HIV, although further documentation of the complete repertoire of the cell types that express the molecule is necessary (see section on "Natural History of HIV Infection," below). The interaction of the HIV envelope protein with the CD4 receptor can be inhibited by antibodies directed at specific determinants on either molecule (Robert-Guroff et al., 1985; Weiss et al., 19851. The structural definition of the molecular components involved in this specific recogni- tion process will be of central importance in the development of vaccines or other prophylactic measures to prevent HIV infection. The HIV envelope protein, through its specific interaction with the CD4 molecule, plays an important role in the cytopathic effect of viral infection on T lymphocytes (Lifson et al., 1986, in press; Sodroski et al., 1986a). The specific inhibition of this interaction, if attainable, may ameliorate the immune deficiency that follows HIV infection. The HIV envelope glycoprotein is unlike the similar components of most other retroviruses, both in its large size and in the extent and pattern of its sequence variability (Coffin, 19861. The primary translation product of the HIV envelope gene is heavily glycosylated during the course of its maturation. The extensive variation that has been observed in the nucleotides (and thus in the predicted amino acid sequences) among independent isolates of HIV is of important theoretical and practical concern in understanding the processes of viral replication and in devel- oping an effective vaccine. Such variation, though striking, is not unex- pected in a virus whose genome is composed of a single strand of nucleic acid. High mutation rates may be an unavoidable consequence of the replication of such genomes. In addition, genomic variation in retroviruses can be affected by the
~ 84 CONFRONTING AIDS insertion, duplication, and deletion of genetic sequences. HIV isolates show significant sequence divergence within the env gene by these mechanisms while their other genes are considerably better conserved between different viral isolates. The variability in the ens gene is concentrated within several "hypervariable" domains, which are inter- spersed between regions that are better conserved (Coffin, 19864. The mechanisms that generate this pattern of variability and conservation of envelope sequences will be important to elucidate, as will the functions of the variable and relatively constant regions. Because all HIV isolates apparently infect cells through the binding of the envelope protein to CD4, the domain of the viral envelope protein that facilitates binding to the cell surface receptor is presumably conserved between viral isolates. Inhibiting this recognition process through immu- nization or other approaches could in principle block infection by all viral isolates. It has been postulated, however, that the variable regions mask the essential constant regions from immunologic attack. Success in developing an HIV vaccine may thus be predicated on understanding and overcoming the problems presented by the genetic mutability and varia- tion of HIV. Following attachment to its cellular receptor, HIV enters the cell by a mechanism as yet poorly defined. The virus is internalized and uncoated, most probably through the normal absorptive endocytosis pathway used by animal cells to internalize cell surface receptors that have bound their respective ligand. As with other animal viruses, a low pH-mediated structural change in the gp 120 molecule in the endocytotic vesicles probably results in the actual uncoating and entrance of the viral genome into the cytoplasm. In the case of HIV, this process may or may not involve a membrane fusion event. Increased understanding of this pro- cess may allow the derivation of drugs that inhibit these early stages of . ~ . . . · ~ Injection In a v~rus-spec~nc manner. Once the retroviral particle has uncoated in the cytoplasm, the critical and characteristic process of reverse transcription of the viral RNA genome into a double-stranded DNA copy ensues. The enzyme that catalyzes this process, reverse transcriptase, provides a specific and potentially very effective target for antiviral therapy. Inhibitors of HIV's reverse transcriptase comprise a number of the candidate drugs under current clinical and laboratory evaluation (see section on "Antiviral Agents," below). The protein components and processing pathways for the synthesis of HIV's reverse transcriptase are receiving substantial analytical attention. The forces that govern the HIV genome's migration to the nucleus and subsequent circularization are not known. The virus-specific integrase encoded by the pol gene is thought to be involved in the process of retroviral integration in other retroviruses. Unlike most
FUTURE RESEARCH NEEDS I 85 retroviruses, HIV shares with the other lentiviruses a pronounced ten- dency to accumulate unintegrated viral DNA in the course of infection in vitro and in vivo. It is thus possible that HIV does not always require integration for virus production to ensue. With the related lentivirus visna virus, dividing cells are reportedly not required for productive infection, although many other retroviruses demonstrate an obligate linkage be- tween cell division and viral integration (Haase, 19861. Whether cell division is required for HIV infection has not been established, but studies of viral production in macrophage cultures in vitro suggest that it may not be necessary (Gartner et al., 19861. The significance and origin of unintegrated HIV DNA are unclear, but they may have significant implications for understanding the cytopathic effects of the virus and for the potential of therapeutic agents to limit infection. Improved understanding of the requirements for and mecha- nisms of HIV integration, increasing availability of enzymatically active viral proteins involved in the process, and the development of cell-free systems for their assay will help in drug design and screening. Following integration into the host cell chromosome a mechanism that needs to be studied further the HIV proviral genome is transcribed into RNA by the cellular RNA polymerase. The potential host or viral factors that control the level of HIV expression are very poorly defined, but they may play a critical role in the persistence of HIV in the infected human host and the rate of immunologic compromise. It has been suggested that immunologic activation of infected T cells stimulates virus production (Hoxie et al., 1985; Zagury et al., 19861. The factors that activate provirus also need to be studied further. Subsequent to transcrip- tion, the processing of the HIV RNA transcripts also uses the host cell's machinery for capping, polyadenylation, and splicing. The tat-III gene is known to be essential for viral replication (Dayton et al., 1986; Fisher et al., 1986a), but there is much uncertainty about its precise mechanism of action. It appears to control the efficiency of translation of viral messages (Feinberg et al., in press; Rosen et al., 1986) and their stability, although effects on viral transcription have also been described. This gene operates through specific sequences contained in HIV mRNAs, and as there are no known host activities of similar character it may provide a specific target for drug attack. A second HIV gene has been described that appears to operate subsequent to the transcription of viral RNA. This gene, variously known as art (Sodroski et al., 1986b) or trs (Feinberg et al., in press), effectively modulates the specific viral mRNAs available for translation. The art (trs) product regulates the pattern of viral RNA expressed, either through differential splicing or specific message stabilization, and is essential for lIIV replication. Its exact role in the in vivo infectious process or
I 86 CONFRONTING AIDS pathology of HIV is unknown, but because it also appears to be an essential and virus-specific activity it is an important potential candidate for antiviral therapy. As with the other viral components, the large-scale preparation and widespread availability of the tat-III and art (trs) proteins would aid in the design of inhibitors for their essential viral functions. Presently, the limiting factors are understanding their actions and developing in vitro assays for their functions. The products of the HIV genes known as sor (orf-l, P', and Q) and 3'-orf (orf-2, E', and F) may not always be essential for virus replication in cell culture (Fisher et al., 1986b; Sodroski et al., 1986c). However, the continued presence of these open reading frames in HIV in the face of the high rate of mutation of retroviruses in general, and of HIV in particular, indicates that there has been strong selection for these genes in the virus infecting the human population. (With an estimated mutation rate of 1 alteration per 10,000 nucleotides per virus replication cycle, mutations would have appeared following only a few rounds of virus replication.) The importance of these genes is also suggested by the presence of antibodies against them in the sera of many HIV-infected persons. Their preservation and serologic recognition suggest that they serve an essential role in the in vivo expression of HIV. Improved understanding of the in vivo role of these novel open reading frames of HIV is extremely important, especially if they are involved in the biological processes underlying the persistent nature or cytopathic consequences of HIV infection. Identifying their functions and develop- ing in vitro assays for their measurement will be of great value in evaluating their candidacy as effective targets for pharmacologic inhibi- tion. If their functions are only evident in infected hosts, the development of animal models will be central to progress in this area (see section on "Animal Models," below). Viral precursor proteins are translated from viral mRNAs by use of host cell ribosomes and translation factors. The gag gene is translated through the synthesis of a protein precursor, Pr55, which is subsequently proteolytically processed into ply, p24, p9, and p7. The pol gene, which encodes the HIV reverse transcriptase, integrate, and probably protease, is thought to be initially translated into a polyprotein precursor, Prl50, which is then processed into p64/53, p22, and p34, respectively. The proteolytic cleavage of the gag and pot precursor proteins may be carried out by the virally encoded protease enzyme. If so, this cleavage might be subject to specific inhibition. The derivation and production of enzymatically active p22 protease would provide a very useful substrate for drug screening and drug design. The primary translation product of the ens mRNA is a protein of
FUTURE RESEARCH NEEDS I 87 approximately 90 kilodaltons (kd). During its transit to the cellular membrane, the envelope precursor is heavily glycosylated, which in- creases its apparent molecular weight to about 160 kd. The extent of this glycosylation of the HIV envelope protein is unprecedented among retroviruses. The observed relative conservation of glycosylation sites between divergent viral isolates suggests that glycosylation of the enve- lope protein plays an important biological role. Unlike that of many other retroviruses, the transmembrane protein (gp41) of HIV is also glycosyl- ated. Glycosylation could conceivably be an important determinant of the structure of the envelope, it could mask functionally important antigenic sites from human immune responses, or it could do both. The mature form of the viral envelope is achieved by the proteolytic cleavage of the gpl60 precursor to gpl20 and gp41. After the gpl20-gp41 complex is inserted in the plasma membrane by cellular processes, there is probably an aggregation of gpl20-gp41 molecules that excludes other cellular membrane proteins. The nature of the chemical bonds maintain- ing the stable interaction between the gpl20 and gp41 molecules should be determined as a possible specific target for drugs. Likewise, the orienta- tion of these proteins in the cell membrane and the virion particle has not been directly determined, but such information is important for drug design. HIV is formed by budding from a modified portion of the cell plasma membrane, during which the viral nucleoid assembles and organizes the gag proteins in association with copies of the genomic RNA and poly- merase components. The formation of the nucleoid involves the aggrega- tion of p24 gag molecules in a virus-specific process. There may be further protein cleavages after budding, a process referred to as maturation in the life-cycle of other retroviruses. As with other viruses the process of HIV assembly depends upon protein-protein interactions. The protein interactions in assembling virions have to be virus specific, otherwise virion production would be exceptionally inefficient or absent. As such, the process of HIV assembly may be subject to chemotherapeutic interference and hence merits further study. Interferons and related molecules have been demonstrated to inhibit the assembly and budding process in other retroviruses, including certain lentiviruses, and similar molecules may have relevance in com- bating HIV infection and spread (Narayan, 19861. Conclusions and Recommendations In the past few years the techniques of molecular biology have provided the starting materials for a detailed evaluation of the replicative pathways of HIV and for the development of therapeutic strategies to inhibit those
I88 CONFRONTING AIDS pathways. Although HIV was only discovered in 1983-1984, research has been so rapid and so successful that almost as much is now known about its molecular virology as about any other retrovirus. However, less is presently known about HIV-cell interactions than for many other retro- viruses. As the pathogenesis of AIDS is clearly related to the interaction of the causative virus with the susceptible host, at the levels of both the target cell and the host organism, much more must be learned about the biology of HIV infection. Although an empirical approach to the development of prophylactic and therapeutic measures for HIV infection and AIDS may work, rational strategies are more likely to succeed, and these will require a good knowledge base. Thus, most of the basic information on HIV and the consequences of its infection may be relevant to the prevention, control, and treatment of AIDS. Furthermore, much of the research on HIV will not only be relevant to AIDS but will increase understanding of the human immune system in health and disease, of pathogenic infections and cancers, and of the basic mechanisms controlling gene expression. · Success in the development of vaccines or drugs to prevent and treat AIDS will be facilitated by a greatly improved understanding of the basic biological processes and consequences of HIV infection. Rational ap- proaches to interfering with the mechanisms of HIV replication and spread present the most hopeful path to the development of effective antiviral agents but will require a substantial increase in the knowledge base concerning the genetic, structural, and biological characteristics of HIV and viruses apparently related to it, such as LAY-2 and HTLV-IV (see Chapter 21. Continuing efforts to fully determine the life-cycle and structure of HIV should be expanded and actively encouraged. The characterization of all viral proteins and their interactions with cellular proteins and processes should have a high priority. These studies should include structural (X-ray crystallographic) studies of HIV, reverse transcriptase, protease, integrate, glycoproteins, and regulatory proteins. · Strategies dedicated to prevention of the clinical manifestations of HIV infection and AIDS must also be predicated on a greatly improved understanding of the normal functioning of the human immune system (see section on "Natural History of HIV Infection," below). · Expansion of the research base on HIV is limited by the availability and adequacy of biological containment facilities, especially for virus production. Adequate financial support should be made available for expansion and creation of appropriate facilities. Coordinated and shared use of containment facilities should be a concern in the minimization of costs and maximization of access. · An extremely important aspect of the entire future AIDS research
FUTURE RESEARCH NEEDS I 89 effort is the wide and free distribution among scientists of various viruses, DNA clones, proteins, cell lines, and other reagents as they are created. This availability will speed the entry of researchers into the field and ensure compatibility and quick dissemination of results. · To facilitate the most productive pursuit of AIDS-related research, it is important to widen the pool of investigators and laboratories engaged in the analysis of HIV. Efforts must be undertaken to encourage an ever-increasing breadth and expertise in the area of AIDS research. · The benefits of the United States' traditional commitment to basic biomedical research are clearly demonstrated in the expeditious identifi- cation of the causative agent of AIDS, the derivation of a test to detect its presence, and the rapidly evolving understanding of its genetic constitu- tion and replicative requirements. The types of basic studies that permit- ted these impressive accomplishments may also be expected to yield valuable insights into therapeutic approaches to limit the establishment and progression of HIV infection. Thus, basic research studies in virology and immunology should be considered an important part of the AIDS research effort and should be fortified in the years ahead (see section on "Funding for Research Related to AIDS and HIV," below). · Given the current and expected level of public interest and anxiety relating to AIDS, scientists must recognize their special obligation in this situation to communicate their results in an accurate, responsible fashion. Biomedical scientists have an important role to play in the public educational efforts related to AIDS. However, the all too prevalent tendency toward simplistic interpretation or sensationalization of AIDS- related research in the media should actively be discouraged. Significant harm to the public's perception of the AIDS threat and to the credibility of the scientific community will also accrue if laboratory, clinical, or epidemiologic results are presented with exaggerated significance. NATURAL HISTORY OF HIV INFECTION Transmission of HIV For an HIV infection to become established, infectious virus must be delivered to a susceptible host. The presence of HIV has been reported in samples of peripheral blood, cell-free plasma, lymph nodes, bone marrow cells, cerebrospinal fluid, brain tissues, semen, cervical and vaginal secretions, lung tissue, saliva, and tears (Gartner et al., 1986; Ho et al., 1985c; Salahuddin et al., 1985; Shaw et al., 19851. Although termed "isolations," some of these reports have, in fact, been based only on transient measurements of reverse-transcriptase-like activities in a tissue culture sample. As such, they have not definitively documented the
190 CONFRONTING AIDS presence of infectious HIV. It is important to validate each of these reported sources of HIV isolation and to attempt to meaningfully estimate the resident viral load and the frequency with which the virus can be detected. Exposure to any fluid containing HIV could, in theory, present a risk for viral transmission. However, a substantial body of epidemiologic data indicates that the virus is spread only from a limited number of sources and received through a limited number of routes (see Chapter 21. These are sexual intercourse with an infected individual involving the exchange of semen or cervical/vaginal secretions, exposure to infected blood or blood products through therapeutic transfusion or the sharing of needles and other equipment that allow the passage of blood from one individual to another, and the perinatal passage of virus from infected mothers to their infants. The relative efficiencies of transmission of HIV from different sources and through different routes are not known in detail, but these are important questions for future epidemiologic and laboratory analyses. Levels of the dose of HIV needed for establishment of an infection may also vary with the mode of spread and the portal of entry. Such differences presumably underlie the apparent discrepancies that exist between the sources of viral isolations and the common routes of transmission that have been defined through epidemiologic studies. Certain host behaviors or factors have been suggested to enhance the transmission of HIV infection, although the presently available data are inadequate to allow certainty. For instance, studies of the risk of infection among African heterosexual populations have suggested that an increased efficiency of HIV spread is associated with the presence of other sexually transmitted diseases (see Chapter 2~. It is not known if there are important interactions occurring in a host during infection with different infectious agents, if local compromise or inflammation of the genital mucosa may be the factors that enhance the viral infection, or if these processes are coincidental. So-called "traumatic" sexual behaviors such as receptive anal inter- course have also been postulated to enhance HIV transmission by permitting the virus to enter the bloodstream through locally damaged blood vessels. While epidemiologic studies clearly show that receptive anal intercourse is associated with an increased risk of HIV infection, interpretation of this observation is difficult in light of the high prevalence of viral infection in the sexual partners of persons engaging in this practice. Recent reports of seroconversion in women who were artificially inseminated by an asymptomatic virus carrier argue against a requirement for "traumatic" sexual intercourse in the transmission of HIV (Stewart et al., 19851. Furthermore, in experimental animal systems, chimpanzees
FUTURE RESEARCH NEEDS 1191 can readily be infected by atraumatic vaginal inoculation with HIV (Fultz et al., 1986a). Thus, trauma may have no biological relevance in HIV transmission, and emphasizing it may obscure the demonstrated potential for both heterosexual and homosexual spread of the virus. However, this issue requires clarification. The relative efficiency of male-to-female compared with female-to-male transmission of HIV is not known, but these efficiencies will be crucial determinants of the rate of spread of HIV infection within the heterosexual population. The state of HIV actually involved in viral transmission, whether cell-free or cell-associated, has not been determined. It is a critical concern, however, for understanding HIV infection and for the develop- ment of an antiviral vaccine. Research on other lentiviruses suggests that viral transmission by cell-to-cell contact provides the major route of exposure for the previously uninfected host (Haase, 1986; Narayan, 19861. Moreover, very little, if any, cell-free virus can be detected in chronically infected animals. Unfortunately, a similar situation may prevail in the transmission of HIV, but this remains to be established. HIV infection of hemophiliacs by virally contaminated clotting-factor concentrates (see Chapter 2) suggests that cell-associated virus may not always be necessary, yet it may still provide the most frequent vehicle for transmission. The cell types that provide the earliest targets for HIV replication in the initiation of an infection are not known. Their location at the time of infection is also unknown, but both of these variables may depend on the route of initial exposure. Both T lymphocytes and macrophages that express the CD4 molecule provide susceptible host cells for HIV infection in vitro, and these may also be the initial in vivo targets for infection. Given the relative longevity and mobility of these cells, the HIV infection may then spread throughout the host. Systematic studies need to be conducted to determine the frequency with which virus can be isolated from various cells and tissues of persons who show antibodies to HIV. The Immune System Response to HIV Infection As described in Chapter 2, the earliest and most easily detectable marker for viral infection is the presence of antibodies reactive with HIV in the serum of an infected individual. However, a very small subset of infected individuals have been reported to remain seronegative for prolonged periods of time (Salahuddin et al., 19841. The available studies cannot establish the length of time that such a virus-positive antibody- negative state exists in those individuals, nor can they estimate the prevalence of such individuals in the population. Future studies must
192 CONFRONTING AIDS address these questions, especially since clarification of this problem is necessary for better protection of the blood supply (see Appendix C). Unlike many other viral infections of humans, antibody synthesis against HIV does not herald the clearing of the infection. With HIV, the virus can continue to be isolated from seropositive persons long after their initial exposure (Feorino et al., 1985; Jaffe et al., 19851. The lentiviruses of other animals are known to cause infections that persist throughout the lifetime of the infected host. Unfortunately, it appears that HIV may similarly establish a lifelong infection in humans. The presence of antibodies against HIV in the serum is thus not a marker of immunity but rather, in most instances, an indication of ongoing infection. It is not yet known if the immune response of some individuals, although unable to eliminate HIV infection, may protect them from the development of immunologic damage and disease. While there may be some change in the pattern of serologic reactivity with specific HIV protein components during the progression from infection to disease (see Appendix B), so far no specific patterns have been found to correlate with protection of health or development of disease. One aspect of the immune response critical to vaccine development efforts is the possible existence of specific antibodies to HIV that could prevent, limit, or eliminate infection by the virus. Such antibodies are referred to as neutralizing antibodies, and an important goal for candidate vaccine immunogens is to elicit these types of antibodies. Neutralizing antibodies have been found in some individuals over the course of their infection with HIV, but they are present in rather low concentrations (Robert-Guroff et al., 1985; Weiss et al., 1985) and appear incapable of curing an individual's infection. With HIV, neutralizing antibodies have been measured using in vitro assays that involve mixing the cell-free virus with a test serum and then evaluating the continued infectivity of the virus preparation by exposing susceptible target cells to it. If, as with other lentiviruses, HIV is spread by transmission from an infected cell to a normal cell, then the presently employed neutralization assays may have little relevance to the immuno- logic mechanisms that would effectively prevent viral infection. In humans infected with HIV, as in other animals infected with lentiviruses, the viral infection persists in the face of a neutralizing antibody response. Whether the currently used assays for HIV neutralization bear any correlation with disease progression remains to be established, but this is an important area for further evaluation. In addition to the antibody response generated by the humoral compo- nents of the host immune system, a protective response to a viral infection can also be mediated through the cellular immune system. With many other types of viruses, the host's cellular immune system is the
FUTURE RESEARCH NEEDS ~ 93 critical determinant of protection from and cure of viral infection. This system consists of specifically immune cytotoxic T cells and a variety of other effecter cells, including natural killer cells, macrophages, and other constitutive killer cells. The disruptive effects inflicted on the human immune system by HIV are well documented with respect to the obvious clinical manifestation of disease. However, the more subtle interactions concerning the specific cellular immune response to HIV and the deter- minants of that response's success or failure are almost a complete mystery. Many questions remain to be answered, including the specific types of cellular immune responses that might afford some protection from HIV infection, potential means of enhancing their antiviral efficacy, and the reasons for their apparent failure in the course of a natural infection (Weissman, 19861. The Immunologic Consequences of HIV Infection Once an HIV infection becomes established, compromise of the immune system can be detected in most, if not all, infected persons. The severity of this compromise may range from abnormalities detectable only by sophisticated immunologic analyses to the virtual collapse of the immune system seen in AIDS. The factors that determine the initiation of this immunologic decline, the rate of its progression, and its ultimate outcome are unknown. While it is possible that specific strains of HIV may be particularly pathogenic or have proclivities to induce specific types of immunologic or necrologic pathology, such correlations have yet to become evident. The possible existence of endogenous (e.g., genetic) or exogenous (e.g., environmental) exposures that modulate the course of HIV-induced immunocompromise is a topic of great interest. Generally referred to as "cofactors," these secondary factors could provide a target for limiting disease development in persons infected with HIV. In vitro studies of HIV expression have suggested that virus production and cytopathic effects are enhanced by the immunologic stimulation of infected T lymphocytes (Hoxie et al., 1985; Zagury et al., 19861. However, it is not known if an in vivo analogy exists for this observation, and epidemiologic analyses have so far failed to identify any environmental cofactors that strongly correlate with disease development (see Chapter 21. Neverthe- less, the relevance of cofactors to disease development remains of exceptional theoretical and practical importance. Well-controlled epide- miologic and laboratory evaluations are needed to assess this issue. The range of cells that harbor virus in vivo during the course of an HIV infection has yet to be completely elucidated. Infection can be docu- mented in T cells and macrophages in the lymphatic tissues, peripheral
194 CONFRONTING AIDS bloodstream, and extralymphatic organs (e.g., the lung). The central nervous system also provides a common target for viral infection and pathology, although the identity of the infected target cells within the brain is controversial (see section on "Neurologic Complications" in Appendix A). The number, type, and location of all cells infected by HIV within an infected person need to be documented from the earliest stages of infection and throughout the development of immunosuppression. Very little is known about the ability of HIV to infect or damage organs critical for T-lymphoid maturation-namely the bone marrow, where T-lymphoid precursors originate, or the thymus, where they are thought to assume antigenic specificity and functional competence. For any potential antiviral drug to have clinical benefit in a highly immunocompromised AIDS patient, it is essential that the immune system remain able to regenerate numerically adequate and functionally active lymphocytes. Should the T-lymphoid progenitor pool in the marrow be exhausted or the thymic environment be irrevocably damaged, this might not be possible. Only a very small percentage (less than 0.01 percent) of peripheral blood and lymph node cells in an HIV-infected person appear to express viral proteins and nucleic acids (Harper et al., 1986b). It is not known whether virus-expressing cells constitute all of the cells infected in vivo or whether HIV infection can assume a latent state wherein the virus, although present in a cell, is not expressed. Latent states in which virus expression is variously restricted at the levels of transcription and translation of viral RNA have been described for other lentiviruses and are thought to play an essential role in the maintenance of their charac- teristic persistent viral infection (Haase, 1986; Narayan, 19861. A latent state of HIV infection has been produced in vitro using a leukemia T-cell line (Folks et al., 1986), but it is not clear if an in vivo parallel exists. A truly latent infection of normal CD4 T cells has not yet been definitively established in vitro; reports of persistent infections of peripheral blood lymphocytes may actually represent smoldering cytopathic processes (Hoxie et al., 1985; Zagury et al., 19861. This may not be surprising, for in the case of certain animal lentiviruses and human and animal herpesviruses the maintenance of viral persistence appears to rely on an interplay of viral and host regulatory contributions (Haase, 19861. Since latent states might facilitate continued viral persistence and host infectiousness, as well as provide a mechanism of escape from antiviral therapies predicated upon the inhibition of HIV replication, their possible existence remains a critical unresolved research issue. It is very impor- tant to determine what fraction of cells in an infected person might be latently infected, what state the virus genome assumes in latently infected cells, how viral expression is maintained in a silent state, and how virus production may be reactivated.
