Infectious Disease Risk to Public Health Posed by Xenografting
This chapter was drawn largely from the workshop Session II: Infectious Issues. Thus, the majority of the chapter summarizes workshop presentations. Where useful for background, some sections have been supplemented with additional information. The chapter, however, is not intended as an in-depth analysis and summary of the field of animal-to-human infectious diseases. The possibility that infections can be transmitted from animals to humans is of concern not only because of the threat to the health of the recipient, but also because such infections may be transmissible to others, creating a public health hazard. Further, such infections may be due to previously unrecognized organisms, making detection difficult if not impossible. If the time from infection to clinical symptoms is long, the risk of widespread transmission is greater, because during this time the new organism may silently spread from person to person, as happened with human immunodeficiency virus (HIV).
Emergence of a new public health risk appears to be a two-step process (Morse, 1995). First, a new infectious agent is introduced into a given human population from other human populations, animals, or environmental exposures. Frequently these new agents are zoonoses, defined as animal microbes that can infect humans as well as the animal species from which they come. The second step is establishment and dissemination of organisms that prove to be infective and transmissible from person to person. The first step, introduction of a potentially transmissible agent into a human, could be accomplished by transplanting an organ that was infected with the agent. It is the second step of establishment and dissemination, however, that raises public health concerns, particularly if the agent is viral since current therapies for viral illnesses are limited.
Four questions can help to analyze these potential public health concerns (Chapman, 1995; and Chapman et al., 1995). First, is there reason to believe a patient can be become infected by xenotransplantation? If so, the second question is whether such infections constitute a threat to the general public health rather than a complication limited to the individual xenotransplant recipient. Third, what options are available for the prevention and control of infectious diseases associated with the use of xenogeneic tissue in humans? Fourth, which are the most appropriate options for xenotransplantation?
Animal Infections and Xenotransplantation
Infections are a major cause of morbidity and mortality after all transplant procedures (Michaels, 1995). They were the cause of serious complications after xenotransplantation in the 1960s when chimpanzee and baboon kidneys were transplanted in two series of experiments carried out by Keith Reemtsma and Thomas Starzl. Bacterial diseases were major contributors to the deaths of five of six of the patients in each of these series. The bacterial infections most likely arose from the recipient's natural flora, but whether infectious agents from the source animal were involved was not investigated.
Consideration of the infectious risks associated with allotransplantation provides some information relevant to xenotransplantation. The major risk factor for infections in patients who receive allotransplants is immunosuppression, which hampers the patient's ability to mount a normal immune response to an infection. Latent infections in the recipient may also be reactivated during immunosuppression. On the other hand, previous infection with an agent no longer present in the recipient can be a resistance factor because the patient may have protective antibodies. The specific organ being transplanted influences the type of infection anticipated. For example, patients who receive a liver have a high incidence of abdominal infections; patients who receive a kidney are susceptible to genitourinary tract infections; and patients who receive a lung or a heart transplant are predisposed to infections of the pulmonary system.
The source of the infectious agents is likewise important. The source of infection can be endogenous flora of the recipient such as skin contaminants that can traverse the skin barrier because of incisions or central catheter lines. Likewise, the source can be endogenous latent infections that might reactivate, such as those caused by Mycobacterium tuberculosis, or members of the herpes viruses family. Nosocomial infections can also be transmitted by health care workers or family members that come in contact with the patient. Opportunistic infections with organisms that are in the environment such as aspergillus or legionella are other potential types of infection. Neither the patient's
endogenous flora nor the microbial environment will be affected by the source of the transplant (human or animal). However, some infections after allotransplantation are recognized as being from the donor organ or accompanying hematopoietic cells and are especially pertinent to xenotransplantation. Donor-associated infections are usually due to latent microbial agents, although occasionally infections occur due to unrecognized, recently acquired, and active bacterial or viral infections in the donor or source animal. Latent organisms do not cause symptoms in the donor and thus will not be recognized unless specific tests to detect them are employed. The herpesviruses cytomegarovirus (CMV) and Epstein-Barr virus (EBV) are the most frequent donor-associated infections. Other herpesviruses, such as herpes simplex virus (HSV) or varicella-zoster virus (VZV), are usually latent in nervous tissue and, therefore, are less likely to be transmitted by a transplanted organ. HIV, hepatitis B virus, and hepatitis C virus have been transmitted by organ transplantation but are less common now that donors are screened carefully to identify seropositive individuals and remove them from the donor pool. Creutzfeldt-Jacob disease (a disease causes by prions) also has been transmitted by transplantation of human tissues. Parasites can be transmitted by transplantation. For example, Toxoplasma gondii has caused infections, particularly after heart transplantation.