FUTURE RESEARCH NEEDS 195 As previously described, HIV has a rapid and obvious cytopathic effect on infected CD4 lymphocytes in vitro. Although the identical population of T lymphocytes is also depleted in vivo, a number of questions remain to be resolved before the mechanism and range of cytopathic conse- quences manifest in an infected person will be understood. Because HIV expression is only detectable in rare cells in vivo, it remains to be established whether direct virus-induced damage adequately explains the decline in CD4 cells, or whether indirect mechanisms are also involved. Postulated indirect effects include autoimmune destruction of T lympho- cytes somehow selected by viral infection (Klatzmann and Montagnier, 1986; Williams et al., 1984), the liberation of immunocytotoxic substances from infected cells, or lymphoid inhibition secondary to the infection of cells with critical supporting roles in the immune response (e.g., macro- phages). While indirect effects may play a critical role in the CD4 T-cell depletion, little experimental support is available for their existence. The slowly progressive tempo of immunologic decline may alternatively result from the persistent direct effects of viral infection resulting in slow but cumulative infliction of damage. The exact mechanism of cell killing by HIV has not been established. In the few other retrovirus systems that have been studied in detail (Temin, 1986), cell killing upon viral infection is correlated with the expression of envelope proteins of particular antigenic type and the accumulation of quantities of unintegrated viral DNA resulting from a continuing process of reinfection. The cytopathic effect of the viral envelope and the presence of unintegrated DNA are also demonstrable in the course of HIV infection, and both may be intimately related to the ability of the virus to bind to the CD4 surface molecule. One of the most obvious cytopathic manifestations of HIV infection in vitro is the formation of multinucleated giant cells arising from the fusion of susceptible T lym- phocytes, which results in their death (Klatzmann et al., 1984; Popovic et al., 1984~. The formation of multinucleated cells, known as syncytia, is known to be a direct consequence of the interaction between the HIV envelope glycoprotein and the CD4 T-cell surface molecule (Lifson et al., in press; Lifson et al., 1986; Sodroski et al., 1986a). This interaction can readily be blocked by specific antibodies directed against the CD4 molecule, but it is only weakly inhibited by sera from infected persons (Lifson et al., in press; Lifson et al., 19861. The relative inability of sera from persons infected with HIV to prevent virus-induced T-lymphocyte cell fusion may allow both the continued spread of viral infection and the cytopathic consequences of viral expression. Also, the ability of one HIV-expressing cell to recruit other uninfected CD4-expressing T lym- phocytes into syncytia may greatly amplify the extent of T-lymphocyte depletion, given only a few virus-expressing cells.
~96 CONFRONTING AIDS Cell fusion is an obvious manifestation of HIV-induced cytopathology, but it is not necessarily the only one. Furthermore, the mechanisms accounting for the depletion of CD4-positive T cells in vivo may also involve cell fusion, but essentially nothing is known about the specific processes of cell killing in an HIV-infected host. Determination of the mechanisms of cell death both in vitro and in vivo will be necessary in future efforts to understand and prevent the pathology of AIDS. The cytopathic consequences of HIV infection may correlate with the level of expression of CD4 on susceptible target cells. This may underlie the ability of certain T lymphocytes and macrophages to support contin- ued virus production (also known as a chronic infection) in vitro (De Rossi et al., 1986; Gartner et al., 1986; Lifson et al., in press; Popovic et al., 19841. It is not known if chronically infected cells can be maintained in vivo and if so how they differ from cells that die upon infection. The cytopathic effects of HIV infection have been attenuated in vitro by artificially constructed viral mutants that compromise envelope expres- sion (Fisher et al., 1986b). The important and as yet unexplored possibil- ity exists that different viral isolates vary in their ability to kill or infect cells and induce immunosuppressive disease. Because the interaction between the HIV envelope protein and the CD4 molecule plays an important role in both viral infectivity and pathogenicity, development of ways to interfere with it may afford valuable strategies for disease prevention. As discussed in Chapter 2, the opportunistic infections and unusual malignancies that characterize AIDS are the most obvious manifestations of a severe underlying state of immunodeficiency. The broad spectrum and serious magnitude of the medical complications seen in HIV-infected persons have emphasized the central regulatory role that T lymphocytes play in the functioning of the immune system and in the preservation of health. However, there remain tremendous gaps in our understanding of the pathogenesis of the immunologic compromise of AIDS. Although apparent in in vitro assays and clinical manifestations, the nature and origin of the relevant immune deficits are not well established. The nature of the damage to the immune system that results from infection is not known, and the mechanisms and cellular targets by which the immuno- logic damage is manifest remain to be delineated. Similarly, the processes by which these specific defects in immune responsiveness and regulation are translated into specific susceptibility to opportunistic infections and malignancies are poorly defined. Much of this uncertainty reflects the limited basic knowledge about the functioning of the healthy human immune system. Future insights and interventions into the state of immunologic compromise in AIDS may necessarily depend on improved understanding of these basic processes.
FUTURE RESEARCH NEEDS 197 Just as the mechanisms of HIV-induced cytopathology remain unre- solved, the precise processes that account for the often profound immu- nologic impairment following viral infection await elucidation. A direct cytopathic insult to the CD4 T lymphocytes that results in their depletion is an attractive hypothesis. However, the simplicity of this hypothesis fails to adequately explain the complexity and pervasiveness of HIV- induced immunopathology. As discussed in Chapter 2, virtually all T-cell functions for which there exist in vivo or in vitro assays are markedly compromised or absent in persons with AIDS. These include delayed cutaneous hypersensitivity reactions, mitogen- and antigen-induced pro- liferative responses, generation of allogenic and autologous mixed lym- phocyte reactions, cytotoxicity, lymphokine production, available help for B-cell antibody synthesis, and expression of interleukin-2 and interleukin-2 receptors (Seligmann et al., 19841. The specific defect accounting for most of these manifestations of immunodeficiency can be explained by the paucity of CD4 helper/inducer T cells, since most of these abnormalities are not measurable when the assay systems are corrected for CD4 T-cell number (Lane et al., 19851. However, an intrinsic defect has been observed in the ability of residual CD4 T cells derived from AIDS patients to recognize and respond to soluble antigens, irrespective of numerical correction (Lane et al., 19851. A measurable defect in reactivity to soluble antigens can also be identified in HIV- infected persons who possess near-normal levels of CD4 T cells. In one study this defect correlated more highly with the progression from persistent generalized lymphadenopathy to clinical AIDS within a one- year period than did the absolute numbers of CD4 T cells (Murray et al., 1985b). It is not known if this specific insensitivity to antigens among CD4 T cells from AIDS patients reflects the preferential depletion of an antigen-responsive population or indirect immunosuppressive factors resulting from HIV infection. The direct inhibitory effects of HIV components themselves have been found in vitro, although their in vivo relevance is unclear (Pahwa et al., 19851. A number of the above-noted immunologic deficits in AIDS patients can be corrected in vitro through the provision of lymphokines that are normally produced by the CD4 T-cell population. Defective specific cytotoxic activity of CD8-positive T cells derived from AIDS patients and nonspecific natural-killer cell function can be partially corrected by exogenous interleukin-2 (Rook et al., 19831. Similarly, many aspects of the observed monocyte defect of AIDS, specifically the compromised intracellular killing of Toxoplasma gondii, the reduced hydrogen peroxide release, and the low expression of HLA class II antigens, can largely be corrected through the provision of gamma interferon (Heagy et al., 1984; Murray et al., 1985a). The lymphokine-mediated correction of these
198 CONFRONTING AIDS abnormalities suggests that they, too, derive from an inadequate inductive influence produced by the numerically limited CD4 T cells. It also suggests that, with greatly increased knowledge, immunoregulatory ther- apy of AIDS may one day be possible. B-cell function is also abnormal in patients with AIDS, marked by the frequent presence of hypergammaglobulinemia and the production of autoantibodies occurring in the absence of an ability to generate specific antibodies upon challenge by novel or recall antigens (Ammann et al., 1984; Lane et al., 19831. It is not known whether failure of appropriate B-cell function derives from the decline of normal regulatory control by CD4 cells, from the proliferative effects of Epstein-Barr virus infections resident in almost all AIDS patients (Quinnan et al., 1984), or from the reported stimulatory potential of HIV virion components (Pahwa et al., 19851. Since many of the opportunistic infections of AIDS patients may result from disrupted humoral immunity, improved understanding of its pathogenesis is needed. Indeed, the increased frequency of aggressive B lymphomas in HIV-infected persons may also result from the aberrant regulation of B-lymphocyte reactivity. The relationship of the appearance of Kaposi's sarcoma to HIV infection is also not understood. Because Kaposi's sarcoma appears with increased frequency in homosexual AIDS patients, it may result from the oncogenic influences of another agent with which they are infected. The lesions of Kaposi's sarcoma seen in AIDS patients appear to arise in a nonclonal, multifocal manner and as such are not conventional malignancies. As discussed in Chapter 2, the appearance of Kaposi's sarcoma in HIV-infected persons does not correlate with the degree of observed immunodeficiency, and it is unclear whether it derives from an absence of normal direct or indirect regulatory influences exerted on the Kaposi's target cell by a specific subpopulation of CD4 T cells or from the failure of immune surveillance against incipient tumors. Since the manifestations of Kaposi's sarcoma may not predict the extent of underlying im- munosuppression, it may be a poor determinant of which patient popu- lation to use in the evaluation of the efficacy of antiviral drugs (see section on "Antiviral Agents," below). However, study of these unusual lesions could contribute greatly to the understanding of AIDS, of the normal interactions between lymphoid and endothelial cell elements, and of the immune response to cancer. Recommendations · Continued study of HIV infections in human beings should include an intensified effort to define the introduction and spread of HIV in vivo, to identify the in vivo reservoirs of infected cells, and to determine the
FUTURE RESEARCH NEEDS 199 effects of HIV on the lymphoid and central nervous systems throughout the course of infection. · The CD4 molecule has been directly implicated as the receptor for HIV infection of cells. Although it was first discussed as a T-lymphocyte marker, several other cell types throughout the body have been found that express the CD4 molecule or the mRNA that potentially encodes it. Because AIDS involves pathologic alteration in several organs previously not thought to contain CD4-expressing cells, and because the whole-body reservoir of HIV-infected cells is unknown, more needs to be learned about the molecule and the full extent of cells expressing it. The developmental biology, function, and lifespan of all cells that express CD4 at some stage in their life history should be studied. · Future evaluations of potential drug therapies for HIV infection and AIDS will require more comprehensive immunologic and virologic anal- yses than are presently practical. Improved analytical techniques in these areas are also of critical importance for future epidemiologic studies of the natural history of infection and the existence of possible cofactors in the development of AIDS. Similarly, a number of presently unresolved issues of tremendous importance to public health will remain inadequately explored until such techniques are developed and widely available. These issues include the definition of the infectious state, the existence and extent of a virus-positive antibody-negative population, the relevance of host or environmental cofactors in the development of AIDS, and the efficiency of heterosexual transmission. EPIDEMIOLOGIC APPROACHES TO UNDERSTANDING THE TRANSMISSION AND NATURAL HISTORY OF HIV INFECTION The current understanding of the natural history and transmission of HIV infection and AIDS is reviewed in Chapter 2. Much, perhaps most, of the available knowledge about the disease derives from epidemiologic research for two reasons: there are presently no entirely suitable animal models in which to study the disease, and certain sorts of experimentation with human beings are precluded by the ethics of research. Much has been learned from epidemiologic research during the first five years of the AIDS epidemic, but future research needs remain great. (The background to the recommendations below is presented in Chapters 2 and 3.) Recommendations on Surveillance · The accuracy and completeness of the reporting of AIDS and ARC should be evaluated periodically; this should be done in several different
200 CONFRONTING AIDS areas because the prevalence of AIDS in those areas may affect that reporting. Such evaluations should be carried out according to a stan- dardized protocol to ensure consistency. These studies should be coor- dinated and supported by the Centers for Disease Control. · Surveillance of AIDS should be augmented by selective reporting of other stages of HIV infection, e.g., ARC, as determined by the Associ- ation of State and Territorial Health Officials in conjunction with the Centers for Disease Control, and by enhanced use of data from serologic testing of potential recruits to the armed forces. · Additional surveillance of HIV infection and AIDS is needed in groups of particular epidemiologic importance (not necessarily high risk), such as heterosexuals, non-IV drug users in high-incidence areas, IV drug users and homosexual men in low-incidence cities, spouses of infected individuals, pregnant women, newborn children of infected mothers, prostitutes throughout the country, and recipients of blood products. · Surveillance efforts for HIV infection and AIDS should not be at the expense of control efforts for other sexually transmitted diseases. · In general, better information is needed to quantify the number of persons infected with HIV. Extensive and repeated surveys of sero- positivity rates are needed to determine the incidence and prevalence of infection by age, race/ethnicity, geographic area, and sex. Such studies should be supported on a national basis, and standardized protocols should be developed to ensure comparability of the collected data. · There is an urgent need to incorporate capabilities for virus isolation in ongoing and future epidemiologic studies. Basic needs such as improving the efficacy of isolating virus from frozen samples, the ability to reproduce isolate virus from the same patient on multiple occasions, and the provision of adequate personnel and resources to isolate virus from large numbers of samples quickly all need to be addressed. · Some critical epidemiologic studies have both surveillance and basic science components and hence do not fall neatly within the ambit of either the Centers for Disease Control or the National Institutes of Health. The broad epidemiologic research program must be coordinated so that important proposals are not allowed to "fall through the cracks." A separate study section for AIDS epidemiology should be temporarily established within the National Institutes of Health. Recommendations on Natural History of HIV Infection · There is a need to determine what proportion of HIV-infected persons will become sick and to define many other features of the disease's natural history. · Additional information is needed on the natural history of HIV in
FUTURE RESEARCH NEEDS 20 ~ infected persons of all AIDS risk groups. The NIH-funded cohort studies of homosexual men should be continued and extended, and new studies should be developed for other current or potential high-risk populations, such as IV drug users and heterosexually active populations. Special attention should be directed to bisexual men and women with multiple sex partners. These studies should be geographically distributed, of sufficient duration (5 to 10 years), and coupled with studies of HIV molecular virology. Careful attention should also be given to the definition and assessment of important predictive parameters of the human immune response. · Prospective studies of individuals with HIV infections should be initiated in order to better assess the incidence of various manifestations and complications of infection with time. Presently, most data concerning clinical manifestations of the infection are limited to those listed at presentation or are derived retrospectively from small series of hospital- ized patients. Such data will be invaluable in assessing the effects of various therapeutic interventions. · Prospective cohort studies should develop standard classification schemes for disease manifestations or use those previously developed by others, such as the Centers for Disease Control or the Walter Reed Army Institute for Research. These schemes should be evaluated for their predictive value. Case-control studies of rare or uncommon findings in AIDS and ARC patients should be developed and supported by the Centers for Disease Control and/or the National Institutes of Health. Such studies are relatively inexpensive and may provide valuable infor- mation on the pathogenesis of HIV and point the way toward additional clinical or laboratory studies. · Studies to determine differences in the presentation, manifestations, and rate of progression of disease between subgroups of AIDS patients based on route of transmission (parenteral versus sexual), age (adult versus pediatric), or area of residence (United States versus Africa) should be started in an attempt to better understand the pathogenesis of the disease. · Possible cofactors should be clinically studied. Possible factors for investigation include coinfection with cytomegalovirus or other agents, stress modification, nutritional intervention, genital ulcers, and HIV superinfection. · Very little is known about the full extent of the introduction, spread, and pathogenic consequences of HIV infection throughout the course of human disease. Every effort should therefore be made to develop means to identify HIV-infected volunteers from whom tissue samples can be obtained in an ethical manner. For instance, logistics and funding should be developed for identifying and carrying out immediate and sterile
202 CONFRONTING AIDS autopsies on HIV-infected individuals who have previously volunteered for such research and who die as a result of some other cause (e.g., accidents, homicide, suicide, myocardial infarction). Recommendations on Transmission of HIV · The range of individual variability in levels of infectious HIV should be quantified, including the levels of virus shed into semen and other potentially infectious body fluids. · Present education programs designed to limit the spread of HIV infection are hampered by the lack of a meaningful definition of "safer sex" practices. There is a tremendous need for the epidemiologic evaluation of how various protective measures influence the transmission of the virus. The effectiveness of barriers, such as condoms, in preventing sexual transmission of HIV should be evaluated epidemiologically. Data on this issue should become available from the current prospective studies of homosexually active men and the spouses of infected hemo- philiacs. Complementary laboratory and epidemiologic research on viricidal lubricants for vaginal and anal use is also desirable, since these could be useful adjuncts to condom use (but almost certainly are not an acceptable primary mode of protection). · The risk of HIV transmission from men to women through vaginal intercourse should be estimated. Factors that affect this risk need to be identified. · The risk of HIV transmission from women to men through vaginal intercourse should likewise be estimated. It is not known if data from central Africa will be directly applicable in this regard to Western countries. Therefore, studies of heterosexual couples in which the female is the initial source of infection are needed to identify possible routes of transmission and factors related to the risk. Such data would also be critical for assessing the possibility that HIV transmission will be perpet- uated among heterosexuals, as are many other sexually transmitted diseases. The ideal subjects for this type of study are the male sexual partners of women who acquired HIV infections through transfusion. · The relative importance of various routes of transmission in central Africa should be determined. It is of particular interest that accurate data be obtained on heterosexual transmission, including risk factors related to sexual activity that may be unique to Africa or quantitatively different from Western countries. Parallel studies should be conducted in Africa and in the United States as far as possible. · Studies to monitor the risk of transmission through accidental needlesticks should continue to be supported. Such data are important in
FUTURE RESEARCH NEEDS 203 assessing the risk of infection in health care settings and may be relevant to the design of intervention trials. · The current risk of HIV infection among the hemophiliac recipients of clotting-factor concentrates in which the plasma donors have all been screened for HIV antibodies and the product has been treated to inactivate residual virus should be monitored. · The risk of HIV transmission within hemodialysis settings should be assessed. · The time and route of maternal-infant transmission and the factors related to the risk of HIV infection in infants should be identified. Data also are needed to confirm the detection of virus in breast milk and to determine whether it can be transmitted orally to infants through breast feeding. In assessing responsibility for the various studies recommended above, policymakers should take account of the fact that certain agencies have traditional strength in certain types of activity and others may be situated to take advantage of unique opportunities. For example, the Department of the Army has the capacity to follow infected individuals over time more readily than other groups. Agencies and policymakers should recognize that collaboration in some project will be essential for example, in follow-up of seropositive military applicants. Recommendations on the Need for Improved Serologic and Virologic Tests Much evidence suggests that the serologic tests now available are generally accurate and useful (see Appendix B). However, there are some obvious needs for their improvement and for other techniques in virus detection and isolation. Improved or new techniques would greatly facilitate studies of the natural history and epidemiology of HIV infection and AIDS. The areas of greatest need are as follows: · Improved tests for antibodies to HIV, particularly tests having greater specificity. (In this area, competitive assays and possibly tests using antigens derived through recombinant DNA techniques appear to hold the greatest promise.) · New and improved methods for the detection of virus, both through the identification of viral antigens and through the detection of infectious viral particles. · Standardized procedures for virus isolation that maximize recovery and that could be used to evaluate other techniques. These areas of work are no less important than others mentioned in this
204 CONFRONTING AIDS chapter, and policymakers may need to consider special arrangements to ensure that work in some of them e.g., viral isolation techniques is not neglected. ANIMAL MODELS New methods have allowed rapid progress to be made in understanding the molecular biology of HIV. However, because AIDS is a uniquely human disease, the biologically relevant aspects of the infection and disease will be far more difficult to unravel. The sequence of cells initially infected; the sites of latency and persistence; the host determinants and cofactors in disease development; the timing, incidence, and sites of nervous system infection; and similar questions cannot be answered in cell cultures and cannot be solved solely by epidemiologic observations. Animal models of immunosuppressive retroviral infections will be crucial in addressing and resolving these biological questions. Similarly, studies of methods to control HIV infection whether by blocking transmission, by chemotherapy of the infected, or by immuni- zation of the uninfected will ultimately have to be done in human populations. Such studies must conform to ethical standards and thus raise problems of research design. In patients with clinical AIDS, the evaluation of potentially dangerous drugs can be justified. But the testing of vaccines in uninfected populations or the testing of potentially danger- ous drugs in seropositive yet healthy persons will pose major practical and ethical problems. Again, animal models are needed. Animal models of HIV infection will be critical components of both vaccine and antiviral development programs. These models provide an important intermediate level of drug evaluation between in vitro studies and testing in humans. Many agents active against HIV in vitro will never have been used on humans, so drug evaluation in other animals will provide essential information about efficacy, pharmacology, penetration of the central nervous system, and toxicity. In the search for experimental counterparts of human diseases, two strategies have been used (Johnson, 19851. One involves the infection of experimental animals with the human agent in hopes of simulating the human disease. The other involves studying analogous diseases in their natural hosts, such as the classic studies of the pathogenesis of ectromelia virus in mice that illuminated the pathogenesis of smallpox. Both ap- proaches have their value. There is no entirely satisfactory animal model for human HIV infection and disease. However, several experimental approaches are being taken that may shed light on the infection process.
FUTURE RESEARCH NEEDS 205 HIV Infection of Chimpanzees The only host thus far found to be susceptible to chronic infection with HIV is the chimpanzee. To date, infected chimpanzees have been shown to develop antiviral antibodies, viremia, and an occasional transient adenopathy but no disease (Alter et al., 1984; Francis et al., 1984; Fultz et al., 1986b; Gajdusek, 19851. This model is therefore of little use in studying the later stages of disease pathogenesis, but it will be indispens- able in the development of HIV vaccines. Chimpanzees are the only animals now known in which HIV vaccines can be tested before clinical application. They can be immunized with a candidate vaccine and subsequently challenged with the human virus to determine if the vaccine provides safe, effective prevention of infection. For example, an envelope protein subunit vaccine could be given and the animals then monitored following challenge with infectious HIV for seropositivity to core proteins as well as for recoverable virus in the blood. Chimpanzees may also play a critical role in the evaluation of the ability of antiviral drugs to prevent, limit, or cure HIV infections. Only about 1,400 chimpanzees are now available for all biomedical research. Housing and maintaining these long-lived animals require major expense. There is also convention among researchers against sacrificing chimpanzees for complete pathologic or virologic studies, which impedes the progress of animal research related to human disease. Increasing the number of chimpanzees available through breeding is a slow process, and their importation is presently not legal because they are designated as an endangered species. A transient HIV viremia has been found occasionally in inoculated macaques, but this would be of very limited experimental value. HIV-Related Viruses in Old World Primates Retroviruses resembling HIV have been isolated from rhesus macaques, African green monkeys, sooty mangabeys, and pig-tail macaques (Desrosiers and Letvin, 19861. These agents (collectively termed STLV-III or SIV) have tropism for T4 cells similar to that of HIV; they also have a similar morphology, antigenic relatedness, and distant sequence homology (less than 75 percent). SIV was initially isolated from captive rhesus monkeys (Macaca mulatta) suffering from a disease resembling AIDS (Daniel et al., 1985; Kanki et al., 1985~. Later, it was found in a large number of healthy African green monkeys (Cercopithecus aethiops) in the wild (Kanki et al., 19851. Although the viruses isolated from the different monkeys appear similar if not identical, they have clearly different effects on their infected
206 CONFRONTING AIDS hosts. Understanding why SIV-infected African green monkeys remain healthy while rhesus monkeys develop profound immunosuppression may provide very useful insights into the pathogenesis of AIDS. Two patterns of disease have been observed in rhesus monkeys inoculated with SIV. In an initial study, four monkeys developed low titers of antibody, hypogammaglobulinemia, inversion of CD4/CD8 cell ratios, and death with opportunistic infections and an encephalitis resem- bling the subacute encephalitis of AIDS. In the initial study two animals developed a good antibody titer, a persistent viremia, a decline in CD4 cells, and lymphadenopathy, but they did not die (Letvin et al., 19851. Subsequent rhesus monkeys have shown a second pattern of nonlethal disease, with wasting, diarrhea, and decreased CD4 cells (N. Letvin, New England Primate Center, personal communication, 1986; Yanagihara et al., 19864. When the mangabey SIV isolate was inoculated into the juvenile rhesus macaques, it induced antibody and viremia with diarrhea and lymphadenopathy (M. Gardner, University of California at Davis, personal communication, 19861. However, a subsequent similar study described additional isolates that did not cause clinical immunodeficiency in either their natural hosts or rhesus macaques (Fultz et al., 1986a). In general, SIV induces a disease in monkeys remarkably similar to human AIDS, making it suitable for studies of pathogenesis, drug trials, and vaccine development. This could be an extremely useful animal model. A detailed description of SIV's character and relevance to HIV infection and AIDS is presented in the background paper prepared for the committee by Letvin and Desrosiers (19861. The recently identified novel human viruses HTLV-IV and LAY-2 demonstrate similarities to SIV. Although HTLV-IV is reported to be apathogenic, LAY-2 was isolated from persons with AIDS. These viruses need better molecular and genetic, as well as biologic, characterization. The variables influencing pathogenicity of these agents may provide important insights into the mechanisms of disease causation and evolu- tionary origins of the immunosuppressive retroviruses related to HIV. Lentiviruses of Ungulates Lentiviruses distantly related to HIV include equine infectious anemia virus (of horses), ovine visna virus (of sheep), and caprine arthritis- encephalitis virus (of goats). HIV shares structural and biologic similar- ities with these other lentiviruses (Gonda et al., 1985; Sonigo et al., 1985~. Although the different animal lentiviruses demonstrate a variety of different pathogenic effects and strategies for evasion of the host immune system during the course of an infection, they provide a wealth of opportunities for understanding or anticipating the complexities of HIV
FUTURE RESEARCH NEEDS 207 infection. The animal lentiviruses and their relevance to HIV infection and AIDS are discussed in the section on "Vaccines" later in this chapter and in the accompanying background papers by Haase (1986) and Narayan (1986~. The animal lentiviruses show epidemiologic features similar to those of HIV, being transferred by blood or mononuclear cells in secretions. These infections largely involve the macrophages, and none of the ungulate viruses directly affect T helper lymphocytes to cause the severe immunosuppression seen in AIDS. These diseases have long incubation periods, persistent viremia, development of weak neutralizing antibody responses, a tendency for the virus to undergo extensive antigenic mutation and antigenic drift, neurotropism, and lytic virus infection of selected leukocyte cell populations. Because lentiviruses are the nearest known relatives of HIV, are partially related in sequence, and have similar biological properties, they can be very useful in understanding HIV-induced disease. Studies of pathogenesis, drug trials, and vaccine strategies could be explored first in these ungulate hosts (Haase, 1986; Narayan, 19861. Conclusions and Recommendations The committee is particularly concerned about the nurturing and preservation of primate animal model resources, which will be essential for drug and vaccine development and for basic investigation of disease pathogenesis and natural history. There are approximately 1,400 chim- panzees in the United States today available for all biomedical research. About 800 of these are outside federal control, and a significant propor- tion of the approximately 600 under federal control (or possibly otherwise available e.g., not in private ownership) are not available for AIDS- related research because they are reserved for breeding programs or are in reserve for use with other emergent diseases. In addition, only those chimpanzees who have already been exposed to non-A, non-B hepatitis virus, making them unsuitable for breeding purposes, are currently being approved for use in AIDS-related research. This leaves a very small number, approximately 200, potentially available for research on AIDS and HIV. An NIH committee following commonly agreed standards of research with animals approves experiments on chimpanzees under federal con- trol. The use of approximately 85 chimpanzees for AIDS-related research has already been approved. As of August 1986, $4.5 million of discretionary funds have been allocated to the NIH Division of Research Resources for a chimpanzee breeding program (Barnes, 19861. The committee is gravely concerned that chimpanzees have been and
208 CONFRONTING AIDS might be used for experiments for which the rationale is not compelling in light of the scarcity and irreplaceable nature of these animals; extreme caution must be exercised in their use. Chimpanzees must be treated as an endangered national resource that will be irreplaceable if squandered. Mechanisms must be found to ensure that AIDS-related experiments with these animals are conducted only if there is a broad consensus among the interested scientific community that the experiment proposed is critically important to development of vaccines or antiviral agents and cannot be conducted in any other species or by any other means. These consider- ations apply equally to animals not under federal control. · Animal models that reproduce or mimic the consequences of HIV infection will play a crucial role in improving our understanding of disease pathogenesis and the development and testing of antiviral drugs and vaccines. Available animal models of HIV infection and the im- munocompromised state must be vigorously supported, and efforts must be made to develop and validate new animal models. Because the most relevant and promising animal models are provided by nonhuman pri- mates, the nation's primate centers should be strengthened to permit the expansion of the primate populations available for AIDS-related research, the development of appropriate biocontainment facilities, and the educa- tion of appropriately trained investigators. · For fiscal and ecological reasons, studies that can be done with relatively well-defined disease models in domestic animals should be carried out in sheep, goats, and horses rather than in primates. · Available supplies of primates will probably be inadequate for future research needs, and the plans for their conservation, expansion, and optimal use appear inadequate. Funding for the systematic evaluation of additional model systems and the validation of existing ones should be a top research priority. Presently available animal resources should be expanded as rapidly as possible to meet expected future demands. Also, a national system should be installed to facilitate appropriate access to test animals for valid experimentation by qualified investigators, regard- less of institutional affiliation. · Naturally occurring retroviral infections leading to significant ac- quired immunodeficiencies have been found in several vertebrate species. Although only a few potentially produce a syndrome quite similar to human AIDS, none has been studied sufficiently to elucidate mechanisms of pathogenesis, so that one could know whether the different types of immunodeficiency result from similar or different basic mechanisms. Therefore, support of known animal systems of retrovirus-induced im- munodeficiency, including murine, feline, and especially simian models, should be increased and broadened.
FUTURE RESEARCH NEEDS 209 · A virus related to HIV that has been isolated from nonhuman primates, SIV, has great potential as a model system to study im- munocompromise resulting from a retroviral infection. However, the generalizability of conclusions derived from SIV must be evaluated with respect to similarities with and differences from HIV infections. Further studies need to be carried out with SIV in rhesus and other breeds of monkeys. If the rhesus model is to be widely used, more biocontainment space must be made available. · A completely analogous animal model of HIV-induced human AIDS has not yet been found. Most of what has been learned about the fundamental aspects of animal virology, immunology, lymphocyte biol- ogy, and, to a lesser extent, neurobiology, has come from extensive investigations with genetically defined inbred strains of rodents. In these times of recombinant DNA manipulations of retroviruses, of tissue culture cells, and of transgenic mouse strains, it is possible that a completely analogous animal model of HIV-induced human diseases might be developed. Therefore, development of possible animal models should be considered that would include both natural models and the deliberate modification of existing species to render them susceptible to AIDS by HIV infection. At the same time the possible biohazards of this type of research need careful consideration. ANTIVIRAL AGENTS Infectious viral agents remain a major threat to human health both in the United States and around the world. Recent years have therefore seen considerable effort directed toward the development of new antiviral agents for the treatment of acute viral infections. However, even though there are now many examples of successful chemotherapeutic conquests of bacterial infections, there are only a few examples of drugs that are elective in the treatment of viral diseases. Much of the difficulty in treating viral infections is a consequence of their nature as intracellular pathogens- that is, they replicate within the cells of their chosen host. Because viruses use many of the host cell's synthetic pathways in their reproductive processes, specifically inhibiting their replication without also severely compromising the metabolic activ- ities and health of their hosts is difficult. As a virus that appears to cause persistent lifelong infection, HIV must be approached as a member of the class of viruses for which successful treatment may be most difficult to find. Furthermore, as a member of the family of retroviruses, HIV represents a type of viral pathogen whose therapy has never before been attempted in humans. Because the contem- plated development of drugs to treat HIV infection and AIDS represents
210 CONFRONTING AIDS such a novel and difficult challenge, predictions of ultimate success are presently impossible to offer. Likewise, the timing of the development and availability of potential drugs cannot be accurately forecast. These notes of caution and even pessimism must be tempered by an appreciation of the remarkable scientific advances that have been made to date in understanding HIV. Similarly, recent developments in molecular biology, virology, immunology, structural chemistry, and drug design have initiated what may be a new era in the treatment of previously intractable human viral infections. In many ways the problems of HIV infection and AIDS may provide the first and most substantial test of these emerging capabilities. Rational drug design based on the specific characteristics of viral processes is still a long-term prospect, but with the impressive skills and commitment of the biomedical research community, this theoretical approach could become a reality. The basic objective of antiviral therapies for AIDS and its related conditions is to effectively inhibit the replication of HIV. Underlying this objective is the belief that continued viral replication is involved in the development and progression of the disease. This belief is corroborated, up to a point, by the limited available studies of the natural history of HIV infection. As discussed in the section on "Natural History of HIV Infec- tion," above, at any given time only a small proportion of cells in an infected person actively express HIV (Harper et al., 1986a). The decline in CD4 T cells, which is thought to account for much of the immunologic compromise of AIDS, appears to be a slowly progressive one. Recent in vitro analyses suggest that the cytopathic effects that HIV exerts on CD4 T lymphocytes result from the active expression of viral gene products. The exact mecha- nisms of CD4 T cell depletion in vivo are not known, but they have been postulated to include a direct cytopathic effect of the viral envelope protein (Lifson et al., in press; Sodroski et al., 1986a) as well as potential indirect, possibly autoimmune processes (Klatzmann and Montagnier, 19864. The persistent, although low-level, expression of the cytopathic deter- minants of HIV may explain the progressive immunologic deficit seen in infected persons. Thus, a chemotherapeutic approach designed to limit viral expression or treat HIV infection may be reasonably expected to minimize the resulting damage to the immune system. However, clinical improvement as a result of antiviral therapy can only be expected if the host immune system has not been irrevocably damaged and maintains regenerative potential. Antiviral therapies may thus have limited utility in persons whose immunologic damage has progressed beyond the point of recovery. Because the regenerative capacities of the adult human immune system are not well understood, the existence or designation of this point is presently impossible to evaluate. Regardless, antiviral therapies might be expected to be most useful in persons infected with HIV but not yet
FUTURE RESEARCH NEEDS 21~ clinically ill. Attention is being focused on potential immunoregulatory and restorative therapies for end-stage HIV infection, yet a great deal must be learned before their practicality can be assessed. A related and important consideration is the potential ability of antiviral agents to beneficially affect the necrologic manifestations of HIV infec- tion. While the consequences of viral infection in the central nervous system are well documented, their potential for reversal is unknown (see section on "Neurologic Complications" in Appendix A). The development of antiviral drugs for HIV infection must also take into account a number of problematic aspects of the behavior of the virus in viva. HIV is known to establish an infection that persists in the face of a host immune response, probably for the lifetime of the infected individual. This behavior, referred to as viral latency, is typical of the lentiretroviruses. The mechanisms that permit the establishment and maintenance of viral latency are not known, but determining them will be extremely important as drug therapies are considered. Should a candidate drug limit, but not eliminate, HIV expression, lifelong therapy may be necessary to achieve the desired protection. If lifelong therapy is required, the already difficult challenge of devising drugs of acceptable toxicity will prove even more severe. In addition to the predictable mechanisms of drug toxicity, AIDS patients have demonstrated unusual toxic reactions upon treatment with anti-infectious or anticancer chemotherapies (Jade et al., 19831. Similar problems not antici- pated in drug development may arise during long-term therapy in persons with AIDS. Prolonged administration of an antiviral agent may also result in the selection of drug-resistant viral variants. However, it is possible that if a drug electively limits replication, the host viral load will progressively decline and ultimately vanish. Successful development of antiviral therapies for HIV must also consider the cellular reservoirs that are involved in viral persistence. Cells of the macrophage/monocyte lineage are known to be susceptible to viral infection in vitro and may also serve as in vivo hosts for viral replication (Gartner et al., 1986~. Given their relative longevity, wide- ranging migration, and apparent relative resistance to the cytopathic effects of viral infection, protracted systemic antiviral therapy may be required for cure or even stabilization. Similarly, the central nervous system is known to be an early and common target of HIV infection (Ho et al., 1985c; Resnick et al., 1985~. Because the brain may carry a substantial viral load (Straw et al., 1985), it may provide a sanctuary for viral persistence. The presence of a reservoir of HIV within the central nervous system requires that antiviral drugs that cross the blood-brain barrier be developed if total viral elimination is to be achieved. A number of possible targets exist for a rational approach to designing drugs to specifically inhibit HIV replication. These include such viral
2 ~ 2 CONFRONTING AIDS processes as assembly and uncoating, integration, reverse transcription, and proteolysis. Drug Evaluation in vitro Candidate drugs for the treatment of HIV infection must fulfill criteria of antiviral efficacy and acceptable toxicity. Certain parameters of these criteria can be measured in vitro, and innovative assay systems have been established to permit the screening of potential candidate drugs. The HIV reverse transcriptase plays an essential role in the viral life-cycle and is presently a primary target of antiviral agents. Its activity can easily be measured in vitro. A1SO7 as discussed earlier in this chapter, the produc- tion of large amounts of active reverse transcriptase by recombinant DNA methodologies is an important goal. Many substances may successfully inhibit purified HIV reverse transcriptase in vitro yet prove unacceptably toxic. Screening methods to simultaneously measure effective inhibition of viral replication and in vitro drug toxicity have recently been described (Mitsuya and Broder, 1986; Mitsuya et al., 19851. Additional assay systems to evaluate agents that inhibit HIV replication need to be developed and validated, as do methods to measure the specific inhibition of the novel viral gene products, as described earlier. Drug Evaluation in Humans Evaluation of the efficacy of antiviral therapies in persons infected by HIV is a critical but difficult task. In contrast to the direct determination of antiviral effects that may be measurable in vitro, it may prove extremely challenging to accurately evaluate the benefit of candidate drugs in infected persons. As opposed to many other viral diseases in humans, where the immediate clinical symptoms directly relate to ongo- ing replication of the causative virus, many of the clinical symptoms of AIDS represent indirect manifestations of prior viral damage. For in- stance, the diagnostic opportunistic infections or typical malignancies of AIDS are the presumed consequences of HIV-induced damage to helper/inducer lymphocyte populations. Even if a drug could effectively halt HIV replication, it might not resolve these immediate clinical signs and symptoms. Once established, certain clinical diseases may be inde- pendent of HIV regulation. By the same token, failure of an antiviral drug to halt the progression of established opportunistic infections or Kaposi's sarcoma may not de facto indicate lack of efficacy. Protection of healthy individuals infected with HIV but showing no AIDS-related infections or malignancies may be the most useful measure of
FUTURE RESEARCH NEEDS 213 antiviral effects. But the evaluation of this protection requires large popula- tions and long time periods. There are also ethical constraints involved in the administration of potentially toxic agents to persons who are not overtly ill. Drugs that may show no beneficial effects in AIDS patients may, in fact, be beneficial in forestalling the immunologic compromise that follows HIV infection. These may, however, be difficult to identify. The clinical and laboratory measures presently in use to evaluate candi- date drugs for antiviral activity against HIV involve virologic, immunologic, and clinical parameters. Virologic measures attempt to measure virus load, but at present they do so rather poorly. Measurements of HIV expression in vivo are not generally applicable, and given the low levels of viral expression they currently detect, they may be at the limits of their resolution (Harper et al., 1986a). Recent advances in the detection of HIV antigens in vivo have suggested that such an approach would be very useful, but it is presently not known to provide adequate sensitivity of viral detection. Current methodol- ogies of viral isolation from infected persons are only semiquantitative, are of limited sensitivity, and require expertise that is not now widely available. Measurement of drug efficacy using in vitro methods for viral isolation is further limited by an inability to identify viral production in vivo, and by the documented variability in the ease of isolation of various HIV strains (Gartner et al., 19861. Measurements of immunologic competence in AIDS patients generally involve determinations of CD4 T-lymphocyte levels, cutaneous hypersensi- tivity testing, and in vitro measures of T-cell responsiveness (see Chapter 2~. It is not yet known which, if any, of these factors provide accurate insights into the most clinically relevant immunologic consequences of HIV infection or can act as usable barometers of the extent of viral replication. The clinical impact of antiviral therapy is ultimately the most important measure of efficacy. As discussed above, however, it also may be the most difficult to quantify meaningfully. Prolonged survival, amelioration of the frequency or severity of opportunistic infections, and freedom from malignant diseases or necrologic sequelae provide the most important guideposts. However, the natural history of HIV infection has not yet been defined completely, so the efficacy of antiviral agents will have to be determined as this definition is being established. Current Antiviral Agents Under Clinical Study The major candidate drugs undergoing preclinical and clinical evalua- tion for efficacy against HIV infection have been identified empirically in studies of other viruses. Although the mechanisms of action of many of them are unknown, some are thought to exert their antiviral effects through inhibition of reverse transcriptase. Following in vitro evaluation
214 CONFRONTING AIDS of their ability to inhibit HIV reverse transcriptase, replication, or cytopathology, these drugs have progressed through subsequent levels of pharmacologic and toxicologic tests before being administered to infected humans. Some agents have demonstrated an antiviral activity in vitro but have yielded disappointing results when tested in AIDS patients. For instance, Suramin treatment, besides failing to improve the immunologic impair- ment seen in AIDS, caused serious toxicity. Thus, agents that appear promising in vitro may yield no benefit, and perhaps even serious tox- icities, when administered to humans. There are currently no drugs with documented efficacy in the treatment of HIV infection. The drugs described below have shown variable degrees of initial promise in vitro and are undergoing clinical evaluation in HIV-infected persons. As previously discussed, dramatic breakthroughs in this pursuit are unlikely, but therapeutic benefits could aid in the increased understanding of the basic mechanisms responsible for the clinical development of AIDS. Suramin Suramin was the first drug reported to have in vitro inhibitory effects on HIV replication (Mitsuya et al., 19841. Suramin is used to treat trypanosomiasis and onchocerciasis (Hawking, 19811. It has also been noted to inhibit the reverse transcriptase activities of a number of murine and avian retroviruses (de Clercq, 19791. Although it is a potent inhibitor of HIV reverse transcriptase, it also significantly inhibits mammalian DNA polymerase alpha (Chandra et al., 1985) and lymphocyte prolifera- tion, thus limiting its therapeutic index. Clinical evaluation of suramin in AIDS patients resulted in a transient inability to recover virus from treated individuals, but it produced little or no evidence of clinical improvement or immunologic recovery (Broder et al., 1985; Levine et al., 1985; Rouvroy et al., 1985; Stein et al., 1986~. Suramin therapy was complicated by fever, rash, renal and liver function abnormalities, and serious adrenal compromise. In addition, the effects of suramin on lymphocyte proliferation may exert a counterproductive immunosuppressive influence at higher doses. Finally, suramin could not be a viable single-drug modality because the drug does not penetrate into the central nervous system. HPA-23 Ammonium 21-tungsto-9-antimoniate, or HPA-23, inhibits the reverse transcriptase activity of a number of murine oncornaviruses in vitro and
FUTURE RESEARCH NEEDS 215 protects against the leukemogenic consequences of their replication in vivo (Chermann et al., 1975; Jasmin et al., 19741. Although HPA-23 is reported to act as a competitive inhibitor of reverse transcriptase, the mechanism of its antiviral activity against HIV is not known. Initial limited clinical evaluations indicated that HPA-23 could exert a transient virustatic effect in patients with AIDS or ARC as measured by failure to isolate HIV following a two-week course of treatment (Rozenbaum et al., 19851. Following cessation of treatment, HIV could again be isolated, and there was no indication of beneficial immunologic effects. Toxicities included thrombocytopenia and transient liver function abnormalities. Although clinical evaluation is continuing in the United States and France, no therapeutic benefit of HPA-23 for persons infected by HIV has been documented. Azidlothymidine (AZT) 3'-Azido-3'-deoxythymidine (AZT), which has been previously re- ferred to as "compound S" and BW A509U, is a thymidine analog modified so that it acts as a chain terminator during DNA synthesis. Although not an inhibitor of reverse transcriptase per se, it can effectively prevent the synthesis of proviral DNA by HIV reverse transcriptase by frequently interrupting the growing viral template. Following conversion to a triphosphate form by cellular enzymes, AZT effectively inhibits the in vitro infectivity and cytopathic effects of HIV (Mitsuya and Broder, 1986; Mitsuya et al., 19851. The drug is well absorbed following oral adminis- tration and effectively penetrates into the central nervous system (Yarchoan et al., 19861. Initial phase I clinical trials defined a relatively short half-life for the drug, with dose-dependent toxicities involving suppression of certain hematologic measures and headaches (Yarchoan et al., 19861. However, it was generally well tolerated. A virustatic effect against HIV was documented at higher doses of AZT. Some patients experienced increases in their levels of CD4 T cells during the course of therapy, and there were occasional indications of partial immunologic improvement involving amelioration of cutaneous energy, clearing of infections, improvement in necrologic symptoms, and weight gain. Surprisingly, clinical and immu- nologic improvements were seen in individuals receiving doses that permitted the continued isolation of HIV. Some patients developed episodes of opportunistic infections during the treatment course. With prolonged administration of AZT, declines in CD4 T-cell levels were seen in some patients, perhaps reflecting pharmacologic suppression of the bone marrow.