In light of this experience with allotransplantation, it is evident that there are several mechanisms by which a recipient could be infected by an organism present in the source animal. The organism could be one that is pathogenic for both humans and animals such as T. gondii . An animal virus could be so similar to the analogous human virus that it is able to bind to human cell receptors and cause disease. An agent that is not infectious in normal humans might cause disease in immunosuppressed organ recipients.
Immunosuppression inhibits the development of specific antibodies, thereby increasing the risk of infectious disease for the patient and also hindering some of the usual methods for detecting infectious disease. Recombination of an animal virus from the transplanted organ with a human virus present in the recipient, although perhaps unlikely to occur, is still of concern because such an event could produce a new virus with more pathogenic properties. The actual risk of human-to-human transmission of agents acquired from animal organs, tissues, or cells is not known but is clearly not zero.
It is important to consider the microbial status of the source animal and how the animal was raised. Was the animal in a specific-pathogen-free environment, a quarantined colony, or a farm? What methods were used to screen for specific pathogens?
The pathogenic potential of any zoonotic infectious agent is a function not only of the organism, but also of an evolutionary adaptation between pathogens and their natural hosts. Therefore, the pathogenic potential of an infectious agent can be modulated unpredictably when the microbe is transmitted from its natural host into another species. For example, cercopithecine herpesvirus, or B virus, has a clinical profile in its natural host, the macaque monkey, that usually does not cause disease—similar to the course of herpes simplex infection in humans. However, B virus infections of humans or other nonmacaque primates (due to a scratch or bite from a macaque) can cause neurologic disease that has a mortality rate of approximately 70 percent. Similarly, pseudorabies virus, the pig herpes simplex-like organism, is benign for adult swine but causes fatal neurologic disease when transmitted to a number of other species. Adult pigs and baboons, the animals most commonly considered as sources of cells, tissues, or organs, are almost universally positive for herpesviruses unless raised under special conditions. Additionally, most adult baboons are positive for another organism, foamy virus. Whether these organisms might have more pathogenic potential when transmitted from their natural host across species lines with xenotransplantation is not yet known.
Basis For Public Health Concern
The second question is whether xenogeneic infections constitute a threat to the general public health or are only a complication of the risk—benefit calculation for the individual xenotransplant recipient (Chapman, 1995; and Chapman et al., 1995). Historic experience with many zoonotic diseases suggests that the potential for human infection with xenogeneic pathogens has implications for the community that extend beyond the individual transplant recipient. Although not all zoonoses can be transmitted from person to person, many noteworthy outbreaks have occurred.
For example, in Germany in 1967, importation of vervet monkeys infected with the Marburg virus resulted in a primarily infected human transmitting the virus to another human, ultimately involving 31 persons and a case fatality rate of 23 percent. In Zaire in 1976, the hospital admission of a single patient infected with Ebola, another zoonotic filovirus, resulted in a large nosocomial outbreak that extended into the surrounding communities. Four successive waves of human-to-human transmission were documented, involving more than 200 persons, with a case fatality rate of 88 percent. Most of this transmission occurred in the hospital and between close family members.
In Pakistan in 1976, a shepherd with gastrointestinal bleeding due to infection with Crimean Congo hemorrhagic fever (CCHF), a zoonotic bunyavirus from domestic animals, underwent an exploratory laparotomy for
what was presumed to be peptic ulcer disease. Human-to-human transmission ensued, involving 17 persons and a case fatality rate of 24 percent.
These examples demonstrate that some zoonotic infections have the potential to extend beyond the individual and into the community. Thus, the risk of xenotransplant-associated infection is not restricted to the xenotransplant recipient alone. The potential for xenogeneic infections to be transmitted through human populations is real and poses a public health concern. Further, the risk for health care workers in close contact with the xenograft recipient is probably higher than for the community at large.