216 CONFRONTING AIDS . The AZT-induced suppression of hematopoiesis and immune reactivity Is correlated with drug-related pyrimidine starvation that accrues with continued administration. These side effects may limit both the accept- able dose of AZT that can be administered and the permissible length of therapy. Shortly before the publication of this report, data were released by the National Institutes of Health and the Burroughs Wellcome Company from a study of azidothymidine administered for 20 weeks to a group of approximately 140 AIDS patients while a similar group received a placebo. The patients were selected for having had no more than one bout of Pneumocystis carinii pneumonia. There was 1 death in the AZT group compared with 16 deaths in the placebo group. Because of the time at which this information became available, the committee was not able to analyze the data from this study in enough detail to judge the risks and benefits of this drug. Further evaluation will be needed to fully determine the side effects of AZT treatment and its long-term efficacy and safety for various categories of patients. More recent analyses of other nucleoside analogs that function as terminators during DNA synthesis have uncovered a variety of com- pounds that are very potent inhibitors of HIV replication and cytopathol- ogy in vitro (Mitsuya and Broder, 19861. Of these agents, 2',3'-dideoxyadenosine and 2',3'-dideoxycytidine have shown the most promise and are currently undergoing toxicity testing in animals in preparation for their evaluation in HIV-infected persons. Ribavirin Ribavirin is a nucleoside that has demonstrated in vitro activity against a number of RNA viruses in vitro and clinical utility in the treatment of influenza A (Knight et al., 1981), respiratory syncytial virus (Hall et al., 1983), and Lassa fever (McCormick et al., 19861. Ribavirin has been noted to exert in vitro antiviral effects against certain bovine and avian retrovi- ruses (Jenkins and Chen, 1981; Sidwell and Smee, 1981) and has been shown to limit retroviral replication in vivo in mice infected with murine leukemia viruses (Shannon, 1977; Sidwell et al., 1975~. It has also been reported to partially inhibit HIV replication as measured by a decreased level of viral reverse transcriptase activity produced following infection of lymphocytes with HIV (McCormick et al., 1984~. In vitro suppression of reverse transcriptase activity, however, was not complete even at doses known to elicit toxic reactions in vivo. Although ribavirin is thought to interfere with the capping of viral RNA transcripts in certain RNA viruses by inhibiting the activity of virally encoded guanylyl transferases, HIV uses cellular capping activities and
FUTURE RESEARCH NEEDS 217 thus would not be subject to this mechanism. Ribavirin also results in the synthesis of poorly translatable viral mRNAs in specific RNA viruses, but it is not yet known if this involves mechanisms independent of compro- mised cap formation (Toltzis and Huang, 19861. In previously reported clinical applications, ribavirin has demonstrated dose-dependent toxicities expressed mainly as nomocytic anemia. The potential in vivo efficacy of ribavirin against HIV infection was suggested in recent phase I clinical studies, where decreased viral replication and enhanced immune function were noted (Roberts et al., 19861. Should ribavirin prove to be an inadequate single-agent therapy, it may still be of use in combination with a drug acting by a presumably distinct mecha- nism, such as an inhibitor of HIV reverse transcriptase. An independent safety and evaluation committee has reviewed the first 12 weeks of a 24-week clinical trial of ribavirin with 373 patients at eight clinical centers in the United States. The committee found that ribavirin's safety profile through 12 weeks is acceptable and that the drug has been well tolerated by the patient groups (ICN Pharmaceuticals, 19861. Foscarnet Trisodium phosphonoformate, also know as foscarnet, is a drug with noncompetitive inhibitory activity against the reverse transcriptases of certain avian and murine oncornaviruses (Sundquist and Oberg, 19791. Also able to inhibit the DNA polymerases of a number of human herpes viruses (Helgstrand et al., 1978), foscarnet has been used to treat cytomegalovirus infection in immunocompromised patients. Recent stud- ies have documented an in vitro inhibition in HIV reverse transcriptase activity at levels pharmacologically acceptable in vivo (Sandstrom et al., 1985; Sarin et al., 19851. Administration of foscarnet is occasionally complicated by acute renal failure. Although the results of clinical evaluation of foscarnet in HIV-infected persons have yet to be reported, clinical trials are in progress. Alpha Interferon Alpha interferon is a member of a rapidly expanding category of naturally occurring proteins, referred to as biological response modifiers, that are finding clinical application. Before HIV was identified as the cause of AIDS, alpha interferon was administered to patients with Kaposi's sarcoma in the hope of exploiting its known activity against certain tumors. The earliest attempts to isolate retroviruses from AIDS patients were, in fact, facilitated by the depletion of alpha interferon from the cell cultures (Gallo et al., 1984), and alpha interferon has recently
218 CONFRONTING AIDS been shown to exert a virustatic effect against HIV in vitro (Ho et al. 1985a). Although it is not known whether alpha interferon exerts similar antiviral action in viva, clinical studies have demonstrated potentially beneficial effects against Kaposi's sarcoma. In early clinical studies of alpha interferon in AIDS patients, no enhancement of immunologic function was detected, but a percentage of persons with Kaposi's sarcoma (especially those with early disease restricted to the skin) achieved a complete remission (Gelmann et al., 1985; Groopman et al., 1984; Krown et al., 19831. Continuing clinical analyses are in progress to evaluate the potential therapeutic efficacy of alpha interferon in the treatment of HIV infection and Kaposi's sarcoma. New Antiviral Agents Against AIDS Although NtH, largely through the National Cancer Institute (NCI), has started a program to test in vitro for agents active against HIV, the program has to date been limited largely to in-house selection of com- pounds collected over the years by NCI for evaluation against cancer in animal models. The program has so far tested approximately 250 com- pounds. There is presently no satisfactory program whereby scientists from academic institutions, research institutes, or the pharmaceutical industry may send significant numbers of candidate antiviral agents for evaluation against HIV in vitro. The problems related to safe handling of the virus have greatly restricted its use in most laboratory settings. It is absolutely essential to future drug development in this area that the NIH testing program be substantially enlarged to serve industrial and academic research and research institutes outside of NIH, under a confidentiality agreement if necessary. These testing results must be readily available to recruit the scientific talent from both the academic community and the private sector. This in vitro testing program must be supported by adequate animal models to determine those agents that are unique candidates for clinical evaluation against HIV. (While the NIH effort should be of adequate size, it should not be turned into a mass screening program as these tend to be inefficient at finding useful drugs.) The NIH drug development program for contract proposals from cooperative multidisciplinary groups is woefully inadequate in funding to meet the stated needs. Research efforts to synthesize and evaluate new agents potentially active against AIDS have been minimal to date. That new agents will be found active in vitro is unquestionable, but how the agents will be selected for clinical trials is presently not clear. Similarly,
FUTURE RESEARCH NEEDS 219 guidelines and animal testing protocols need to be established for new candidate drugs for potential clinical evaluation against AIDS. Conclusions and Recommendations Development of therapy for HIV infection will most likely be a difficult and long-term process with no presently available guarantees for success. The ideal AIDS antiviral drug must fulfill a number of requirements. It must be conveniently administered, preferably orally, and it must be sufficiently nontoxic to be used for prolonged periods perhaps for a lifetime by asymptomatic individuals. In addition, it must not only be active in peripheral lymphocytes but also in the central nervous system, as HIV may infect the nervous system early in the disease process. Clearly, the ideal AIDS antiviral drug has not yet been identified, but several drugs are under clinical evaluation. The testing of these agents may be of value in developing subsequent, more useful agents. · The committee recognizes the urgent need to develop and test experimental agents for the treatment of HIV infection and resulting clinical disorders and the need to make active agents widely available as soon as possible. The committee believes that randomized clinical trials using a placebo control group are required until the first agent is identified that is both safe and effective. After an effective agent is identified, newer drugs can and should be compared to it rather than to a placebo. The committee agrees strongly with current practice whereby subjects partic- ipating in any controlled trial who have been in the placebo groups should be offered subsequent treatment with the more active agent if one is identified. The committee recognizes the desire of some to forgo use of placebo controls and immediately test experimental drugs against one another. A compelling argument against such a stratagem is the possibility that promising agents will turn out to be harmful to patients. This was, in fact, the unfortunate experience with suramin in patients with AIDS and ARC. When drugs may offer slight, yet clinically important, benefit or harm, it is especially important to establish that the benefits outweigh the risks in a placebo-controlled randomized trial. This is the quickest, most efficient, and least-biased way to ensure the most efficacious treatment possible for present and future AIDS patients. Many patients, including those with ARC as well as AIDS, have a good intermediate prognosis (up to several months), and these patients deserve to have access to drugs with documented benefits that outweigh toxicity. · Decisions on the design of studies to test new drugs for HIV infection must be made on a case-by-case basis. Such decisions should take into account the results of further studies on the efficacy and toxicity of AZT,
220 CONFRONTING AIDS the category of patients to whom the drug under consideration would be given, and preliminary information on the safety and efficacy of the drug. · The ethical aspects of the design of antiviral drug and vaccine trials should be kept under review by the National Commission on AIDS proposed in Chapter 1. · Efforts should be undertaken now to ensure that appropriate levels of organizational and financial support are in place to permit the expeditious drug evaluation through phase I, II, and III clinical trials of promising therapeutic agents against HIV infection as they become available. · A mechanism for providing equal patient access to clinical trial enrollment without regard to area of residence and patient demographics is most desirable. A system such as Physicians' Data Query, which provides on-line access to information about existing cancer chemother- apy trials, should be put in place. However, it may not be realistic to expect that all patients with AIDS or ARC can be accommodated in clinical trials. If such accommodation is impossible, policies providing access to trials should be as equitable as possible. In analogous situations, age, stage of disease, overall health, and absence of specific contraindica- tions to the use of the experimental drugs have been used as entry criteria. This is a complex issue that should be addressed by ethicists and experts in the conduct of clinical trials. · The magnitude and importance of the AIDS epidemic dictate that all of the potential technical research and development capabilities of governmental agencies, university researchers, pharmaceutical compa- nies, research institutes, and other organizations from the private sector become active in the overall drug development effort as a major approach to the treatment of HIV infection. · A conference should be convened as soon as possible to bring together researchers from industry, academia, and the Public Health Service to consider the key issues necessary for the development of antiviral drugs. Such a meeting could consider the unique capabilities of NIH including its drug screening program, its treatment evaluation units, and its provision of special animal models to facilitate industry's development efforts. Such a conference could also review the experience gained to date from clinical studies with antiviral drugs and strategies for development based on knowledge of the molecular biology of HIV. · Cooperative and coordinated efforts should be undertaken in the realms of basic and applied research to provide a solid experimental foundation for future drug development efforts. · Government-sponsored in vitro and in vivo testing and evaluation programs for potential agents active against HIV should be established. These programs should include antiviral agents from all sources, includ- ing the private sector.
FUTURE RESEARCH NEEDS 221 · Mass screening approaches to identifying drugs have in the past not been very productive. The committee favors a more rational approach based on the selection of unique viral processes or proteins as drug targets. · NIH-sponsored drug development involving multidisciplinary efforts should continue to be strongly supported for the discovery of new agents active against HIV. · Guidelines should be established regarding the acceptability of new antiviral agents for clinical evaluation against AIDS. These should include the efficacy of in vitro and in viva models, toxicology and preclinical pharmacology, and the effects of new agents on the immune system. · Studies to establish, validate, and standardize measures of relevant parameters of the human immune response should be actively encour- aged. Likewise, techniques and facilities to definitively detect HIV infection and improve the isolation of viruses from infected persons should be developed. In the absence of such improvements, the thera- peutic efficacy of novel agents (drugs and vaccines) may be difficult to establish, validate, or compare with alternative treatments. VACCINES Development of an effective vaccine to prevent HIV infection must be a prominent goal in any program designed to halt the continuing spread of the AIDS epidemic. However, it is also likely to be one of the most difficult to realize. Active immunization has proved to be an extremely effective means to limit or eliminate the exceptional morbidity and mortality inflicted upon human populations by many types of viruses, but the derivation of an effective vaccine against a human retrovirus has never been seriously attempted, much less achieved. Similarly, experi- ence in the production of vaccines for retroviruses of other animals has been rather limited and often disappointing. Programs to develop an HIV vaccine face many difficult challenges. Biologically, the characteristic genomic diversity and persistence of infection by HIV may present serious obstacles to the generation of broadly effective immunity. Vaccine development is also constrained by the presently limited basic understanding of the immune response to HIV infection, its apparent impotence in clearing the viral load, and the ways it might be bolstered through protective immunization. Should the biological and scientific obstacles be surmounted, there remain a number of other factors that may delay or limit the availability of an HIV vaccine. As discussed earlier in this chapter, the scarcity of available chimpanzees to test the safety and efficacy of candidate vac- cines may compromise or delay adequate preclinical evaluation. Initiation
222 CONFRONTING AIDS of testing in human populations will also present serious ethical and practical considerations, which will undoubtedly affect the clinical eval- uation of an HIV vaccine. Traditionally, vaccine manufacturers have looked to the federal gov- ernment to develop the basic understanding of the pathogen and disease before they become actively involved in vaccine development. Their decision to become involved often depends on a balance of economic and social factors. Given the extremely high cost of vaccine development programs and the present concerns over liability for vaccine-related injuries, many manufacturers may be unwilling to initiate or pursue the derivation or distribution of a vaccine to prevent AIDS. Animal Retrovirus Vaccines In the course of natural infections, certain retroviruses establish transient infections that are eventually cleared by the host immune system through the development of protective immunity. Other retrovi- ruses establish infections that persist in the face of the host's active but ineffective immune response. Viruses of the former category include members of the oncornavirus subfamily, as may be typified by feline leukemia virus, while retroviruses of the latter category include all known members of the lentivirus subfamily, including HIV. (Classification schemes for retroviruses may, in the near future, be undergoing revision based on better knowledge of their genetic relationships.) Feline leukemia virus (FeLV) is a naturally occurring retrovirus that is relatively common in domestic cats and causes a spectrum of pathologic consequences, ranging from inapparent infections to immunosuppression and leukemia. Many naturally infected cats recover and are immune from subsequent reinfection. Various approaches have been pursued in the development of an effective vaccine against FeLV, and some have achieved a high level of efficacy. Although early vaccines against FeLV infection were not completely protective (Jarrett et al., 1975; Yohn et al., 1976) and some actually enhanced the process of infection (Olsen et al., 1977), recent preparations induce protection from laboratory or experi- mental challenges (Lewis et al., 1981; Olsen et al., 1980; Osterhaus et al., 19851. Effective vaccines have been formulated consisting of prepared aggregates of the viral envelope glycoproteins shed from infected cells and combined with immunostimulatory adjuvants. Protection from FeLV infection has been found to correlate with the presence of virus-neutralizing antibodies directed against the viral envelope glycopro- tein (Hardy et al., 1976; Lutz et al., 19801. Similar approaches have been pursued in mouse model systems, in which vaccines have been developed that effectively prevent infection and pathology induced by certain murine . . . .
FUTURE RESEARCH NEEDS 223 leukemia viruses (Hunsmann et al., 19811. In the murine retrovirus systems as well, protection from infection can be induced by immuniza- tion with preparations of viral envelope glycoproteins designed to elicit neutralizing antibodies. In contrast to the feline and murine oncornaviruses, the development of effective vaccines against other types of retroviral infections has been very difficult, although such efforts have been under way largely in just the last few years. The most elusive category includes the viruses that, like HIV, cause chronic lifelong infections. Consideration of these viruses highlights some of the practical and theoretical difficulties in preparing an HIV vaccine. For instance, bovine leukemia virus (BLV) is a retrovirus that results in chronic latent infections of cattle. It is thought to spread among animals by the transmission of virus-infected cells. Although not phylogenetically related to HIV, it is related to the other T-lymphotropic retroviruses of humans, HTLV-I and HTLV-II.- Attempts to protect cattle from BLV infection through use of vaccines have had mixed, but disappointing, results. Immunization with purified inactivated virus in- duced antibodies that neutralized the infectivity of BLV in vitro and reportedly could protect from low levels (Patrascu et al., 1980) but not higher levels (Miller et al., 1985) of challenge with virally infected cells. Thus, neutralizing antibodies alone may not be sufficient to protect against retroviral infections transmitted through cellular intermediates. A similar situation may prevail in humans infected with HTLV-I, where high titers of neutralizing antibodies are detectable using sera from infected persons in in vitro assays, but the infection persists in vivo (Ho et al., 1985b; Robert-Guroffet al., 1985; Weiss et al., 19851. Experiences with protection from lentiviral infection through immuni- zation are extremely limited. Preliminary attempts involving immuniza- tion of goats with caprine arthritis-encephalitis virus have failed to produce protective immunity (McGuire et al., 19861. Although disappoint- ing, this result is too limited in scope and in its demonstrated relevance to HIV to draw conclusions about the feasibility of an AIDS vaccine. As discussed above in the section on "Animal Models," the ungulate lentiviruses may provide valuable model systems for basic studies con- cerning an AIDS vaccine. Given appropriate experimental support, increased activity can reasonably be expected in this area. Recent advances in the isolation and characterization of a number of previously unknown primate retroviruses oBer potential avenues for explorations of biological issues relevant to an HIV vaccine. Although not closely related to HIV, a type D primate retrovirus has been isolated that causes severe syndromes of immunosuppression and chronic wasting (Desrosiers and Letvin, 1986~. Viral infection and disease resulting from inoculation with this virus can be successfully prevented by active
224 CONFRONTING AIDS immunization (M. Gardner, University of California at Davis, personal communication, 19861. In addition, simian immunodeficiency viruses may provide excellent animal models for HIV infection and AIDS. Attempts to prevent SIV infection and disease in rhesus macaques are currently in progress (N. Letvin, New England Primate Center, personal communication, 19861. Study of the natural history, virology, and immunobiology of the recently discovered human retroviruses HTLV-IV and LAY-2 may also yield important insights to further the HIV developmental efforts (Kanki et al., 1986~. For instance, because infection with HTLV-IV reportedly does not cause any obvious disease, epidemiologic studies of its distribu- tion in Africa may indicate whether prior HTLV-IV infection protects against superinfection with HIV. The mechanisms by which the lentiviruses evade the immune response of their hosts appear rather diverse, and it is not clear which of these lentiviruses, if any, employ tactics similar to HIV (Haase, 1986; Narayan, 1986~. The visna-maedi virus causes chronic progressive interstitial pneumo- nia and a severe demylinating encephalomyelitis in sheep. Persistent infec- tion of sheep is accompanied by the progressive accumulation of mutations in the envelope glycoprotein, resulting in a process of antigenic drift wherein novel viral variants emerge that can escape neutralization by previously existing antiviral antibodies (Narayan et al., 1978; Thormar et al., 19831. Although new antigenic viral variants arise in the course of infection by visna viruses, the original viral strain is not replaced by them, suggesting that antigenic variation is not necessary for the maintenance of a persistent infection. Rather, the low titers and low affinity of virus-neutralizing anti- bodies seen following visna virus infection appear unable to prevent viral infection and spread (Kennedy-Stoskopf and Narayan, 19861. Equine infectious anemia virus, which can maintain a chronic infection in horses and results in intermittent episodes of acute fever, weight loss, and anemia, appears to employ a different mechanism of persistence. Antigenic variation in the envelope gene of this virus also occurs, but unlike with visna the new variants that emerge in a cyclic fashion during the course of an infection escape the host immune response while previous types are replaced (Montelaro et al., 1984~. However, between cycles of viremia and neutralization, viral persistence is maintained by latent infection of macrophages (Cheevers and McGuire, 19854. Caprine arthritis-encephalitis virus, which causes a progressive leukoencephalomyelitis in goats, appears to use yet another mechanism. In this case, infection of susceptible animals fails to elicit detectable neutralizing antibodies, and, in the absence of an effective immunity, viral infection may persist unabated (Narayan et al., 19841.