The initial drama of a zoonotic outbreak predicts the likelihood that it will receive attention but does not predict its eventual public health importance. In fact, the public health consequences of a xenogeneic infection may be most significant when the immediate pathogenicity is least evident. The filovirus and bunyavirus outbreaks were very dramatic, but their public health impact was limited because the infections had a short incubation period and were easily identified, enabling prompt initiation of public health measures to control them. Other infections with very long incubation periods are less easily controlled. For example, initial infections in humans with HIV-1 during the 1970s or earlier resulted in more than a decade of insidious transmission before AIDS was even suspected as a public health problem for the first time in the 1980s. Arguments exist to suggest that the HIV epidemic in humans resulted from a simian retrovirus introduced across species lines into human hosts, where it adapted and was then transmitted (Allan, 1995b). Several species of African monkeys carry an HIV-like virus called SIV (simian immunodeficiency virus) that is part of the animal's normal microbial flora and apparently nonpathogenic in its normal host. Many genetically distinct SIVs have been isolated and named according to the different monkey species in which they are found. Humans have two distinct HIVs, HIV-1 and HIV-2. It is hypothesized, based on molecular biological evidence, that HIV-1 was derived from SIV found in chimpanzees. One interpretation is that only two cross-species transmissions into humans created an epidemic that now has infected 18 million to 20 million people. In some areas of Africa, up to 30 percent of sexually active persons are infected. Stronger molecular evidence supports the idea that HIV-2 was derived from SIV in sooty mangabey monkeys. In addition, the geographic distribution of the HIV-2 epidemic in West Africa among humans parallels the natural habitat of the sooty mangabey.
SIV has been proven to be transmissible to humans. An active and ongoing SIV infection in a human who was working with nonhuman primates has been confirmed by evidence of seroconversion and persistent seropositivity with increasing antibody titers, identification of the seroreactivity of new viral gene products over time, and isolation of the virus from the infected person (Khabbaz et al., 1994). This human SIV infection is not an isolated event. An anonymous serological survey of primate workers demonstrated antibodies to
SIV in 3 out of 472 tested. Not only is there nonhuman primate to human transmission of SIV, but when captive nonhuman primates are housed together, horizontal transmission of retroviruses across species lines occurs on occasion.
The potential for the introduction of a new retrovirus into human hosts via implanted xenogeneic tissue is of public health concern due to the long period of clinical latency associated with all known human retroviral infections. This long latency period provides the opportunity for silent person-to-person transmission to occur before pathogenicity is evident. The pathogenic potential of exogenous retroviruses such as SIV and the baboon simian T-lymphotropic viruses (STLVs) are of concern because they are genetically similar to human exogenous retroviruses HIV-2 and HTLV (human T-cell lukemia virus). The genetic relatedness of humans to nonhuman primates may make transmission more likely with xenotransplants from primates than from more disparate species such as pigs. Some observers consider the use of baboons as a source of organs to be especially dangerous for this reason (Allan, 1995a,b). However, retroviruses are not limited to primates and have been found in many other animals including horses, minks, and cats, with an unknown potential for causing disease in humans.
All eukaryotic species examined, including baboons and pigs) have been found to harbor endogenous retroviral-like DNA (deoxyribonucleic acid) sequences that pose uncertain risks. Endogenous retroviruses are retroviruses that are fixed in the germ line and transmitted as part of the genetic inheritance of the offspring. In the host species, the retroviruses commonly are clinically benign and defective, but some are known to be xenotropic: the virus does not replicate in the host species but is capable of infecting related species. Endogenous retrovirus proviral DNA can be detected in the tissue of all animals examined thus far, including humans, nonhuman primates, and pigs. Baboon retrovirus can be isolated from baboons by cocultivation with human cells. Although no true recombinant of animal endogenous retrovirus with other animal viruses has been described, phenotypic mixing between different viruses can occur under appropriate experimental conditions, which suggests that this and other endogenous retroviruses may have pathogenic potential under conditions associated with xenotransplantation.
Safety generally has been presumed to be increased when swine tissue is used, rather than tissue from nonhuman primates. Swine retroviruses have, nevertheless, been identified, although they are still incompletely characterized. The potential for retroviruses that are latent in porcine tissue to infect an immunosuppressed human host, to rescue replication-defective viruses, or to recombine with latent viruses to create a hybrid is unknown. Further, swine harbor other infectious agents, some of which are known human pathogens. Therefore, the use of swine for xenotransplantation is not without infectious risk.