FUTURE RESEARCH NEEDS 225 Vaccines Against HIV Virus-neutralizing antibodies are detectable in persons infected with HIV, although they are present in rather low titers (Ho et al., 1985b; Robert-Guroff et al., 1985; Weiss et al., 19851. As discussed above in the section on "Natural History of HIV Infection," these antibodies are most frequently detected using in vitro assays of the inhibition of free-virus infectivity. Should HIV be transmitted via infected cells, as appears likely, the relevance of the presently used neutralization assays is unclear. Similar assays are poor predictors of protection or immunity from lentivirus systems (Haase, 1986; Narayan, 19861. The developmen- tal effort for an HIV vaccine is handicapped by the lack of a meaningful in vitro measure of immunologic protection from infection. The extent of antigenic variation generated during lentivirus replication presents a major concern for development of an HIV vaccine. Isolates of HIV derived from different individuals demonstrate substantial variation in the nucleotide sequences of their envelope glycoprotein genes, al- though it is not yet known if these are translated into biologically significant antigenic variations (Coffin, 19861. Preliminary studies indicate that detectable immunologic differences do exist between HIV isolates; antibodies raised in animals and directed against the envelope glycopro- tein purified from virions (Matthews et al., in press) or produced by recombinant DNA methods (Berman et al., 1986) will neutralize the virus type used for immunization but not divergent isolates. Independent isolates from single infected humans also demonstrate genomic variation of a similar type, but to a lesser extent (Hahn et al., 19861. It is not yet known if the observed genomic variation of HIV is involved in the virus's resistance to immune clearance, but the existence of many different strains of HIV creates a major potential difficulty for the generation of a broadly cross-reactive and protective vaccine. The regions of the viral envelope gene that are well conserved between isolates may provide targets for protective immunization, although their immunologic signifi- cance has not yet been established. Similarly, there seems to be less variation in HIV's internal structural and regulatory proteins. The internal proteins probably have functions that do not allow so much variation in protein sequence. Unfortunately, these proteins may not be readily available to neutralizing antibodies because they are localized within cells or virions. Preliminary reports suggest that sera from HIV-infected persons are able to neutralize variant isolates in vitro (Robert-Guroff et al., 1985; Weiss et al., 1985), but this result requires more extensive survey and exploration to establish its in viva relevance. Evaluation of the spectrum
226 CONFRONTING AIDS of protection provided by vaccine candidates will require testing in relevant animal models, primarily chimpanzees. Although most attention has focused on the antibody response against HIV as a potential agent of protection from infection, evaluation of the cellular response to virally infected cells has not been explored in detail. Since the cellular arm of the immune response is thought to be most relevant to the elimination of virally infected cells, its recruitment may be the most effective target for vaccination strategies (Weissman, 19861. The cellular immune response is known to be involved in the immune clearance of a number of types of viruses. In several of these instances, including certain retroviruses, cytotoxic T cells recognize conserved viral internal proteins expressed on the surfaces of infected cells (Holt et al., 1986; Pillemer et al., 1986; Townsend et al., 19851. If a similar reactivity exists to the conserved core proteins in HIV-infected cells, the problems posed by envelope sequence variation may be less significant. Models of Vaccine Delivery If antigens capable of inducing immunity to HIV infection can be identified, possible systems for vaccine delivery must be considered. The vaccines currently in use in humans generally fall into the broad catego- ries of live or killed vaccines. Killed-virus vaccines may consist of either inactivated whole virus or viral subunits. Live vaccines against HIV would fall into two categories: those using attenuated live HIV or HIV-related viruses and those using attenuated live viral vectors into which the genes coding for appropriate HIV antigens have been inserted. Because retroviruses in general and HIV in particular have a pro nounced tendency for recombination and mutation, live attenuated retro- viruses may be likely to revert to virulent forms. Problems of reversion have been overcome in the development of other live vaccines-for example, poliomyelitis. Although the problems of ensuring the safety of a live HIV vaccine would be very difficult, they should not be assumed to be insurmountable. However, other candidates, if promising, might be more attractive. Live viral vaccines, such as vaccinia virus, carrying inserted genetic sequences encoding HIV antigens have been considered for use (Chakrabarti et al., 1986; Hu et al., 19861. Even if excellent expression of appropriate viral antigens were obtained, there are concerns about the use of these antigens that might require extensive research, development, and testing to resolve. Vaccinia vaccination with most strains may itself cause rare but serious complications, especially when given to persons who are immunocompromised. These concerns may be compounded if the effec
FUTURE RESEARCH NEEDS 227 tive immunogen (e.g., the envelope glycoproteins) exerts additional cytopathic effects on exposed cells. Killed-virus vaccines may consist of inactivated viruses, of antigens extracted from the whole virus, or of antigens produced in the laboratory through recombinant DNA technologies or chemical synthesis. Killed whole HIV might not be effective against the antigenically diverse spectrum of HIVs unless an appropriate mixture of viral strains could be identified and included. This approach seems unlikely at present, given the lack of a clear definition of the extent or significance of viral heterogeneity. As noted above, the most likely approach to an HIV vaccine is through the discovery of meaningful, broadly protective immunizing antigens and epitopes and their expression by recombinant DNA technology. Chemi- cally synthesized peptide antigens, which are most useful in defining epitopes that are related to immunity, suffer from poor immunizing capability and are less ready for practical application than are more complex antigens made by recombinant DNA technologies. The systems presently used for the production of recombinant proteins employ bacte- ria, yeast, or mammalian cells. However, the candidate immunogens, including the external glycoproteins of HIV, are heavily glycosylated, which may influence the choice of organism used to produce antigens. Bacteria do not glycosylate recombinant proteins, and yeast do so quite differently than do mammalian cells. If glycosylation of an HIV antigen is necessary for appropriate immunogenicity, then production in mamma- lian cells may be required. Recently, the production of a large segment of the HIV envelope glycoprotein has been achieved in mammalian tissue culture cells (Berman et al., 19861. But antigen production in mammalian cells, however useful, is plagued by continuing concerns over safety. The Food and Drug Administration has approved Chinese hamster ovary cells for the production of an investigational hepatitis B vaccine. These cells have also provided usable substrates for HIV envelope production (Berman et al., 19861. Anti-idiotype antibodies, which are produced by immunization against the variable region of specific antibodies and present an image of the original antigen, have also been explored for vaccine potential, because it is possible that the important epitopes of HIV may be discontinuous and require close proximity of more than a single continuous amino acid sequence. In the practical sense, however, there is no present example of a human anti-idiotype vaccine, just as there is no example of an adequate synthetic peptide vaccine for human use. Any killed-virus or subunit vaccine against HIV would probably require, or at least benefit from, immunopotentiation (optimal antigen presentation) by being combined with adjuvants. Although alum is
228 CONFRONTING AIDS accepted as a safe adjuvant, if it proves insufficient other approaches will require evaluation and approval. Approaches to HIV Vaccine Development and Evaluation Testing of an HIV vaccine in human populations will present a number of difficult logistical challenges. Initial vaccine evaluation must address issues of safety and immunogenicity and is traditionally carried out in volunteer populations not at risk of infection so as to avoid confusion with immune responses that would accompany infection. This is particularly a problem when the immune response to the vaccine would be difficult to distinguish from that to a natural infection, as would likely be the case with inactivated- or attenuated-virus vaccines. With subunit vaccines, probably derived from recombinant DNA technology, the immune re- sponse would be distinguishable from natural infection. Some have argued that it might be possible for this reason to compress the usual evaluation schedule to use those at risk of infection in early safety and immunogenicity trials. This issue needs further and early discussions as actual vaccine candidates appear more promising. The selection of populations for subsequent testing for vaccine efficacy will require even more careful advance planning. To permit trials of manageable size, populations must be identified that demonstrate a significant rate of incident HIV infections. These test populations should at the same time be representative of other target populations if the results obtained in them are to be broadly relevant. Rigorous determinations of the immunologic and virologic status of participants will be necessary at the outset to ensure that they have not already been exposed to HIV. Persons who are infected with HIV but not yet seropositive may present a difficult confounding variable. Testing of vaccine efficacy will require double-blind, randomized, placebo-controlled trials. However, the nature and severity of HIV infection may make it difficult to design such trials so that they meet the standards of ethics commonly accepted or required for research. It can be argued that allowing susceptible persons at high risk of HIV infection to persevere in their high-risk behaviors is distinctly unethical when health education could lessen their likelihood of HIV infection and disease. Participation in the vaccine trials may also positively or negatively affect the frequency of high-risk behaviors practiced by test subjects, influ- encing their rate of incident infection independent of vaccine effective- ness. These difficulties must begin to be addressed now. Another difficulty involves the unwillingness of many pharmaceutical companies to commit to a significant financial and scientific investment in vaccine development in the face of present liability threats and insurance
FUTURE RESEARCH NEEDS 229 considerations. Concerns exist over liability both during the testing phase of vaccine development and after a vaccine is licensed but are perhaps greater during the latter period. Unless problems of vaccine liability are dealt with swiftly and effectively, no manufacturer may be willing to produce HIV vaccine for use in the American market. The liability issue is critical, but the degree of its impact will depend on the proposed approaches and target populations for vaccine development. Should a vaccine be targeted solely at defined groups at high risk, specific statements of risk-benefit analysis might serve to lessen liability concerns. Perhaps a vaccine candidate, once proven useful, could be made available to significant numbers of individuals as a subsidized, semipermanent, investigational product. However, any proposed strategy of general use- whether for adolescents or for the general public immediately raises the multitude of problems experienced during the mass immunization initia- tive for swine influenza. Most notably, the temporal coincidence of diseases of unknown etiology with mass immunization programs predict- ably raises the level of litigiousness to a point where even federal resources for compensation of perceived vaccine-related injuries can be strained. In addition, the real or imagined risk of administration of an HIV vaccine might be considered much greater than for other vaccines. Similarly, the consequences of failure of vaccine protection might be considered more severe in the case of an HIV vaccine. These issues must be addressed in advance of any such efforts with respect to HIV vaccination. Conclusions and Recommendations Developing a vaccine to prevent HIV infection and AIDS presents a number of scientific challenges that have never before been responded to successfully. As a result, an effective vaccine may be very difficult, if not impossible, to produce. Should an effective vaccine candidate become available, there are significant social concerns that may limit or prevent its testing and use. Therefore, a vaccine may not be reasonably expected to be available in less than 5 years. Even for the next 5 to 10 years, the committee generally believes that the probability of a licensed vaccine becoming available is low. · An aggressive basic and applied research endeavor is essential to evaluate the possibilities and prospects for effective vaccines. It is clear that a successful vaccine development program will depend on a greatly expanded foundation of basic research knowledge concerning HIV. Because of the long developmental process for vaccines, basic and applied studies should be adequately supported and effectively organized.
230 CONFRONTING AIDS · Vaccine development programs will benefit from the active and interactive participation of government, academia, and industry. New methods to encourage the interchange of information and minimize proprietary considerations should be evaluated. · The committee finds that there has been inadequate federal coordi- nation of vaccine development. The NIH has recently reorganized its efforts on AIDS, and the committee encourages the appointment of strong leadership to the vaccine program, with authority and responsibility to develop a strategy for a broad-ranging vaccine development program. This program should take advantage of the strengths available within NIH, in the external scientific community, and in industry. · Industry is fearful of involvement in the development of vaccines because of the potential liability it must accept in distributing them. Creative options for the governmental support of industrial research, guarantees of vaccine purchase, and assumption of reasonable liability should therefore be explored. (The only alternative to commercial pro- duction is the establishment of government production facilities and total assumption of liability.) For instance, states could enact malpractice and product liability reform laws that would encourage programs of HIV vaccine development. The enactment of a reasonable financial limitation or "cap" on court and jury awards in personal injury and product liability cases arising out of the clinical testing of an HIV vaccine or out of the use of a licensed vaccine on the market may provide one reasonable ap- proach. · It is imperative that ethical and pragmatic problems be addressed simultaneously with scientific ones, so that possible success in developing a vaccine candidate can be exploited expeditiously in a way that meets ethical and legal criteria. The committee therefore recommends that its proposed National Commission on AIDS consider establishing a special subcommittee on the ethical, legal, and social issues involved in the testing of HIV vaccines. SOCIAL SCIENCE RESEARCH NEEDS As much as AIDS is a medical and biological issue, it also has important social dimensions. HIV infection is spread through particular types of behavior, and presently the best hope for stopping the epidemic spread of the virus is through changes in the types-of behavior responsible for its continued transmission. Yet the forces that shape human behavior, and the best approaches to influencing behavior to protect health, are among the most complex and poorly understood aspects of society's response to the AIDS epidemic. It is instructive to note that virologic research, having received reasonable levels of funding over the years, was well poised to
FUTURE RESEARCH NEEDS 23 ~ begin addressing the many biological problems posed by HIV. In con- trast, the knowledge base in the behavioral and social sciences needed to design approaches to encouraging behavioral change is more rudimentary because of chronic inadequate funding. This lack of behavioral and social science research generates some of the most important and immediate research questions surrounding the epidemic. Social science research can play a number of valuable roles in meeting the challenge of AIDS. It can help in the development of effective education programs to encourage changes in behavior that will break the chain of HIV transmission. It can contribute to the development of informed public policies that reduce the public's fear of AIDS and discriminatory practices toward AIDS sufferers. And it can guide the establishment of improved health care and social services that further the ability to treat AIDS patients effectively, humanely, and at reasonable cost. To date, there has been little social science research specifically focusing on HIV infection and AIDS, but there have been studies of the factors influencing behavior change, risk perception, attitudes toward civil liberties, tolerance and discrimination, communication, and the organization of health care that are relevant to the AIDS epidemic. These studies suggest avenues for research more directly related to AIDS. Such research can be useful in guiding short-term administrative and social responses to the epidemic and in providing the clarification necessary for developing longer-term measures for coping with AIDS and, more gen- erally, with future health crises. Breaking the Chain of Transmission As discussed in Chapter 4, educational efforts aimed at providing accurate information to persons at risk of infection are currently the best available public health measures to stop the spread of HIV infection. However, a great deal must be learned before the optimal educational approaches can be derived, accurately targeted, and effectively transmit- ted. The literature on behavioral risk modification generally concedes the extraordinary difficulty of modifying behavior, even when there is clear demonstration of risk (Leventhal and Cleary, 1980~. Much of this research focuses on alcoholism, cigarette smoking, and dietary risks; it is therefore only indirectly related to the risks of infectious disease. Even research on other sexually transmitted diseases is not entirely analogous to AIDS, because of AIDS' exceptionally dire consequences and the lack of effective medical therapies. In all of these areas, a common assumption (and a common assumption in discussions of AIDS) is that repetitive media information will induce
232 CONFRONTING AIDS people to change their behavior to avoid risk. While research on this hypothesis is scattered and results are inconclusive, several useful points emerge. There are areas where extensive media reports (for example, those relating cholesterol-producing foods with heart disease) have con- tributed to positive changes in consumer behavior over a long time period. Similarly, media publicity about the possible adverse effects of birth control pills and intrauterine devices (IUDs) also resulted in a significant decline in their use (Jones et al., 19801. However, these and other examples of behavioral responses to information are all in areas where alternatives are easily available. By the same token, the risks may be lower and less poorly defined in many of these areas as compared with the risks associated with HIV infection. Deliberate efforts to use the media to influence behavior have not necessarily achieved their desired result. Despite extensive public infor- mation about the Salk polio vaccine in the late 1950s when it was first available, relatively few individuals agreed to be vaccinated at that time (Robinson, 19631. Similarly, media coverage of the 1964 Surgeon General's report on smoking and cancer had little direct effect on smoking habits in the short term (Troyer and Markle, 19831. Although people seek information to guide even the most personal decisions, they use this information mainly when it corresponds to prior inclinations or when it is reinforced by their social situation and the beliefs and attitudes of their reference groups. Review of the literature on the effect of risk information on smoking behavior shows that communication is generally ineffective unless it is directed to specifically defined target populations, contains an "action plan," comes from credible sources, and is combined with community and peer support (Leventhal and Cleary, 1980~. Other analyses (Lichtenstein, 1982) emphasize the importance of peer leaders and social support networks in effecting behavior change. Studies of drug treatment and rehabilitation programs confirm these findings. For example, a study of narcotics treatment programs by Caloff (1967) identified different types of narcotics users and stressed the need to adapt information and rehabilitation programs appropriately. In a study of methadone mainte- nance programs, Nelkin (1973) indicated that without an adequate social support system and appropriate changes in the environment, the effect of methadone on the behavior of heroin addicts has been limited. The importance of group support and pressure has been evident in the relative success of groups and programs such as Alcoholics Anonymous in changing life-styles (Roman and Trice, 19721. Research in other social science areas yields similar results. Persuasion research (often performed in laboratory settings) offers compelling evi- dence of the importance of group pressure and credible sources as
FUTURE RESEARCH NEEDS 233 variables in influencing behavior (Bostrom, 1983; Rolod and Miller, 19801. Similarly, research on the diffusion of innovation finds that acceptance of new ideas depends on the role of personal networks and opinion leaders who are trusted, especially when innovation requires changes in social or cultural norms (Rogers, 19831. This research consistently emphasizes, first, the importance of shaping information appropriately for specific groups and distributing it through credible and trusted sources within the target community. Second, it points to the need for providing social reinforcement to maintain behavior change. Enough is known to suggest that education and other techniques of persuasion may achieve success in inducing behavior change if pursued intensively and systematically (Farquhar et al., 1984), but it is also necessary to strive to improve these intervention strategies. Several directions for social science research specifically relevant to AIDS and HIV transmission need to be pursued, in addition to the evaluation of the effectiveness of various education programs described in Chapter 4. Knowledge acquired through this research will enable interventions to be better designed and directed. A first necessary task would be to develop a demography of HIV infection that would identify not only high-risk groups but also spouses, sexual partners, children of infected parents, family members, and so on. In essence, a more detailed, representative, and contemporary evaluation of sexual behavior analogous to the Kinsey Report is needed to assess the range and varieties of sexual behaviors in both the homosexual and heterosexual communities. Increased knowledge of sexual behaviors and the factors that affect those behaviors will be necessary to design improved approaches to inducing behavior change. Areas for study include the development of sexual orientation, the selection of sexual partners and practices, and choices about methods for safeguarding health or preventing pregnancy in various groups. It is desirable to gather information on these topics not only for adults but also for adolescents. Additionally, it is essential to know the size of those groups that engage in various sexual practices. It would then be essential to map out the existing and potential sources of health information available to these groups and to target credible sources (i.e., ax-addicts, organizations within the homosexual community, churches, social networks, neighbor- hood groups) that will be trusted as conveyers of information and that can over social support. Similarly, more needs to be known about the reasons why people begin using drugs, especially IV drugs, about the influences on choice of drugs and routes of administration, about what motivates cessation of drug use, about the factors that can reinforce choices to quit, and about the
234 CONFRONTING AIDS variation of these in various groups of drug users. Influences on behavior in adolescents as well as adults should be studied. International compar- isons will be necessary to ascertain the best methods for combating drug use and the efforts of various interventions (e.g., needle availability) in reducing HIV transmission, since the full range of methods cannot be tried in any one country. It will also be very important to study the social dynamics, rituals, and practices of various risk populations. The point of such studies would be to analyze and develop effective means to reach people at risk, delineate the obstacles to behavior change (for example, rituals concerning the sharing of needles and syringes among IV drug users), and determine an effective language and style of communication. The use of appropriate language is particularly critical given the diverse background of the populations at risk. What is meaningful to the gay community may be quite irrelevant for IV drug users. Just as AIDS clearly affects certain high-risk groups, it also has the potential to affect the whole of society. It is necessary to reach individuals in groups where the prevalence of HIV infection is presently high, but it is also imperative to reach individuals in lower-prevalence communities, many of whom do not consider them- selves at risk of infection or AIDS. Information aimed at high-risk groups may in delivery and content miss other persons potentially at risk, such as men who consider their bisexual behavior as safe, sexual partners of IV drug users or bisexual men, heterosexually active persons in high-risk areas, or the clientele of prostitutes. Specifically targeted information will not be enough to stop the spread of AIDS. Novel approaches must be evaluated to reach and inform all individuals and social groups. Another type of research should monitor the adoption of safer sex practices and other positive changes in behavior by those at risk. Similarly, studies of sex education programs should be under way to evaluate the means of communicating information about sex, the effec- tiveness of various approaches in fostering protective behavioral change, and the factors that make such programs politically viable as well as effective. Finally, social experiments could be devised to develop strat- egies for intervention (e.g., information campaigns, distribution of free sterile needles and syringes to IV drug users) to evaluate their effective- ness and to better understand the variables that encourage safe behavior. Reducing Public Fear and Its Effects The public response to AIDS has been one of fear, often reflected in excessive caution, discriminatory behavior, and recommendations for drastic policy measures that are unwarranted in terms of what is known about the actual risk. As described in Chapter 4, national polls of public
FUTURE RESEARCH NEEDS 235 attitudes about AIDS indicate that most people are highly attentive to news about AIDS and that many are inclined to overestimate risks (Singer and Rogers, 19861. Important dimensions of the public response to AIDS are suggested in a study by McClosky and Brell (1983) of tolerance in the United States. The study addresses the fragility of tolerance; most Americans voice support of civil liberties, but they often reject concrete applications. The study found persistent intolerance and fear of homosexuals and only fragile respect for their rights, a fact that certain groups concerned about "moral erosion" are able to exploit politically. Finally, the study suggests that opinion leaders clergy, media, community leaders are likely to be more tolerant and protective of civil liberties than is the public at large. An important research question involves the dynamics of discrimina- tory practices and behavior relating to HIV infection and AIDS. For instance, why do some schools or workplaces exclude AIDS sufferers, while others accept and help them? Because factors other than the actual extent of risk enter into public perceptions, studies of the response to AIDS should probe well beyond people's factual understanding. How and by whom were employers or colleagues informed? How do the back- grounds of the persons involved in a dispute influence their perceptions? What were their prior relations in the work setting? Do different re- sponses reflect the structure of the social setting or the mode of dissem- inating information? The object of such research is to develop effective means of public communication that will counter discriminatory practices and lead to workable decision-making procedures predicated on the most accurate scientific understanding available rather than prejudice resulting 1 from inadequate or inaccurate information. Public attitudes that tend to overestimate the risks surrounding AIDS are consistent with research findings about public perceptions. Studies of risk perception suggest that the public underestimates familiar risks and overestimates those that are unfamiliar, involuntary, invisible, and po- tentially catastrophic (Fischoff et al., 19814. Fear is often attributed to people's inability to deal with uncertainty. However, this generalization has been questioned by those who observe the considerable ability to deal in a probabilistic manner with uncertain variables in many occupations (Douglas, 19861. AIDS appears to be a new problem, one of uncertain origin and rapid but silent spread. It thus presents enormous challenges regarding the adjustment of public perceptions to the accumulating medical and scientific understanding of AIDS and HIV transmission. Sociological studies also suggest that perceptions of risk are heavily influenced by political and social attitudes. Thus, different cultures or social groups will emphasize certain risks and minimize others (Douglas and Wildavsky, 19821. Perceptions of risk are also closely connected to
236 CONFRONTING AIDS legitimizing moral principles (Berger and Luckman, 19661. In the United States, the strength of religious feeling and morality has turned issues of abortion, animal welfare, evolution, and reproductive technologies into major policy disputes and has of course influenced the public response to AIDS as well. (It should also be noted that some churches have been in the forefront of promoting a positive, constructive, and compassionate attitude toward AIDS sufferers. They can play an important role in educating their members, as mentioned in Chapter 4.) Finally, risk research indicates that political factors, in particular trust in authority and expertise, also shape perceptions and fears (Nelkin and Brown, 19841. Mistrust of experts and failure to distribute information through locally available and respected community leaders have contributed to public fear of AIDS for instance, in the exacerbation of disputes over allowing children with AIDS to attend public schools (Nelkin and Hilgartner, 19861. A rhetoric of accountability and blame often pervades the discourse about risks. Many have observed the tendency to blame the victim as a means of silencing indictments of the social order. The sick are often blamed for their ill health, the poor for their economic plight (Navarro, 1977; Donzelot, 19791. For instance, there has been a striking contrast between media accounts of AIDS patients in pediatric and transfusion- associated cases, who are often portrayed as "innocent victims," as opposed to the depiction of implicit responsibility in cases among homosexual men and IV drug users. Efforts to influence risk perception usually begin with proposals to improve public education (Kunreuther, 19781. Yet many studies suggest that educational measures alone do little to change public attitudes (Slovic et al., 19811. People interpret information in highly selective ways, a fact well documented in the studies of the effect of media information on attitudes (Klapper, 19601. Information from the media is absorbed and assimilated in different ways depending on the predispositions of the reader, the influence of peers and opinion leaders, and the trust in the credibility of sources (Cohen, 19641. However, the press does help to establish a framework of expectations, so that isolated events take on meaning as public issues. By widening the base of public information, the press defines a context for public policy and forces a policy response (Lang and Kurt, 1983; Tuchman, 19781. A major challenge is to provide information in a way that will create the sense of urgency necessary without causing undue panic. The communi- cation literature suggests that effective information campaigns must distribute the judgments of experts through credible and trusted opinion leaders who can effectively channel information to various targeted communities.