There are concerns that xenogeneic viruses may recombine or reassort with viruses latent in human tissues and result in variants that possess either a broader host range or an increased pathogenic potential. For example, the periodic emergence of new pandemic influenza strains is thought to occur by a process of reassortment between animal and human influenza viruses. In addition, the oncogenic potential of animal DNA viruses introduced into immunosuppressed humans remains undefined, although cross-species infections do exist. Indeed, one needs to ask whether a pig producer who has received a pig organ xenotransplant could be a source of a novel virus if his transplant were to become infected with a pig virus at the same time that the rest of his body was infected with a similar human virus. If viral recombination occurred, he might (or theoretically could) inadvertently pass this new variant to his own pigs. Thus, a new infection could arise in pigs, causing an emerging swine health problem, in addition to the potential human health problem. Finally, few data exist on the presence in potential animal organ sources of prion-associated disease, similar to Creutzfeldt-Jacob disease or bovine spongiform encephalopathy (Allan, 1995a,b). Prion disease, not yet described in pigs tissue, is of concern, for example, in applications using pig neuronal tissue in parkinsonian patients.
Methods Of Risk Evaluation
It is clear that the potential for infectious hazards exists with xenotransplantation. Controlling the risks to human health posed by various hazards is a four-step process involving research, risk assessment, risk management, and risk communication. First, evidence must be collected on the effects of exposure to a particular microbe by an individual and a population. Second, a risk assessment must be performed that includes three related activities:
- Hazard identification determines if the microbe actually causes an adverse effect.
- The relationship between exposure and an adverse effect must be determined (i.e., the threshold dose for infectivity).
- Exposure assessment determines transmissibility of the infectious agent and the number of people likely to be exposed.
A complete risk assessment provides an estimate of the incidence of an adverse effect in a given population.
The third step is risk management, which is the development of guidelines, rules, and regulations. This step should include evaluation of the effects of various approaches in terms of their impacts on public health and on economic, social, and political factors, including a cost—benefit analysis.
Further, risk management may have to proceed in the absence of a complete and quantitative risk assessment. Finally, there should be risk communication, which involves communicating in clear, ordinary language to the affected population and to political leaders, media, and other spokespeople, the results of the risk assessment detailed above. Such communication involves relating relative risk and likelihood to common experience as exemplified by the statement ''It is safer to fly than it is to drive a car the same distance."
Regarding the infectious disease risk to public health posed by xenotransplantation, data are missing on several of the links required to perform an adequate risk assessment. The information summarized in the first portion of this chapter permits identification of some of the known potential hazards and has shown that transmission of microorganisms occurs with xenotransplantation from animals to humans. However, hazard identification is incomplete because there are almost certainly viruses present in the animal that are potential human pathogens and that have not yet been identified. Further, some viruses have been detected, such as those in the spumavirus family (foamy viruses), that constitute an uncertain hazard because to date no disease has been linked to the presence of foamy virus in an individual. Neither the number of organisms (threshold) nor the mix of various organisms required to produce a disease in the patient is known. Exposure assessment involves consideration of the presence or absence of various organisms in the animal that was the source of the cells, tissue, or organs; the route of transmission of the microbe; and the types of contacts between the patient and other people. In many cases, it is difficult—if not impossible—to quantitate all of these factors. Thus, quantitative risk characterization is not possible given the current state of knowledge, and risk management will have to proceed in the absence of complete characterization. Risk communication is also difficult because of the lack of data and certainty. Thus, risk management depends on professional judgment, and the public should be informed that the management plan represents a best guess for minimizing risks. Participation of the public in setting guidelines will be important.
What options are available for risk management of xenotransplant-associated infectious public health risk? One option is to eliminate all risk by avoiding all use of xenogeneic tissue in humans. This option, however, would mean the death of many patients each year who are eligible for transplantation, but for whom there are no human donors available and for whom xenotransplants might be effective. Another option would be to dismiss all concerns with risk in the name of medical progress—an option that is unacceptable in light of the considerable data indicating the existence of a public health risk. A different option, which the committee found preferable, is to seek a balance between concerns for public health risks and concerns for desperately ill patients who may be aided by progress in xenotransplantation. This approach requires performing a risk characterization by estimating, to the extent
possible, what risks to the public health may exist and then utilizing the risk management tools available to decrease or contain those risks.
The first available tool is pretransplant screening of animal sources of the cells, tissues, or organs for known zoonotic pathogens. The risk of infecting the recipient with previously recognized zoonotic pathogens can be controlled by adequate survey of the animal source, by screening and quarantine the individual animal or developing well-characterized source colonies where feasible, and by attending to the circumstances and methodology of tissue retrieval. Screening should also be performed on the recipient pretransplant (baseline) and after transplant. Posttransplant prospective surveillance of the patient for the occurrence of symptoms suggesting infection and for the presence of microbes in cultured body fluids should be conducted. Both health care workers and family contacts should be followed to detect the occurrence of disease. Finally, tissue from both the animal and the recipient should be archived for future use.