FUTURE RESEARCH NEEDS 237 Organizing Health and Social Services There are several important directions for research on providing services to AIDS patients. A necessary first step will be to map the now poorly defined multiplicity of state, federal, and local agencies and private and public institutions involved in health and social policies relating to AIDS. Goals of such a project would be to identify areas where institu- tional fragmentation or professional specialization are counterproductive, to understand the effects of the present system of financing medical care and social services, and to develop better and more cost-effective means of coordination. Similarly, there is a need to document and evaluate the various types of services that are presently available to persons infected with HIV in different communities, ranging from hospitals to terminal care facilities and home services. It will be important to evaluate their effectiveness, their acceptability to patients and the community, their cost, and their distribution of costs. Comparative studies of hospital management practices-for example, the allocation of beds, the distribution of resources within a hospital, and the ways in which the staff copes with the stress of working with patients with a fatal and intractable disease- could be useful in improving the effectiveness of hospital practices. Social experiments based on different models of patient care developed in high-incidence areas (see Chapter 5) will permit evaluation of their applicability to other areas and will provide a foundation on which to build locally relevant programs. Studies are also needed that recognize the difficulties HIV-infected patients have in gaining access to services, the effects of homophobia and negative attitudes toward drug users, and the conflicts that may arise between individuals in high-risk groups over access to care. The problem of providing treatment and care to IV drug users is especially intractable but has been virtually ignored as a focus of research since the develop- ment of methadone and other treatment programs. Finally, understanding the policy and social service response to AIDS would be enriched by comparative international studies that could pro- vide alternative models and new ideas. What legal, social, fiscal, and administrative structures have been instituted in response to AIDS in different countries? How do those countries provide services for patients; report, contain, or restrict high-risk groups; try to protect the general population; or deal with testing, blood donation, employment, and the allocation of resources for health services and research? How do their national policies and practices reflect the legal context, the cultural and social definitions of disease, the cultural biases about homosexuality and drug abuse, concepts of civil liberties, and the general structure of their social and medical services? If not alternative models, such comparative
238 CONFRONTING AIDS studies could at least provide new ideas for approaching AIDS within the United States and fostering international cooperation. Conclusions and Recommendations Because effective drugs or vaccines to counter HIV infection are not now available, effective educational interventions are essential to limit the spread of the virus. However, previous experience with educational programs to promote behavioral change for disease prevention is rather discouraging. There is a great need to explore novel approaches in AIDS-related education and to vigorously examine the relative efficacy of various approaches. Social science research should be directed toward the following goals: · Establishing the demography of heterosexual, homosexual, and IV drug use behaviors and the characteristics of the groups that practice different patterns of behavior; · Identifying credible information sources for various groups and opinion leaders; · Conducting experiments and demonstration projects on approaches to behavior change to understand what does or does not work; · Tracking discriminatory practices and their dynamics (Why is dis- crimination a problem in some areas and situations and not in others?; · Evaluating treatment, social service programs, and hospital manage- ment practices to understand what works and is cost-effective; · Studying the special problems of caring for addicted individuals and preventing transmission among IV drug users; 0 Linking experimental educational programs with epidemiologic eval- uation of their effectiveness in reducing the rate of seroconversion across the spectrum of populations at risk; · Making comparative studies of international responses to the epi- demic; · Studying the public's understanding of and attitudes toward AIDS and related issues in order to better design interventions to promote accurate awareness; · Studying and analyzing the ethical and legal aspects of the AIDS . . epl( .emlc. FUNDING FOR RESEARCH RELATED TO AIDS AND HIV The current national commitment to AIDS and HIV research must be considered in light of the nation's overall commitment to biomedical research. In 1985 the United States spent an estimated $13.5 billion for
FUTURE RESEARCH NEEDS 239 health research and development (Wyngaarden, 19851. The federal govern- ment contributed $6.8 billion-about one half of the total. The National Institutes of Health alone provided $4.8 billion, or 70 percent of the total federal investment in biomedical research. Thus, the NIH was responsible for over 35 percent of the funding from all sources in the nation. In 1985, industry spent about $5 billion on health research and development, or 37 percent of the national total. About $3.3 billion of industry's expenses were for biomedical research and development for human health. Large pharmaceutical firms contributed the major portion of this investment in biomedical research. A number of large firms have emphasized the new biotechnologies by creating new institutional ar- rangements. A variety of novel and innovative partnerships have been formed between industrial firms, private research institutes, and the academic research community. In addition, a number of relatively small research-intensive companies have arisen in recent years and have invested significant energies and funds in the new biotechnologies. Foundations and other private sources contributed about 8 percent of the total national effort in biomedical research. An important component of this contribution is research and training grants. Current Levels of and Mechanisms for Funding The total amount spent on research on AIDS and HIV is a very small portion of the overall national effort for biomedical research and devel- opment. The largest commitment has been made by the federal govern- ment through the Public Health Service (PHS), which includes the National Institutes of Health, the Centers for Disease Control (CDC), the Food and Drug Administration (FDA), the Alcohol, Drug Abuse, and Mental Health Administration (ADAMHA), and the Health Resources and Services Administration. The actual PHS expenditure in 1985 was about $108 million. Out of this, the NIH spent $63 million, the CDC $33 million, FDA $9 million, and ADAMHA $2.6 million. For 1986, Congress appropriated the following amounts to PHS agencies: CDC, $64.9 million; NIH, $140.7 million; FDA, $10 million; and ADAMHA, $12.7 million, for a total of $244.3 million (including expenditures in the Office of the Secretary of Health and Human Services). Centers for Disease Control Funds allocated to the CDC are used for many purposes, including research. Most research is applied, reflecting the CDC's primary role in disease detection, surveillance, and education. Each CDC grant for demonstration projects aimed at AIDS prevention must contain an
240 CONFRONTING AIDS evaluation research component, but the size and scope of these research components are difficult to assess. A number of cooperative agreements have been made between the CDC and various states and localities for specific epidemiologic studies of AIDS. Considerable central direction and protocol development are involved in these studies, but ideas for the studies may originate from persons outside the CDC's staff. The cooperative agreements are awarded after open competition based on published requests for proposals. National Institutes of Health Funds assigned to the NIH are distributed through institutes. Most AIDS-related funding is provided by the National Institute of Allergy and Infectious Diseases (NIAID) and the National Cancer Institute. Funds are distributed in the form of peer-reviewed research grants, contracts, and cooperative agreements and to NIH staff scientists. Other Federal and State Agencies The Food and Drug Administration has a key responsibility in the development of new therapeutics and devices such as the HIV antibody test. Recent changes in the agency are facilitating the investigation of new therapeutics for AIDS. In addition, the Alcohol, Drug Abuse, and Mental Health Administration will be handling social sciences research in all three of its institutes. The Department of Defense has a significant effort under way in monitoring military personnel who have been identified as being infected with HIV. Significant research capacity resides in military institutes such as the Walter Reed Army Institute for Research, some of which is being focused on AIDS. In FY 1986, approximately $37 million has been allocated to the Department of the Army for the support of AIDS-related research. However, the committee is not aware of projections for AIDS-related research funding in the military budget for future years. Some state governments have also become directly involved in the support of AIDS-related research activities, including those of California, Massachusetts, and New York. Other states are also considering the support of AIDS research, either as it relates to specific issues within their jurisdic- tions or as it relates to broader basic or applied research questions. Industry It is not possible to estimate the contribution by industry to AIDS research and development, but the amount must be only a small portion
FUTURE RESEARCH NEEDS 241 of industry's total biomedical research and development expenses. The smaller biotechnology firms have the manpower and capabilities to contribute much to this effort. A number of these companies have programs to develop AIDS vaccines or new diagnostic tests. Other small biotechnology firms have ongoing programs of research on interleukin-2, interferon, and various immunomodulators. The commitment of pharmaceutical and biotechnology companies to AIDS research must be considered in terms of the research and develop- ment investment risks involved. In a number of these areas, the devel- opmental costs are quite high and must be weighed against the anticipated commercial returns of new products. A number of large pharmaceutical firms also have research and development programs for diagnostic tests and new treatment and pre- vention strategies for AIDS. Some of these companies have been able to redirect their existing research capabilities to address the AIDS problem. Among the areas under investigation are possible antiviral drugs for individuals already infected with HIV, drugs for treating opportunistic infections, and new kinds of diagnostic tests. The creation, testing, and distribution of new drugs and vaccines require numerous and diverse types of expertise. Strategies for the development of agents to treat or prevent HIV infection and AIDS will greatly benefit from the most effective use of all of these needed resources. Large pharmaceutical firms can make a tremendous contribu- tion by investment of their demonstrated skill toward the evaluation, production, distribution, and marketing of candidate AIDS drugs or vaccines. Many smaller companies could also provide innovative contri- butions to the derivation of such agents. Thus, the involvement of commercial companies in vaccine and treatment efforts is highly desirable and may well be essential in the nation's response to the AIDS epidemic. In addition, a commitment to AIDS-related research is not understood fully by the management of most companies; usually, the motivation comes from the scientific stab. The effort is usually viewed as a public service, and the therapeutics being developed are perceived to have a limited orphan drug status. However, the development of new therapeu- tics for AIDS, ARC, and HIV infection, such as safe and effective antiviral drugs for seropositive individuals, would have an exceptionally large national and international market potential (see Chapters 3 and 71. Current NIH Funding Mechanisms A detailed analysis of the history and evolution of NIH AIDS funding was commissioned by the committee (Stoto et al., 19861. In FY 1982, 47 percent of NIH AIDS funds went to intramural research, and 39 percent
242 CONFRONTING AIDS went to extramural research grants. Since then, there has been an increasing emphasis on contracts and cooperative agreements, as work in more applied areas, such as clinical trials of new drugs, increases. Most of the extra congressional appropriation for AIDS in FY 1986 ($70 million) was for contracts or cooperative agreements. In FY 1986, $22.7 million (19 percent) of NIH AIDS funds went for support of new or continuing research grants (including some cooperative agreements); $28.1 million (23 percent) went for intramural research; and $64.1 million went for research contracts, not counting an additional $13.9 million still in the NIH director's office that will probably also go to contracts. It is thus apparent that, as a proportion of the total amount allocated, distribution by the contract mechanism has proportionally grown by the greatest amount, and well over 50 percent of NIH AIDS funds are now used to support contracts. Some of this increase is represented by the advent of funding of clinical testing programs for the evaluation of AIDS therapeu- tics, but this represents only a part of the total growth in this category. Between FY 1983 and 1986, there was a steady decline in the proportion of NIH funds spent for investigator-initiated research grants and an increasing proportion expended for NIH-designed contracts. This changing emphasis on the distribution of research funding is cause for some concern. Much of the most important research that needs to be addressed is of a basic nature. The major NIH support for basic research is provided through investigator-initiated, peer-reviewed grants. This system traditionally has been used to evaluate the scientific merit of a research proposal and, by comparing it with competing grants, to prior- itize the allocation of funds. It has been widely acknowledged as an effective means of promoting innovative, novel, and important work. This mechanism stimulates the broadest possible involvement of high-caliber nonfederal researchers in the development and evaluation of projects. With contract funding, the scope of work and approaches used tend to be specified by federal scientists, and the critique and prioritization of applications tend to involve scientists from the nonfederal sector to a lesser extent. Emphasis on the contract mechanism for the channeling of funds for AIDS and HIV research means that a lower proportion of the total funding is available for investigator-initiated proposals. Hence, fewer proposals and less interest are generated in the nonfederal research community. Additionally, the influence of that community an acknowl- edged major source of scientific expertise-in setting the national re- search agenda is substantially diminished. Notwithstanding the quality of researchers and policymakers in federal agencies, the problem of AIDS demands that the broadest range of expertise be involved. A more balanced growth of support is desirable in
FUTURE RESEARCH NEEDS 243 coming funding cycles to promote the involvement of the nonfederal research community to a greater extent. Reliance on centrally planned studies is advantageous in certain situations, particularly where many federal agencies are involved and where the problem clearly requires a well-coordinated effort. A central organization can be expected to be particularly useful when studies require many different types of expertise and where needed actions can be clearly defined. How- ever, an overcommitment to centrally planned studies may prove detrimen- tal to the nation's broader research effort and potential. If the tendency toward centralized research studies continues to the exclusion of the creative scientific input from researchers outside of the NIH, or if it results in the compromise of research funding for investigator-initiated, peer-reviewed grants, extremely deleterious consequences may result. The intramural and extramural NIH efforts must be mutually interactive and beneficial if the nation's research on AIDS and HIV is to be optimally productive. An additional problem with NIH's funding for AIDS research in recent years is that it has been derived largely through reprogramming of funds from other health areas (Krause, 1986; Stoto et al., 19861. These funds were diverted from ongoing NIH activities, in the form of personnel positions and research support, to satisfy expectations of progress on AIDS in the absence of a commensurate provision of adequate funding. As a result, many of the NIH's non-AIDS programs have been detrimen- tally affected. However, these areas have not diminished in importance or urgency. Hence, reprogramming of funds is not an appropriate response to research funding for AIDS and HIV, because it delays progress toward controlling other health problems and toward establishing the knowledge base needed to deal with AIDS. It is important that funds and personnel positions be restored to these areas. The current situation at the NIH and its resultant ability to respond to the AIDS epidemic are further constrained by a hiring freeze that affects the recruitment of new investigators. The essential point in this regard is that HIV infection and its conse- quences are a new and additional problem. Other problems have not diminished, nor has the public expectation diminished that they will be pursued actively by the biomedical research community. The ability to act flexibly to exploit new information and pursue new ideas is essential in this rapidly moving field. For the reasons noted above, future reprogramming of monies would damage the capacity to maintain stability of funding for other important areas. Congress and the administration should continue to be open to consideration of supplemen- tal requests to ensure that research on prevention and treatment of HIV infection is not delayed or other areas penalized. Decisions on the appropriate level of overall funding for research on AIDS and HIV need to take into account all of the criteria listed above,
244 CONFRONTING AIDS and no simple formula exists to assist this process. The National Commission on AIDS proposed by the committee could provide a useful service by monitoring the distribution and needs for research funding in these areas. Decision makers should keep these activities under review, and the total allocation of funds to research on AIDS and HIV should be reexamined periodically. Distribution of Funds Among Agencies and to Specific Research Areas The committee identified a number of research areas in need of support that fall under various agencies, but it did not attempt to assess the distribution of funds among agencies. Nor did it attempt to address in detail a variety of specific questions, such as the relative claims of research on vaccines versus drug development or research versus educa- tional efforts (in each case both efforts are justifiable). Although no comprehensive attempt was made to prioritize funding needs, some specific needs were thought to be particularly important. The committee noted a particular dearth of research in social science and behavioral research related to AIDS and HIV. It recommends that its assessment of research needs and funding levels act merely as a starting point for an extensive evaluation of these topics. Some funding for research on AIDS and HIV has, in recent years, been provided by states, and this has been critically important in certain areas. An assessment is needed of the respective roles and responsibilities of federal agencies, states, and other funding sources in the support of this research. Assessing Desirable Levels of Research Support The task of determining the appropriate level of federal funding for any health research area is exceedingly complex. Consideration of the prob- lem usually starts with an assessment of the impact of the disease, sometimes in comparison with other health problems. The numbers of cases can be tabulated in a variety of ways mortality, morbidity, duration, severity, disability, type of complication, or sequelae. Such considerations can include not only the severest consequences but also the burden of milder cases. Some measures of burden address the age distribution of disease, including early deaths and years of life lost. Particular types of morbidity may be weighted particularly heavily- e.g., neuropsychiatric impairment. The burden of disease can also be calculated in terms of the health care costs or other impacts on the health
FUTURE RESEARCH NEEDS 245 system e.g., the percentage of beds occupied by AIDS cases. To these considerations must be added some weighting related to knowledge about the trends and spread of the disease, including the uncertainty and public anxiety over these factors. The total federal funding initially allocated for research on AIDS and HIV for FY 1986 (exclusive of the military allocation of $37 million) was about $192.8 million. This was later increased to approximately $240 million. This amount compares to research allocations of about $1,500 million in FY 1986 for cancer; $650 million for heart disease; $59 million for motor vehicle accidents; and $27 million for sickle cell disease. A research investment ratio can be calculated for these diseases, in which the present federal investment is compared to estimated deaths by each disease category during the next decade (Graham, 19861. Such analytical measures are useful, but they provide only one factor to be considered in the overall decision processes determining the distribu- tion and relative levels of research funding. Other factors include · The strength of desire to ameliorate the problems; · The perceived opportunities and likelihood that funding will result in applicable results; · The spillover effect for basic research in which findings have appli- cability to other health problems (For example, the rapid progress on HIV was largely a result of funding for basic research in cancer virology in the late 1960s and 1970s.~; · The likely quality of research that will result if funds are allocated (In this area the question is whether additional funds can be effectively used. Peer review priority scores for NIH-funded grants related to AIDS are now equal to those in other areas and are likely to improve further if active recruitment of more high-caliber individuals to the area is attempted.~; · Inducements (or the lack of them) other than federal funding to conduct research, including the market size for potential products and their patent or liability protection (e.g., for vaccines). A compelling reason for increasing AIDS funding is the nature of the uncertainty about the future course of the epidemic. It is clear that the problem will worsen considerably over the next 5 to 10 years, and without much question it will persist into the next century. The presently uncertain rate of spread of HIV infection further into the heterosexual population will determine the ultimate magnitude of the epidemic. Rather than waiting to see how bad the problem will become, all eventualities must be prepared for now by putting a very high priority on AIDS and HIV research. If the epidemic worsens dramatically, as is quite conceiv- able, it may be too late to mount the required effort. The United States has a special responsibility to assist the rest of the
246 CONFRONTING AIDS world in handling disease problems. This country has the largest biomed- ical research community in the world, has been the main source of progress in biomedical research for the world since the l950s, and is looked to as a major intellectual and financial resource for dealing with international health crises. As discussed in Chapter 7, AIDS is presently a health crisis of extraordinary magnitude in Africa and is rapidly spreading to many countries in the world. Much is expected of the United States, and U.S. efforts have the potential to be of great benefit in the world's struggle against AIDS. Taking into account the projected impact of the disease and balancing the opportunities and implications of research on this and other diseases, research related to AIDS and HIV ranks high on all of the scales normally used to assess research funding levels. However, it is a new disease, recently added to the existing agenda for health research. The rapid accumulation of knowledge regarding the magnitude and nature of the problem has meant that unforeseen needs and opportunities have constantly been arising. The committee was not able to identify any currently funded area that is not potentially useful. However, as noted earlier in this chapter, it did identify many areas of basic and applied biomedical, epidemiologic, and social science research that are urgently in need of extra emphasis. Because of the necessarily long lag time from basic research to the benefits of that research, and because of the increasing magnitude of the problem, invest- ments in research on AIDS and HIV should be made immediately. Thus, the level of funding for such research should be increased substantially. Those assessing the committee's proposed increase in research funds may question whether the base of scientific personnel and facilities in the United States is large enough to usefully absorb this extra funding. The answer is clearly yes. As discussed above, funds are needed for a variety of efforts and must be distributed to a diverse set of research questions. Furthermore, the money should not all be appropriated through the NIH. Some should go to the CDC; some should go to the ADAMHA; and some should go through the National Science Foundation to social science research. The quality of institutional resources and facilities available for the job is another important question. The U.S. biomedical research establish- ment is physically in poor shape and cannot take full advantage of modern instrumentation because of policies that have limited funds to upgrade equipment in recent years. The ability to deal with the continuing problem of HIV infection and other existing and new health problems may be drastically impaired within the next decade if major investments are not made in facilities, equipment, and training programs. It is apparent that AIDS is a multifaceted problem that will not be solved easily or quickly and that many of the areas requiring attention are fundamental, needing long-term research to yield any benefit. Thus, all
FUTURE RESEARCH NEEDS 247 groups funding research (e.g., the Public Health Service, the Department of Defense, states, foundations) need to recognize the desirability of a long-term commitment to substantial programs of research on AIDS and HIV. It is prudent to act as though the problem will escalate and to increase research funding accordingly. The committee did not attempt to build a detailed budget that could be used, for example, in the appropriations process. Rather, it strove to assess the desirable overall magnitude of such a budget. The PHS request to the U.S. Department of Health and Human Services for AIDS-related research in FY 1988 was $471 million. If appropriated, this budget request would represent about a doubling of funds from FY 1986 to FY 1988. The NSF spends just over $50 million annually on social science research, but presently a very small amount of this is on studies related to AIDS. The committee believes that there are sufficient areas of need and opportunity to double research funding again by 1990, leading to an approximately $1-billion budget in that year. The areas of clear need include the following: · High-Containment Facilities for Primate Research Upgrading pri- mate centers will require millions of dollars for construction and in- creased support. · Better Containment Facilities for Universities and Research Institutes Improved containment facilities are needed to foster wider involvement of the non-NIH research community. Certain types of AIDS research cannot be carried out under normal laboratory conditions. · Training Funds Fewer biomedical scientists are being trained under federal programs every year, which will eventually result in fewer new biomedical investigators. The AIDS research program must have a strong training component to ensure that sufficient talent exists for future programs. Also, training in specific areas, such as primate physiology, will be needed. Construction and Renovation Funds The facilities for biomedical research in the United States are in great need of renovation because of years of chronic underfunding of their maintenance. · Equipment Funding for equipment, which has been neglected by sponsors in recent years, also needs to be part of a balanced AIDS research program. · Social Science and Behavioral Research Funding for the various types of social science and behavioral research needed, especially that for mounting effective public education programs, is seriously deficient. · Vaccine and Drug Development Efforts to develop both vaccines and drugs will have to expand as new leads develop and new candidates are identified.
248 CONFRONTING AIDS · International Studies Surveillance, epidemiologic research, and educational assistance are all needed to understand conditions abroad and to ameliorate the disease burden in affected countries. · Basic Research Efforts A massive effort is needed to understand at the molecular level every aspect of HIV and a host's response to infection. This will involve characterizing the structure and function of each of the viral components and greatly expanded in viva research efforts. · Epidemiologic Studies Greatly expanded epidemiologic research will be needed to monitor the spread of the infection in various groups and to better understand its natural history. · Return of Reprogrammed Funds and Personnel The AIDS program was initiated by borrowing funds and facilities from other ongoing extramural and intramural NIH research efforts (Krause, 19861. In FY 1986 this amounted to perhaps $70 million. Also, almost all of the intramural personnel working on AIDS were reprogrammed from other duties; there has been a net reduction in NIH personnel over the last few years (Krause, 19861. To maintain the strength of research efforts on other diseases and to ensure that knowledge gained from basic research is as extensive as possible, these reprogrammed funds and people should revert back to their original purposes. The committee discussed amounts likely to be needed in specific research areas by 1990 and believes its estimate of an annual total of $1 billion to be realistic. However, it refrained from identifying specific amounts for particular research areas because others will be in a better position to assess relative needs and the agencies through which these should be channeled as the time to assign these funds approaches. Recommendations · The committee recommends that the federal appropriations for research on AIDS and HIV continue to increase toward a goal of at least $1 billion annually by 1990. These funds must be new appropriations, not a reprogramming of existing PHS funds. The PHS, through the diversity of its activities, is responsible for research and public health activities relating to a wide variety of important medical problems. The funding for these activities should not be compromised by increased expenditures on AIDS research. In addition, funds should be restored to the NIH programs that have suffered from the personnel and funding diversions to the ongoing intramural AIDS effort. · The $1 billion cited above is needed to support ongoing efforts and to provide for additional research studies on AIDS and HIV. Necessary studies include those designed to improve the understanding of the
FUTURE RESEARCH NEEDS 249 natural history of HIV infection and those directed toward therapeutic interventions. There is a great need for applied research in the pursuit of effective drugs and vaccines, but, as previously discussed, much of the effort must focus on basic studies. The contributions of both the federal research effort and the extramural scientific communities should be actively supported. The potential input provided by independent investigator-initiated proposals in the elucidation of basic research ques- tions has not yet been fully developed and should be emphasized in future funding. Necessary studies include, but are not limited to, increased epidemiologic surveillance and study, increased examination of multiple routes to vaccine or drug development, increased investigation of the nature of the virus and its effects on the immune system, increased study of the immune system itself, increased investigation of animal models, increased study of modes of human behavior modification, investigation of human sexuality, and analysis of ethical and legal options in responding to the AIDS epidemic. Many of these are expensive areas of research; $1 billion may prove to be insufficient. Ways are needed immediately to prevent the spread of HIV and to cure AIDS, and a very high priority must be given to finding them. · The level of funding for investigator-initiated studies in all areas (including non-AIDS studies) must be adequate to continue to attract the most able younger scientists to clinical, social science, and basic biomed- ical research or the quality and productivity of the scientific enterprise will suffer. · The development of new fundamental knowledge in the areas of immunology, virology, and developmental and cellular neurobiology has traditionally come primarily from investigator-initiated grants to labora- tories. These laboratories are generally staffed by graduate students, medical students, and postdoctoral fellows. The trend in PHS funding for the past several years, and acutely in 1986, is to limit or decrease the numbers of fellowships for graduate students, M.D.-Ph.D. students, and postdoctoral fellows. Because they will play an important role in the future research effort on AIDS, new investigators should receive ade- quate support for their training and education. REFERENCES Allan, J. S., J. E. Coligan, T. H. Lee, M. F. McCane, P. J. Kanki, J. E. Groopman, and M. Essex. 1985. A new HTLV-III/LAV encoded antigen detected by antibodies from AIDS patients. Science 230:810-813. Alter, H. J., J. W. Eichberg, H. Masur, W. C. Saxinger, R. Gallo, A. M. Macher, H. C. Lane, and A. S. Fauci. 1984. Transmission of HTLV-III infection from human plasma to chimpanzees: An animal model for AIDS. Science 226:549-552.
250 CONFRONTING AIDS Ammann, A. J., G. Schiffman, D. Abrams, P. Volberding, J. Ziegler, and M. Conant. 1984. B-cell immunodeficiency in acquired immune deficiency syndrome. JAMA 251: 1447-1449. Arya, S. K., C. Guo, S. F. Joseph, and F. Wong-Staal. 1985. Trans-activator gene of human T-lymphotropic virus type III (HTLV-III). Science 229:69-73. Barnes, D. M. 1986. The challenge of testing potential AIDS vaccines. Science 233: 1151. Berger, P., and T. Luckman. 1966. The Social Construction of Reality. New York: Doubleday. Berman, P. W., W. Nunes, C. Fennie, and L. A. Lasky. 1986. Expression of recombinant AIDS retrovirus envelope protein analogues in genetically engineered cell lines. P. 15 in Abstracts of the Second International Conference on AIDS, Paris, June 23-25, 1986. Bostrom, R. 1983. Persuasion. New Jersey: Prentice Hall. Broder, S., R. Yarchoan, J. M. Collins, H. C. Lane, P. P. Markham, R. W. Klecker, R. R. Redfield, H. Mitsuya, D. F. Hoth, E. Gellman, J. E. Groopman, L. Resnick, R. C. Gallo, C. E. Myers, and A. S. Fauci. 1985. Effects of suramin on HTLV-III/LAV infection presenting Kaposi's sarcoma or AIDS-related complex: Clinical pharmacology and suppression of virus replication in vivo. Lancet II:627-630. Caloff, J. 1967. A Study of Four Voluntary Treatment and Rehabilitation Programs for New York City's Narcotics Addicts. New York: Community Service Society. Chakrabarti, S., M. Robert-Guroff, F. Wong-Staal, R. C. Gallo, and B. Moss. 1986. Expression of the HTLV-III envelope gene by a recombinant vaccinia virus. Nature 320: 535-537. Chandra, P., A. Vogel, and T. Gerber. 1985. Inhibitors of retroviral DNA polymerase: Their implication in the treatment of AIDS. Cancer Res. 45(suppl.):4677s-4684s. Cheevers, W. P., and T. C. McGuire. 1985. Equine infectious anemia v~rus: Im- munopathogenesis and persistence. Rev. Infect. Dis. 7:83-88. Chermann, J. C., F. Sinoussi, and C. Jasmin. 1975. Inhibition of RNA-dependent DNA polymerase of murine oncornaviruses by 5-tungsto-2-antimoniate. Biochem. Biophys. Res. Commun. 65:1229-1236. Coffin, J. M. 1986. Genetic variation in AIDS viruses. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Cohen, A. 1964. Attitude Change: Social Influences. New York: Basic Books. Dalgleish, A. G., P. C. L. Beverley, P. R. Clapham, D. H. C;-awford, M. F. Greaves, and R. A. Weiss. 1984. The CD4 (T4) antigen is an essential component antigen of the receptor for the AIDS retrovirus. Nature 312:763-767. Daniel, M. D., N. L. Letvin, N. W. King, M. Kannagi, P. K. Sehgal, R. D. Hunt, P. J. Kanki, M. Essex, and R. C. Desrosiers. 1985. Isolation of T-cell tropic HTLV-III-like retrovirus from macaques. Science 228:1201-1204. Dayton, A. I., J. G. Sodroski, C. A. Rosen, W. C. Goh, and W. A. Haseltine. 1986. The trans-activator gene of the human T cell lymphotropic virus type III is required for replication. Cell 44:941-947. de Clercq, E. 1979. Suramin: A potent inhibitor of the reverse transcriptase of RNA tumor viruses. Cancer Lett. 8:9-22. de Clercq, E. 1979. Suramin: A potent inhibitor of the reverse transcriptase of RNA tumor viruses. Cancer Lett. 8:9-22. De Rossi, A., G. Franchini, A. Aldovini, A. DelMistro, L. Chreco-Bianchi, R. C. Gallo, and F. Wong-Staal. 1986. Differential response to the cytopathic effects of HTLV-III superinfection in T4+ (helper) and T8+ (suppressor) T-cell clones transformed by HTLV-I. Proc. Natl. Acad. Sci. USA 83:4297-4301. Desrosiers, R., and N. L. Letvin. 1986. Animal models for AIDS and their use for vaccine and drug development. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Donzelot, V. J. 1979. The Policing of Families. New York: Pantheon Books.
FUTURE RESEARCH NEEDS 251 Douglas, M. 1986. Risk Acceptability According to the Social Sciences. New York: Russell Sage. Douglas, M., and A. Wildavsky. 1982. Risk and Culture. Berkeley: University of California Press. Farquhar, J. M., N. Macoby, and D. S. Solomon. 1984. Community applications of behavioral medicine. Pp. 437-480 in Handbook of Behavioral Medicine, W. D. Gentry, ed. New York: Guildford Press. Feinberg, M. B., R. F. Jarrett, A. Aldovini, R. C. Gallo, and F. Wong-Staal. In press. HTLV-III expression and production involve complex regulation at the levels of splicing and translation of viral RNA. Cell. Feorino, P. M., H. W. Jaffe, E. Palmer, T. A. Peterman, D. P. Francis, V. S. Kaly anaraman, R. A. Weinstein, R. L. Stoneburner, W. J. Alexander, C. Raevsky, J. P. Getchell, D. Warfield, H. W. Haverkos, B. W. Kilbourne, J. K. A. Nicholson, and J. W. Curran. 1985. Transfusion-associated acquired immunodeficiency syndrome: Evidence for persistent infection in blood donors. N. Engl. J. Med. 312:1293-1296. Fischoff, B., S. Lichtenstein, P. Slovic, S. Derby, and R. Keeney. 1981. Acceptable Risk. New York: Cambridge University Press. Fisher, A. G., M. B. Feinberg, S. F. Josephs, M. E. Harper, I. M. Marselle, G. Reyes, M. A. Gonda, A. Aldovin, C. Debouk, R. C. Gallo, and F. Wong-Staal. 1986a. The trans-activator gene of HTLV-III is essential for a virus replication. Nature 320:367-371. Fisher, A. G., L. Ratner, H. Mits~ya, L. M. Marselle, M. E. Harper, S. Broder, R. C. Gallo, and F. Wong-Staal. 1986b. Infectious mutants of HTLV-III with changes in the 3' region and markedly reduced cytopathic effects. Science 233:655-659. Folks, T., D. M. Powell, M. M. Lightfoote, S. Benn, M. A. Martin, and A. S. Fauci. 1986. Induction of HTLV-III/LAV from a nonvirus-producing T-cell line: Implications for latency. Science 231 :600-602. Francis, D. P., P. M. Feorino, J. R. Broderson, H. M. McClure, J. P. Getchell, C. R. McGrath, B. Swenson, J. S. McDougal, E. L. Palmer, A. K. Harrison, F. Barre-Sinoussi, J.-C. Chermann, L. Montagnier, J. W. Curran, C. D. Cabradilla, and V. S. Kalyana- raman. 1984. Infection of chimpanzees with lymphadenopathy-associated virus. Lancet II: 1276-1277. Fultz, P. N., H. M. McClure, D. C. Anderson, R. B. Swenson, R. Anand, and A. Srinivasan. 1986a. Isolation of a T-lymphotropic retrovirus from naturally infected sooty mangabay monkeys (Cerecocebus atys). Proc. Natl. Acad. Sc~. USA 83:5286-5290. Fultz, P. N., H. M. McClure, R. B. Swenson, C. R. McGrath, A. Brodie, J. P. Getchell, F. C. Jensen, D. C. Anderson, J. R. Broderson, and D. P. Francis. 1986b. Persistent infection of chimpanzees with human T-lymphocyte retrovirus in brains and other tissues from AIDS patients. Lancet 1:55-56. Gajdusek, D. C., H. L. Amyx, C. J. Gibbs, Jr., D. M. Asher, P. Rodgers-Johnson, L. G. Epstein, P. S. Sarin, R. C. Gallo, A. Maluish, L. O. Arthur, L. Montagnier, and D. Mildvan. 1985. Infection of chimpanzees by human T-lymphocyte retrovirus in brain and other tissues from AIDS patients. Lancet 1:55-56. Gallo, R. C., S. Z. Salahuddin, M. Popovic, G. M. Shearer, M. Kaplan, B. F. Haynes, T. J. Palker, R. Redfield, J. Oleske, B. Safai, G. White, P. Foster, and P. D. Markham. 1984. Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science 224:500-503. Gartner, S., P. Markovitz, D. M. Markovitz, M. H. Kaplan, R. C. Gallo, and M. Popovic. 1986. The role of mononuclear phagocytes in HTLV-III/LAV infection. Science 233:215- 219. Gelmann, E. P., O. T. Preble, R. Steis, H. C. Lane, A. H. Rook, M. Wesley, J. Jacob, A. Fauci, H. Masur, and D. Longo. 1985. Human lymphoblastoid interferon treatment of
252 CONFRONTING AIDS Kaposi's sarcoma in the acquired immunodeficiency syndrome: Clinical response and prognostic parameters. Am. J. Med. 78:737-741. Gonda, M. A., F. Wong-Staal, and R. C. Gallo. 1985. Sequence homology and morphologic similarity of HTLV-III and visna virus, a pathogenic lentivirus. Science 227:173-177. Graham, J. D. 1986. AIDS and biomedical research: A comparative analysis. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Groopman, J. E., M. S. Gottlieb, J. Goodman, R. T. Mitsuyasu, M. A. Conant, H. Prince, J. L. Fahey, M. Derezin, W. M. Weinstein, C. Casavante, J. Rothman, S. A. Rudnick, and P. A. Volberding. 1984. Recombinant alpha-2 interferon therapy for Kaposi's sarcoma associated with the acquired immunodeficiency syndrome. Ann. Intern. Med. 100:671-676. Haase, A. T. 1986. Pathogenesis of lentivirus infection. Nature 322:130-136. Hahn, B. H., G. M. Shaw, M. E. Taylor, P. D. McNeeley, W. P. Parks, S. Modrow, R. C. Gallo, and F. Wong-Staal. 1986. Nature and extent of genetic variation in HTLV- III/LAV. P. 16 in Abstracts of the Second International Conference on AIDS, Paris, June 23-25, 1986. Hall, C. B., J. T. McBride, E. E. Walsh, D. M. Bell, C. L. Gala, S. Hildreth, L. G. Ten Eyck, and W. J. Hall. 1983. Aerosolized ribavirin treatment of infants with respiratory syncytial viral infection: A randomized double-blind study. N. Engl. J. Med. 308:1443- 1447. Hardy, W. D., P. W. Hess, E. MacEwen, E. McClelland, E. Zuckerman, M. Essex, S. M. Cotter, and O. Jarret. 1976. Biology of feline leukemia virus in the natural environment. Cancer Res. 36:466-847. Harper, M. E., L. M. Marselle, K. J. Chayt, R. C. Gallo, M. H. Kaplan, and F. Wong-Staal. 1986a. Detection of concomitant infection of HTLV-I and HTLV-III in situ hybridization. J. Cell. Biochem. Suppl. 10A:189. Harper, M. E., I. M. Marselle, R. C. Gallo, and F. Wong-Staal. 1986b. Detection of lymphocytes expressing human T-lymphotropic virus type III in lymph nodes and peripheral blood from infected individuals by in situ hybridization. Proc. Natl. Acad. Sci. USA 83:772-776. Hawking, F. 1981. Chemotherapy of filariasis. Antibiot. Chemother. 30:135-162. Heagy, W., V. E. Kelley, T. B. Strom, K. Mayer, H. M. Shapiro, R. Mandel, and R. Fineberg. 1984. Decreased expression of human class II antigens on monocytes from patients with the acquired immune deficiency syndrome: Increased expression with interferon gamma. J. Clin. Invest. 74:2089-2096. Helgstrand, E., B. Eriksson, N. G. Johansson, B. Lannero, A. Larsson, A. Misiorny, J. O. Noren, B. Sjoberg, K. Stenberg, G. Stening, S. Stridh, B. Oberg, S. Alenius, and L. Philipson. 1978. Trisodium phosphonoformate, a new antiviral compound. Science 201: 819-821. Ho, D. D., K. L. Hartshorn, T. R. Rota, C. A. Andrews, J. C. Kaplan, R. T. Schooley, and M. S. Hirsch. 1985a. Recombinant human interferon alfa-A suppresses HTLV-III replication in vitro. Lancet 1:602-604. Ho, D. D., T. R. Rota, and M. S. Hirsch. 1985b. Antibody to lymphadenopathy-associated virus in AIDS. N. Engl. J. Med. 312:649-650. Ho, D. D., T. R. Rota, R. T. Schooley, J. C. Kaplan, J. D. Allan, J. E. Groopman, L. Resnick, D. Felsenstein, C. A. Andrews, and M. S. Hirsch. 1985c. Isolation of HTLV-III from cerebrospinal fluid and neural tissues of patients with neurologic syndromes related to the acquired immunodeficiency syndrome. N. Engl. J. Med. 313:1493-1497. Hogle, J. M., M. Chow, and D. J. Filman. 1985. Three-dimensional structure of poliovirus at 2.9 A resolution. Science 229:1358-1365.
FUTURE RESEARCH NEEDS 253 Holt, C. A., K. Osorio, and F. Lilly. 1986. Friend virus-specific cytotoxic T lymphocytes recognize both gag and ens gene-encoded specificities. J. Exp. Med. 164:211-226. Hoxie, J. A., B. S. Haggarty, J. L. Rackowski, N. Pillsbury, and J. A. Levy. 1985. Persistent noncytopathic infection of normal T lymphocytes with AIDS-associated retrovirus. Science 229:1400-1402. Hu, S.-L., S. G. Kosowski, A. F. Purchio, and J. M. Dalrymple. 1986. Expression of AIDS virus envelope gene in recombinant vaccinia virus. P. 9 in Abstracts of the Second International Conference on AIDS, Paris, June 23-25, 1986. Hunsmann, G., J. Schneider, and A. Schultz. 1981. Immunoprevention of Friend virus- induced erythroleukemia by vaccination with viral envelope glycoprotein complexes. Virology 113:602-612. ICN Pharmaceuticals. 1986. Safety of Ribavirin. News Release, October 3, 1986, Costa Mesa, Calif. Jaffe, H. W., D. J. Abrams, A. J. Amman, B. J. Lewis, and J. A. Golden. 1983. Complications of co-trimoxazole in treatment of AIDS-related Pneumocystis carinii pneumonia in homosexual men. Lancet II:1109-1111. Jaffe, H. W., P. M. Feorino, W. W. Darrow, P. M. O'Malley, J. P. Getchell, D. T. Waffield, B. M. Jones, D. F. Echenberg, D. P. Francis, and J. W. Curran. 1985. Persistent infection with human T-lymphotropic virus type III/lymphadenopathy-associated virus in appar- ently healthy homosexual men. Ann. Intern. Med. 102:627-628. Jarrett, W., O. Jarrett, L. Mackey, H. Laird, C. Hood, and D. Hay. 1975. Vaccination against feline leukemia virus using a cell membrane antigen system. Int. J. Cancer 16: 134- 139. Jasmin, C., J. C. Chermann, C. Herve, A. Teze, P. Souchay, C. Boy-Soustau, N. Raybaud, F. Sinoussi, and M. Raynaud. 1974. In vivo inhibition of murine leukemia and sarcoma viruses by heteropolyanion 5-tungsto-2-antimoniate. J. Natl. Cancer Inst. 53:469-474. Jenkins, F. J., and Y. C. Chen. 1981. Effect of ribavirin on Rous sarcoma virus transfor- mation. Antimicrob. Agents Chemother. 19:364-368. Johnson~ R. T. 1985. Nononcogenic retrovirus infections as models for chronic and relaps~ng numan disease: Introduction. Rev. Infect. Dis. 7:66-67. Jones, E., J. Beniger, and C. Westoff. 1980. Pill and IUD discontinuation in the United States, 1970-1975: The influence of the media. Fam. Plann. Perspect. 12:293-300. Kan, N. C., G. Franchini, F. Wong-Staal, G. C. Dubois, W. G. Robey, J. A. Lautenberger, and T. S. Papis. 1986. Identification of HTLV-III/LAV sor gene product and detection of antibodies in human sera. Science 231:1553-1555. Kanki, P. J., J. Alroy, and M. Essex. 1985. Isolation of a T-lymphotropic retrovirus related to HTLV-III/LAV from wild-caught African green monkeys. Science 230:951-954. Kanki, P. J., F. Barin, S. M'Boup, J. S. Allan, J. L. Romet-Lemonne, R. Marlink, M. F. McLane, T.-H. Lee, B. Arbeille, F. Denis, and M. Essex. 1986. New human T-cell retrovirus related to simian T-lymphotropic virus type III (STLV-IIIAGM). Science 232: 238-243. Kennedy-Stoskopf, S., and O. Narayan. 1986. Neutralizing antibodies to visna lentivirus: Mechanism of action and possible role in virus persistence. J. Virol. 59:37-44. Klapper, J. 1960. The Effects of Mass Communication. Glencoe, Ill.: The Free Press. Klatzmann, D., and L. Montagnier. 1986. Approaches to AIDS therapy. Nature 319: 10-11. Klatzmann, D., E. Champagne, S. Chamaret, J. Gruest, D. Guetard, T. Hercend, J. C. Gluckman, and L. Montagnier. 1984. T-lymphocyte T4 molecule behaves as the receptor for human retrovirus LAV. Nature 312:767-768. Knight, V., H. W. McClung, S. Z. Wilson, B. K. Waters, J. M. Quarles, R. W. Cameron,
254 CONFRONTING AIDS S. E. Greggs, J. M. Zerwas, and R. B. Couch. 1981. Ribavirin small-particle aerosol treatment of influenza. Lancet II: 945-949. Krause, R. 1986. The NIH research effort on AIDS: To what extent has it been achieved by reprogramming funds originally allotted for other purposes? Background paper. Wash- ington, D.C.: Committee on a National Strategy for AIDS. Krown, S. E., F. X. Real, S. Cunningham-Rundles, P. L. Myskowski, B. Koziner, S. Fein, A. Mittelman, H. F. Oettgen, and B. Safai. 1983. Preliminary observations on the effect of recombinant leukocyte A interferon in homosexual men with Kaposi's sarcoma. N. Engl. J. Med. 308:1071-1076. Kunreuther, H. 1978. Disaster Insurance Protection: Public Policy Lessons. New York: Wiley. Lane, H. C., H. Masur, L. C. Edgar, G. Whalen, A. H. Rook, and A. S. Fauci. 1983. Abnormalities of B-cell activation and immunoregulation in patients with the acquired immunodeficiency syndrome. N. Engl. J. Med. 309:453-458. Lane, H. C., J. M. Depper, W. S. Greene, G. Whalen, T. A. Waldmann, and A. S. Fauci. 1985. Qualitative analysis of immune function in patients with the acquired immunodefi- ciency syndrome: Evidence for a selective defect in soluble antigen recognition. N. Engl. J. Med. 313:79-84. Lang, G., and L. Kurt. 1983. The Battle for Public Opinion. New York: Columbia University Press. Lee, T. H., J. E. Coligan, J. S. Allan, M. F. McLane, J. E. Groopman, and M. Essex. 1985. A new HTLV-III/LAV protein encoded by a gene found in cytopathic retroviruses. Science 236:1546-1549. Letvin, N. L., and R. D. Desrosiers. 1986. Animal models for AIDS and their use for vaccine and drug development. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Letvin, N. L., M. D. Daniel, P. K. Sehgal, R. C. Desrosiers, R. D. Hunt, L. M. Waldron, J. J. MacKey, D. K. Schmidt, L. V. Chalifoux, and N. W. King. 1985. Induction of AIDS-like disease in macaque monkeys with T-cell tropic retrovirus STLV-III. Science 230:71-73. Levine, A. M., P. S. Gill, J. Cohen, J. G. Hawkins, S. C. Formenti, S. Aguilar, P. R. Meyer, M. Krailo, J. Parker, and S. Rasheed. 1985. Suramin antiviral therapy in the acquired immunodeficiency syndrome. Clinical, immunologic, and virologic results. Ann. Intern. Med. 105:32-37. Leventhal, H., and P. Cleary. 1980. Review of the research: Theory of behavioral risk modification. Psychol. Bull. 88:370-405. Lewis, M. G., L. E. Mathes, and R. G. Olsen. 1981. Protection against feline leukemia by vaccination with a subunit vaccine. Infect. Immun. 34:808-812. Lichtenstein, E. 1982. The smoking problem. J. Consult. Clin. Psychol. 50:804-819. Lifson, J. D., M. B. Feinberg, G. R. Reyes, L. Rabin, B. Banapour, S. Chakrabarti, B. Moss, F. Wong-Staal, K. S. Steimer, and E. Engleman. In press. Inductions of CD4 dependent cell fusion by the HTLV-III/LAV envelope glycoprotein. Nature. Lifson, J. D., G. R. Reyes, B. S. Stein, and E. G. Engleman. 1986. AIDS retrovirus-induced cytopathology: Giant cell formation and involvement of CD4 antigen. Science 232:1123- 1127. Lutz, H., N. Pederson, J. Higgins, V. Hubscher, A. Troy, and G. H. Therten. 1980. Humoral immune reactivity to feline leukemia virus and associated antigens in cats naturally infected with feline leukemia virus. Cancer Res. 40:3642-3646. Matthews, T. J., A. J. Langlois, W. G. Rovey, N. T. Chang, R. C. Gallo, P. J. Fischinger, and D. P. Bolognesi. In press. Restricted neutralization of divergent HTLV-III/LAV isolates by antibodies to the major envelope glycoprotein. Proc. Natl. Acad. Sci. USA.