The animal organ source should be screened for known zoonotic organisms that are infectious for humans, such as Brucella and Erysipelothrix in swine, or organisms that can be present in swine or baboon, such as T. gondii, encephalomyocarditis virus, lymphocytic choriomeningitis virus, and Mycobacterium species. Also, screening should be conducted for types of organisms that could be transmissible or could recombine in humans, such as herpesviruses or retroviruses.
Laboratories should use methods of virus testing that are species specific (e.g., baboon specific for baboons) when available and should consider the use of broad range polymerase chain reaction as discussed in the next section on detection methods. It is undesirable to rely on tests that were devised for human viruses because these methods may be insensitive and may miss important viruses in the animal that is to be the source of transplant material. For example, in screening baboons for viruses that might be of concern after xenotransplantation, reasonable concordance was found between two laboratories using antibody tests for cytomegalovirus as well as for simian agent 8, a baboon herpesvirus, but significant discordance was found when looking for evidence of Epstein-Barr-like virus (Herpes papio). These two laboratories relied on detecting baboon H. papio by cross-reacting serology with human reagents. One laboratory that had very few positive tests utilized an enzyme-linked immunosorbent assay (ELISA) test directed against human EBV nuclear antigen (EBNA). However, recent studies (Michaels and Simmons, 1994) have shown that baboon H. papio EBNA 2 is quite different from human EBNA, which may explain the lack of cross-reactivity. The other laboratory relied on an immunofluorescence antibody test directed against the EBV viral capsid antigen, which is a much more conserved portion of this virus. This experience highlights the need to develop tests that are specific for organisms in the
animal rather than relying simply on cross-reactivity. Again, it is important to note that nearly all adult male baboons, if not raised in a quarantined environment, are positive for three herpesviruses (simian agent 8, cytomegalovirus, and H. papio) and for foamy virus.
When nonhuman primates such as baboons are to be sources of the transplanted material, they should be screened for other organisms that, although unlikely to be found in animals raised in captivity in the United States, would preclude their use as a source of organs. Examples of such organisms include Ebola, Marburg, and Reston, as well as viruses that cause simian hemorrhagic fever and monkey pox. There should also be screening for the retrovirus family of including lentiviruses such as SIV, oncoviruses such as STLV, and simian retrovirus (SRV), as well as spumaviruses (foamy viruses). Many virologists and infectious disease specialists agree that the presence of lentiviruses or oncoviruses should preclude the use of an animal as a source of organs. There are also some who would exclude primates that carry spumaviruses, although these viruses have not been associated with disease even in infected humans.
It is important to look for organisms that are unknown by performing a broad array of viral cultures on a number of different cell lines. A comprehensive strategy for culturing viruses should be developed. This strategy could include establishing cell lines from the animal organ, induction of latent viruses in such a cell culture with inducing agents, and coculturing with a variety of sensitive indicator cells. Any viruses that are found should be preserved so that they can be used at a later date for surveillance purposes when following the patients prospectively. Bacterial stool cultures should also be performed to identify pathogens that have the potential for systemic infection; for example, Salmonella species, Yersinia species, and other potential pathogens that might have been in the gastrointestinal tract. Blood cultures of the animal should be obtained well in advance of the transplant. This is difficult to do in human donor transplantation but can be accomplished readily in xenotransplantation because the surgical procedure can be planned in advance. Appropriately stained smears of the animal's blood could also be examined for the presence of parasites and intracytoplasmic inclusion in addition to culturing the blood. At the time the organ is harvested, a full necropsy should be performed and, ideally, a number of tissues should be sampled and cultured. Blood and tissue samples should also be archived for surveillance purposes.
Potential recipients should be screened serologically for prior exposure to infectious agents, and serum samples from the patients should be stored to use as pretransplant infectious controls. Recipient surveillance protocols should involve serial serologic screening and viral cultures. Intensive culturing should
be employed for fever evaluations. Finally, all biopsy and autopsy specimens should be archived.
Public health guidelines exist that are intended to minimize the risk of transmission of known pathogens by human-to-human transplantation, and similar guidelines are under development that address xenotransplantation. In addition, when human infections with known zoonotic pathogens occur, standard diagnostic testing procedures and disease descriptions may be helpful. Likewise, therapeutic options are available for many of these pathogens. The intense immunosuppression of patients undergoing xenotransplantation may complicate the task of posttransplant monitoring, and of clinical disease recognition, by decreasing the dependability of diagnostic antibody testing and perhaps by altering disease presentation or response to therapeutic intervention. However, immunosuppression will not alter either the culture techniques or molecular diagnostics needed to identify the pathogen or the control interventions available to reduce person-to-person transmission once the infection is recognized.
One method for decreasing the infectious risk posed by xenotransplantation is to use specific-pathogen-free (SPF) animals as a source of transplant material. However, there are several difficulties with this approach. It is critical that SPF be defined for each animal: that is, what pathogen is the animal certified to be free of and what methods were used for the certification? It is not possible to have completely pathogen-free animals, even those derived by cesarean section, because some potentially infectious agents are passed in the genome and others may be passed transplacentally. Gnotobiotic animals (raised in germ-free conditions) are as free of pathogens as possible, but still would have endogenous viruses and might have other transplacentally transmitted organisms. In addition, gnotobiotic animals are very expensive to maintain and, in the case of swine, do not thrive. Raising primates in gnotobiotic environments has not been attempted. Likewise, it would be very expensive and time-consuming to produce an SPF colony of baboons because these animals have a relatively long generation time (they reach sexual maturity in five to seven years). In addition, cesarean-derived animals must be reared in isolation, which is labor intensive, and such a colony would require new, isolated housing facilities separate from other baboons. Animal welfare issues raised by rearing animals in isolation would require considerable attention, and mature animals would have to be trained in the care of newborns for the successful creation of an SPF colony. However, SPF baboon colonies could be achieved. It has been estimated that it would take 7 to 10 years and cost approximately $8 million to $10 million to produce a baboon colony large enough to provide 100 SPF baboons per year. If this approach is taken, animals should be certified to be retrovirus free (except for endogenous retroviruses) and herpesvirus free.
Production of swine that could be used for transplantation would be easier because they become sexually mature in six months and have a four-month gestation time, producing a litter of 3 to 13 offspring, so that after sexual maturity, two litters per year would be possible (Swindle, 1995). These animals grow rapidly, reaching a size suitable for adult transplantation in three to six months. SPF is a term that is currently defined in swine husbandry to denote freedom from certain specified pathogens. Under the current system, herds of SPF swine are raised on farms and are inspected every 90 days to ensure adherence to standards. Vaccination of swine is not allowed if the herd is to receive the SPF designation. However, such swine are a good place to start to develop animals suitable for transplantation purposes. It would probably not be necessary to derive swine by cesarean to establish a pathogen-free herd, because vaccines exist for many swine diseases and there are no rules prohibiting vaccination while developing such a herd for xenotransplantation. A pathogen-free herd would most likely be housed under conditions approaching laboratory animal housing instead of usual farm conditions, although the location of the facility and personnel access procedures would be important concerns as well as the SPF surveillance program.
Six colonies of SPF rhesus monkeys have been developed (Keeling, 1995). These animals are free of SIV, SRV 1–5, STLV, and cercopithecine herpesvirus 1 (B virus). This definition does not include several viruses known to infect nonhuman primates such as spumavirus, herpesviruses other than B virus, adenovirus, reovirus, and spongiform viruses, or other organisms such as T. gondii. These animals are in general too small to be useful in human transplantation and may be too phylogenetically distant from an immunological point of view. Production of these animals, with a relatively limited definition of SPF and shorter times to sexual maturity than baboons, takes five years and is expensive—each 2-year-old SPF animal costs about $3,000 (Keeling, 1995). In 1995, non-SPF baboons cost approximately $1,500.
In summary, SPF baboons would not be entirely free of infectious agents, would be expensive to produce and maintain, would require considerable time to develop, and would be an economically stressed approach to decreasing the infectious disease potential of baboons at present. If further research shows that baboons are a good source of cells, tissues, or organs for xenotransplantation, it may be worth reconsidering the production of pathogen-free baboons. Producing pathogen-free swine (the term xenografic-pathogen-free or XPF has been proposed) is feasible (although they too will not be entirely free of infectious agents) and has been done in several areas of the country. The biggest challenge with swine will be to develop immunological strategies to prevent rejection. Transgenic swine that may produce tissues and organs that are better tolerated by the human are being developed (see Chapter 2).