FUTURE RESEARCH NEEDS 255 McClosky, H., and A. Brell. 1983. Dimensions of Tolerance. New York: Russell Sage Foundation. McCormick, J. B., S. W. Mitchell, J. P. Getchell, and D. R. Hicks. 1984. Ribavirin suppresses replication of lymphadenopathy-associated virus in culture of human lympho- cytes. Lancet II:1367-1369. McCormick, J. B., I. J. King, P. A. Webb, C. L. Schribner, R. B. Craven, K. M. Johnson, L. H. Elliott, and R. Belmont-Williams. 1986. Lassa fever: Effective therapy with ribavirin. N. Engl. J. Med. 314:20-26. McGuire, T. C., D. S. Adams, G. C. Johnson, P. Klevjer-Anderson, D. D. Barbe, and J. R. Gorham. 1986. Acute arthritis in caprine arthritis encephalitis virus challenge exposure of vaccinated or persistently infected goats. Am. J. Vet. Res. 47:537-540. Meusing, M. A., D. H. Smith, C. D. Cabradilla, C. V. Benton, L. A. Lasky, and D. J. Capon. 1985. Nucleic acid structure and expression of the human AIDS/lymphadenopathy retrovirus. Nature 313 :450-457. Miller, J. M., M. J. Van der Masten, and M. J. F. Schmen. 1985. Vaccination of cattle with binary ethylenimine treated bovine leukemia virus. Am. J. Vet. Res. 44:64-67. Mitsuya, H., and S. Broder. 1986. Inhibition of the in vitro infectivity and cytopathic effect of HTLV-III/LAV by 2',3'-dideoxynucleosides. Proc. Natl. Acad. Sci. USA 83:1911-1915. Mitsuya, H., M. Popovic, R. Yarchoan, S. Matsushita, R. C. Gallo, and S. Broder. 1984. Suramin protection of T cells in vitro against infectivity and cytopathic effect of HTLV-III. Science 266:172-174. Mitsuya, H., K. J. Weinhold, P. A. Furman, M. H. St. Clair, S. N. Lehrman, R. C. Gallo, D. Bolognesi, D. W. Barry, and S. Broder. 1985. 3'-Azido-3'-deoxythymidine (BW A509U): Antiviral agent that inhibits the infectivity and cytopathic effect of human T-lymphotropic virus type III/lymphadenopathy-associated virus in vitro. Proc. Natl. Acad. Sci. USA 82:7096-7100. Montelaro, R. C., B. Parekh, A. Orrego, and C. J. Issel. 1984. Antigenic infectious anemia virus, a retrovirus. J. Biol. Chem. 259:10539-10544. Murray, H. W., R. A. Gellene, D. M. Libby, C. D. Rothermel, and B. Y. Rubin. 1985a. Activation of tissue macrophages from AIDS patients: In vitro response of alveolar macrophages to lymphokines and interferon-gamma. J. Immunol. 153:2374-2378. Murray, H. W., J. K. lIillman, B. Y. Rubin, C. D. Kelley, J. L. Jacobs, L. W. Tyler, D. M. Donelly, S. M. Carriero, J. H. Godbold, and R. B. Roberts. 1985b. Patients at risk for AIDS-related opportunistic infections: Clinical manifestations and impaired gamma interferon production. N. Engl. J. Med. 313: 1504-1509. Narayan, O. 1986. Models of AIDS in the light of other natural lentivirus infections in domestic animals. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Narayan, O., D. E. Griffin, and J. E. Clements. 1978. Virus mutation during "slow infection": Temporal development and characterization of mutants of visna virus recov- ered from sheep. J. Gen. Virol. 41:343-352. Narayan, O., D. Sheffer, D. E. Griffin, J. Clements, and J. Hess. 1984. Antibodies to caprine arthritis-encephalitis lentivirus in persistently infected goats can be overcome by immu- nization with inactivated Mycobacterium tuberculosis. J. Virol. 49:349-355. Navarro, V. 1977. Political power, the state and their implications in medicine. Rev. Rad. Polit. Econ. 9:61-80. Nelkin, D. 1973. Methadone Maintenance: A Technological Fix. New York: George Braziller. Nelkin, D., and M. Brown. 1984. Workers at Risk. Chicago: University of Chicago Press. Nelkin, D., and S. Hilgartner. 1986. Disputed dimensions of risk. Milbank Mem. Fund Q. Sept/Oct.
256 CONFRONTING AIDS Olsen, R. G., E. A. Hoover, J. P. Schaller, L. E. Martes, and L. H. Wolff. 1977. Abrogation of resistance to feline oncornavirus disease by immunization with killed feline leukemia virus. Cancer Res. 37:2082-2096. Olsen, R. G., M. Lewis, L. E. Marthes, and W. Hause. 1980. Feline leukemia vaccine: Efficacy testing in a large multicat household. Feline Pract. 10:13-16. Osterhaus, A., K. Weijer, F. Uytdehaag, O. Jarrett, B. Sundquist, and B. Morein. 1985. Induction of protective immune response in cats by vaccination with feline leukemia virus iscom. J. Immunol. 135:591-596. Pahwa, S., R. Pahwa, C. Saxinger, R. C. Gallo, and R. A. Good. 1985. Influence of HTLV-III/LAV on functions of human lymphocytes: Evidence for immunosuppressive effects and polyclonal B cell activation by banded viral preparations. Proc. Natl. Acad. Sci. USA 82:8198-8202. Patrascu, I. V., S. Coman, I. Sandu, P. Stiube, I. Munteanu, T. Coman, M. Ionescu, D. Popescu, and D. Mihailescu. 1980. Specific protection against bovine leukemia virus infection conferred on cattle by the Romanian inactivated vaccine BL-VACC-RO. Virologie 31 :95-102. Pillemer, E. A., D. A. Kooistra, O. N. Witte, and I. L. Weissman. 1986. Monoclonal antibody to the amino-terminal L sequence of murine leukemia virus glycosylated gag polyproteins demonstrates their usual orientation in the cell membrane. J. Virol. 57:413- 421. Popovic, M., M. G. Sarngadharan, E. Read, and R. C. Gallo. 1984. Detection, isolation, and continuous production of cytopathic retroviruses (HTLV-III) from patients with AIDS and pre-AIDS. Science 224:497-500. Quinnan, G. V., H. Masur, A. H. Rook, G. Armstrong, W. R. Frederick, J. Epstein, J. F. Manischewitz, A. M. Macher, L. Jackson, J. Ames, H. A. Smith, M. Parker, G. R. Pearson, J. Parrillo, C. Mitchell, and S. E. Strauss. 1984. Herpes virus infections in the acquired immune deficiency syndrome. JAMA 252:72-77. Rabson, A. B., D. F. Daugherty, S. Ven Katesan, K. E. Boulukos, S. I. Benn, T. M. Folks, P. Feorino, and M. A. Martin. 1985. Transcription of novel open reading frames of AIDS retrovirus during infection of lymphocytes. Science 229:1388-1390. Ratner, L., W. Haseltine, R. Patarca, K. J. Livak, B. Starcich, S. F. Josephs, E. R. Doran, J. Antoni Rafalski, E. A. Whitehorn, K. Baumeister, L. Ivanoff, S. R. Petteway, M. L. Pearson, J. A. Lautenberg, T. K. Papas, J. Ghrayeb, N. T. Chang, R. C. Gallo, and F. Wong-Staal. 1985. Complete nucleotide sequence of the AIDS virus, HTLV-III. Nature 313:277-284. Resnick, L., F. diMarzo-Veronese, J. Schupbach, W. W. Tourtellotte, D. D. Ho, F. Muller, P. Shapshak, M. Vogt, J. E. Groopman, P. D. Markham, and R. C. Gallo. 1985. Intra-blood-brain-barrier synthesis of HTLV-III specific IgG in patients with neurologic symptoms associated with AIDS or AIDS-related complex. N. Engl. J. Med. 313:1498- 1504. Robert-Guroff, M., M. Brown, and R. C. Gallo. 1985. HTLV-III-neutralizing antibodies in patients with AIDS and AIDS-related complex. Nature 316:72-74. Roberts, R. B., D. Scavuzzo, J. Laurence, O. Laskin, Y. Kim, and H. W. Murray. 1986. Effects of short term oral ribavirin in high risk patients for AIDS. P. 68 in Abstracts of the Second International Conference on AIDS, Paris, June 23-25, 1986. Robinson, E. J. 1963. Analyzing the impact of science reporting. Journalist Quart. 40:306- 314. Rogers, E. 1983. Diffusion of Innovation. New York: The Free Press. Roloff, M., and G. R. Miller. 1980. Persuasion: New Directions in Theory and Research. Beverly Hills: Sage Publications.
FUTURE RESEARCH NEEDS 257 Roman, P., and H. Trice. 1972. Spirits and Demons at Work. Ithaca: New York State Industrial and Labor Relations Press. Rook, A. H., H. Masur, H. C. Lane, F. Frederick, T. Kasahara, A. M. Macher, J. Y. Djeu, J. F. Manischewitz, L. Jackson, A. S. Fauci, and G. V. Quinnan, Jr. 1983. Interleukin-2 enhances the depressed natural killer and cytomegalovirus-specific cytotoxic activities of lymphocytes from patients with the acquired immune deficiency syndrome. J. Clin. Invest. 72:398-403. Rosen, C. A., J. G. Sodroski, W. C. Goh, A. I. Dayton, J. Lippe, and W. A. Haseltine. 1986. Post-transcriptional regulation accounts for the trans-activation of the human T-lymphocyte virus type III. Nature 319:555-559. Rossman, M., E. Arnold, J. W. Erikson, E. A. Frankenberger, J. P. Griffith, H. J. Hecht, J. E. Johnson, G. Kamer, M. Luo, A. G. Moser, R. R. Ruekert, B. Sherry, and G. Vriend. 1985. The structure of a human common cold virus and functional relationship to other picornaviruses. Nature 317: 145-153. Rouvroy, D., J. Bogaerts, J. B. Habyarimana, O. Nzaramba, and P. Van de Perre. 1985. Short-term results with suramin for AIDS-related conditions. Lancet 1:878-879. Rozenbaum, W., D. Dormont, B. Spire, E. Vilmer, M. Gentilini, C. Griscelli, L. Mon- tagnier, F. Barre-Sinoussi, and J. C. Chermann. 1985. Antimoniotungstate (HPA 23) treatment of three patients with AIDS and one with prodrome. Lancet 1:450-451. Salahuddin, S. Z., P. D. Markham, R. R. Redfield, M. Essex, J. E. Groopman, M. G. Sarngadharan, M. F. McLane, A. Sliski, and R. C. Gallo. 1984. HTLV-III in symptom- free seronegative persons. Lancet 1:1418-1420. Salahuddin, S. Z., P. D. Markham, M. Popovic, M. G. Sarngadharan, S. Orndorff, A. Fladagar, A. Patel, J. Gold, and R. C. Gallo. 1985. Isolation of infectious human T-cell leukemia/lymphotropic virus type III (HTLV-III) from patients with acqu~red immuno- deficiency syndrome (AIDS) or AIDS-related complex (ARC) and from healthy carriers: A study of risk groups and tissue sources. Proc. Natl. Acad. Sci. USA 82~5530-5534. Sanchez-Pescador, R., M. D. Power, P. J. Barr, K. S. Steiner, M. M. Stempien, S. L. Brown-Shimer, W. W. Gee, A. Renard, A. Randolph, J. A. Levy, D. Dina, and P. A. Luciw. 1985. Nucleotide sequence and expression of an AIDS-associated retrovirus (ARV-2). Science 227:484-492. Sandstrom, E. G., J. C. Kaplan, R. E. Byington, and M. S. Hirsch. 1985. Inhibition of human T-cell lymphotropic virus type III in vitro by phosphonoformate. Lancet I: 1480- 1482. Sarin, P. S., Y. Taguchi, D. Sun, A. Thornton, R. C. Gallo, and B. Oberg. 1985. Inhibition of HTLV-III/LAV replication by foscarnet. Biochem. Pharmacol. 34:4075-4079. Seligmann, M., L. Chess, J. L. Fahey, A. S. Fauci, P. J. Lachmann, J. L'Age-Stehr, J. Ngu, A. J. Pinching, F. S. Rosen, T. J. Spira, and J. Wybran. 1984. AIDS an immunologic reevaluation. N. Engl. J. Med. 311:1286-1292. Shannon, W. M. 1977. Selective inhibition of RNA tumor virus replication in vitro and evaluation of candidate antiviral agents in vivo. Ann. N.Y. Acad. Sci. 284:472-507. Shaw, G. M., M. E. Harper, B. H. Hahn, L. G. Epstein, D. C. Gajdusek, R. W. Price, B. A. Navia, C. K. Petito, C. J. O'Hara, E.-S. Cho, J. M. Oleske, F. Wong-Stall, and R. C. Gallo. 1985. HTLV-III infection in brains of children and adults with AIDS encephalop- athy. Science 227:177-182. Sidwell, R. W., and D. F. Smee. 1981. Bovine leukemia virus inhibition in vitro by ribavirin. Antiviral. Res. 1:47-53. Sidwell, R. W., L. B. Allen, J. H. Huffman, J. T. Witkowski, and L. N. Simon. 1975. Effect of 1-~-D-ribofuranosyl-1,2,4-triazole-3-carboxamide (ribavirin) of friend leukemia virus infections in mice (38647). Proc. Soc. Exp. Biol. Med. 148:854-858.
258 CONFRONTING AIDS Singer, E., and T. Rogers. 1986. Public opinion about AIDS. AIDS and Public Policy J. 1:1. Slovic, P., B. Fischoff, and S. Lichtenstein. 1981. Perceived risk: Psychological factors and social implications. Proc. R. Soc. London A376: 17-34. Sodroski, J. G., C. Rosen, and W. A. Haseltine. 1984. Trans-activating transcriptional activation of the long terminal repeat of human T-lymphotropic virus in infected cells. Science 225:381-385. Sodroski, J., W. C. Goh, C. Rosen, K. Campbell, and W. A. Haseltine. 1986a. Role of HTLV-III/LAV envelope in syncytium formation and cytopathicity. Nature 322:470-474. Sodroski, J. G., W. C. Goh, C. Rosen, A. Dayton, E. Terwilliger, and W. Haseltine. 1986b. A second post-transcriptional trans-activator gene required for HTLV-III replication. Nature 32:412-417. Sodroski, J. G., W. C. Goh, C. Rosen, A. Tartar, D. Portetella, A. Burny, and W. Haseltine. 1986c. Replicative and cytopathic potential of HTLV-III/LAV with sor gene deletions. Science 231:1549-1553. Sonigo, P., M. Alizon, K. Staskus, D. Klatzman, S. Cole, D. Danos, E. Retzel, P. Tiollais, A. Haase, and S. Wain-Hobson. 1985. Nucleotide sequence of visna lentivirus: Relation- ship to the AIDS virus. Cell 42:369-382. Stein, C. A., W. Saville, R. Yarchoan, S. Border, and E. P. Gelmann. 1986. Suramin and function of the adrenal cortex. Ann. Intern. Med. 104:286-287. Stewart, G. J., J. P. P. Tyler, A. L. Cunningham, J. A. Barr, G. L. Driscoll, J. Gold, and B. J. Lamont. 1985. Transmission of human T-cell lymphotropic virus type III (HTLV-III) by artificial insemination by donor. Lancet II:581-584. Stoto, M. A., D. Blumenthal, J. S. Durch, and P. H. Feldman. 1986. Federal funding for AIDS: Decision process and results. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Sundquist, B., and B. Oberg. 1979. Phosphonoformate inhibits reverse transcriptase. J. Gen. Virol. 45:273-281. Temin, H. M. 1986. Mechanisms of cell killing cytopathic effects by retrovirus. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Thormar, H. M., M. R. Barshatzky, K. Arnesen, and P. B. Kozlowski. 1983. The emergence of antigenic variants is a rare event in long term visna virus infection in vivo. J. Gen. Virol. 64:1427-1432. Toltzis, R., and A. S. Huang. 1986. Effect of ribavirin on macromolecular synthesis in vesicular stomatitis virus-infected cells. Antimicrob. Agents Chemother. 29:1010-1016. Townsend, A. R. M., F. M. Gotch, and J. Davey. 1985. Cytotoxic cells recognize fragments of the influence nucleoprotein. Cell 42:457-467. Troyer, R., and G. Markle. 1983. Cigarettes. New Brunswick: Rutgers University Press. Tuchman, G. 1978. Making News. New York: The Free Press. Wain-Hobson, S., P. Sonigo, O. Danos, S. Cole, and M. Alizon. 1985. Nucleotide sequence of the AIDS virus, LAV. Cell 40:9-17. Weiss, R., N. Teich, H. Varmus, and J. Coffin, eds. 1982. The Molecular Biology of Tumor Viruses, 2d ea.: RNA Tumor Viruses. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. Weiss, R. A., P. R. Clapham, R. Cheingsong-Popov, A. G. Dalgleish, C. A. Came, I. V. Weller, and R. S. Tedder. 1985. Neutralization of human T-lymphotropic virus type III by sera of AIDS and AIDS-risk patients. Nature 316:69-72. Weissman, I. 1986. Approaches to understanding the pathogenic mechanisms in AIDS. Background paper. Washington, D.C.: Committee on a National Strategy for AIDS. Williams, R. C., H. Masur, and T. J. Spiral 1984. Lymphocyte-reactive antibodies in acquired immune deficiency syndrome. J. Clin. Immunol. 4:118-123.
FUTURE RESEARCH NEEDS 259 Wilson, I. A., J. J. Skehel, and D. C. Wiley. 1981. Structure of the haemoagglutinin membrane glycoprotein of influenza virus at 3 A resolution. Nature 289:366-373. Wyngaarden, J. B. 1985. Federal Biomedical Policy. Address before the New York Science Policy Association at the National Academy of Sciences, Washington, D.C., November 15, 1985. Yanagihara, R., D. M. Asher, L. G. Epstein, A. V. Wolff, C. J. Gibbs, Jr., and D. C. Gajdusek. 1986. Current status of attempts to transmit acquired immunodeficiency syndrome (AIDS) to new and old world monkeys. P. 19 in Abstracts of the Second International Conference on AIDS, Paris, June 23-25, 1986. Yarchoan, R., R. W. Klecker, K. J. Weinhold, P. D. Markham, H. K. Lyerly, D. T. Durack, E. Gelmann, S. N. Lehrman, R. M. Blum, D. W. Barry, G. M. Shearer, M. Fischl, H. Mitsuya, R. C. Gallo, J. M. Collins, D. P. Bolognesi, C. E. Myers, and S. Broder. 1986. Administration of 3'azido-3'-deoxythymidine, an inhibitor of HTLV-III replication, to patients with AIDS or AIDS-related complex. Lancet 1:575-580. Yohn, D. S., R. G. Olsen, J. P. Schaller, E. A. Hoover, L. E. Mathes, L. Heding, and G. W. Davis. 1976. Experimental oncornav~rus vaccines in the cat. Cancer Res. 36:646-651. Zagury, D., J. Bernard, R. Leonard, R. Cheynier, M. Feldman, P. S. Sarin, and R. C. Gallo. 1986. Immune induction of T cell death in long term culture of HTLV-III infected T cells: A cytopathogenic model for AIDS T-cell depletion. Science 231:850-853.