It is the potential for human infections with xenogeneic pathogens not previously described or not previously recognized to pose hazards to human hosts that is of greatest public health concern. Public health surveillance, the primary prevention tool applicable to all emerging infections including risk from xenotransplantation, is useful here. Such surveillance involves ongoing monitoring of a population (recipients and contacts) for events of public health concern, tracking the trends in the rate of occurrence of those events, investigating the causes of any observed increase in trends, and instituting public health interventions combined with continuous monitoring to document further changes in the rate of occurrence of those events after the intervention.
According to a 1992 Institute of Medicine report,
The key to recognizing new or emerging infectious diseases, and to tracking the prevalence of more established ones, is surveillance. A well-designed, well-implemented surveillance program can provide the means to detect unusual clusters of disease, document the geographic and demographic spread of an outbreak, and estimate the magnitude of the problem. It can also help to describe the natural history of a disease, identify factors responsible for emergence, facilitate laboratory and epidemiological research and assess the success of specific intervention efforts. (p. 2)
Most public health surveillance systems monitor for discrete and definable adverse health events. However, public health surveillance for infections with new or newly recognized pathogens, which may be associated with xenotransplantation in humans, requires monitoring for the unknown. Two questions arise: (1) How can surveillance be conducted for what is not known or recognized? (2) How can the significance of the outcome of such surveillance be assessed?
Xenotransplant recipients will all have underlying illnesses that set them apart from the normal population prior to transplantation. In this setting, surveillance must monitor for adverse health events that are unexpected, unexplained, possibly infectious, and occurring at a higher than normal rate, which is difficult to determine for a distinctly abnormal population. What surveillance approaches are most appropriate to use in xenotransplantation? Surveillance of populations of xenotransplant recipients for clustering of adverse health outcomes and monitoring of individual recipients for unexplained illnesses are probably the most useful public health tools available. This is particularly true in the case of xenotransplantation, because the patients will be identified and surveillance can be focused on these known individuals and their contacts.
There are no standard diagnostic tests available to detect infections with previously unrecognized pathogens. In addition, antibody monitoring is hampered by the immunologic suppression that is required in transplantation. Standard culture techniques may not identify new xenogeneic pathogens. Identification of such pathogens will require the use of a variety of methods that are new or being developed (see section on detection methods below). As an example of the difficulties associated with culturing techniques, in 1976, when a group of attendees at the American Legion convention died in Philadelphia from Legionnaire's disease, it took more than five months of intensive effort before Legionella was cultured for the first time.
Surveillance should also focus on those most likely to become infected by an organism from the patient, such as hospital staff or close personal contacts of the patient. The types of organisms that may be transmitted by transplanted tissue are likely to be transmissible through blood or through secretions. Hence, universal precautions should be strictly enforced in all patient contacts. Archiving serum taken from hospital staff and from other people who come in contact with patients posttransplant should be required.
The concern for xenozoonoses leads to the need to develop carefully validated methods that are both sensitive enough to detect known infectious agents and specific enough to distinguish whether an organism is of human or animal origin (Persing, 1995). Sensitive tests that may be of low specificity can be used to screen for a range of etiologic agents. Traditional methods include electron microscopy to identify structures similar to known pathogens, search for cytopathology in cell cultures inoculated with material suspected of carrying a microbe, and injection of material into animals. Many nucleic acid-based methods developed over the past decade can be quite sensitive and useful in detecting new infectious agents and monitoring infection with these agents. These methods include shotgun cloning, polymerase chain reaction (PCR), representational difference analysis (RDA), sequence-independent single primer amplification (SISPA), and others that are being developed in this rapidly changing field. Some of the more conventional techniques, such as culture and animal model to create a registry recovery techniques, and the technological framework for some of the nucleic acid techniques are discussed below.
Improvements in culture techniques have resulted in the recovery of many organisms not detectable with standard methods, including Legionella pneumophila, Bartonella henselae, Borrelia species, Mycobacterium genevensii, and other pathogens. These improvements are important advances because,
even though nucleic acid techniques are very useful for identifying and characterizing organisms, only by recovery of an organism in culture can it be characterized both immunologically and biochemically. Microorganisms have been recovered by injecting material from animals into immunodeficient rodents, such as SCID (severe combined immunodeficiency) mice infused with portions of the human immune system (SCID-hu mice). Such techniques have permitted recovery of human organisms not capable of growing in normal rodents, such as Plasmodium falciparum, Epstein-Barr virus, and Schistosoma mansoni.
A nucleic acid technique called representational difference analysis RDA, based on subtractive hybridization, has been developed.1 The RDA technique has been used to identify several organisms including human herpesvirus 8, which is associated with Kaposi's sarcoma, and a new hepatitis-associated flavivirus. This technique can be used to sort out nucleic acids in a complex mixture and to identify unique sequences that are associated with pathogenic organism. It is a very powerful technique likely to have broad application in identifying not only bacteria, fungi, and protozoa, but also endogenous retroviruses and other viral or novel pathogens.
Another nucleic acid technique is broad-range PCR, which is based on amplifying sequences that are common to a wide variety of species of microorganisms within a phylogenetic family. This technique has been used to identify a broad range of species within a given family: Rickettsia species, Bartonella species, Babesia, Mycobacterium genevensii, Mycobacterium X, lentiviruses, various human papilloma viruses, and hepatitis viruses. Broad-range PCR depends on the development of PCR primers that can identify
highly conserved ribosomal sequences common to many or all members of a viral family. Species-specific regions in the ribosomal sequences are then identified by using species-specific primers. This method, if automated, might become a routine test in clinical laboratories and permit the identification of a wide range of microorganisms, including uncultured microorganisms. Broad-range PCR could even lead to a new scheme for characterization of microorganisms in which an uncultured organism is identified and phylogenetic analysis is performed. In this scheme, the closest relative that can be grown readily in culture or in an animal is selected on the basis of phylogenetic testing. Serologic reactivity to this available organism is then determined. Such a procedure could permit epidemiologic and seroprevalence surveys for a new organism in baboons or swine that are to be the source of tissue for xenotransplants. Another by-product of this approach is that useful scientific information on infectious agents and the diversity of their natural relatives is likely to result.
One major drawback of PCR-based methods for diagnostic and discovery purposes is contamination. The sensitivity of the method will permit the detection of exquisitely small amounts of contaminating material present in the original sample along with the desired sequence. This key problem can be overcome through rigorous technique, appropriate use of controls, and corroborative methods.
The ideal virology and microbiology laboratory supporting a program of xenotransplantation should combine standard approaches to diagnosis using specific, validated tests and state-of-the-art research methods, which would be sensitive and designed to detect potential pathogens. Such a laboratory should be capable of screening for a long list of known animal and human microorganisms and would be useful in examining posttransplant patients with fever or other symptoms or signs that suggest infection or in evaluating clusters of adverse outcomes. Some workshop participants felt that a central national laboratory should be established to develop and validate new tests and methods. The other option would be to develop collaborative arrangements between laboratories at various institutions.
Need For A Registry
In order to follow patients and their contacts carefully for potential development of infections that present public health hazards, it will be critical to create a registry (Chapman, 1995). A xenotransplantation national registry would collect data on xenotransplant recipients that permitted documentation of significant commonalities among them. Such a registry ideally should be able to capture patients' records automatically and electronically, while safeguarding patient identity. Key clinical and laboratory data about every
xenotransplant recipient in the United States would be updated periodically with responses to a standard set of health screening questions and laboratory assays. Because it is necessary to maintain such a registry to protect the public health, prospective xenotransplant recipients would be required to give informed consent to surveillance, most likely for the rest of their lives. Some types of safeguards must be considered for this mandatory lifelong surveillance, however, because it implies the need for enforcement under some circumstances and presents the possibility of limitations to individual freedom and privacy.
Such a national or international registry would allow prospective monitoring of individual xenotransplant recipients and of the recipient population as a whole, as well as retrospective tracking of epidemiologically linked recipients for investigation of adverse events, if they should occur. As another benefit, the registry would allow research on other aspects of xenotransplantation, such as the effectiveness of various immunosuppressive regimes. Archiving of biological specimens from both animals and patients prior to and following the transplants would allow retrospective laboratory investigation of new agents if unexplained illnesses arise.
Investigations of individual instances of unusual illnesses can be revealing, although usually a cluster of cases must occur before a problem can be recognized, a new etiologic agent identified, or associated risk factors delineated. For example, although scattered individual cases had occurred throughout the country for more than a decade, the AIDS epidemic was not even suspected until the spring of 1981 when similar sexual activity histories were noted among seven young men with Pneumocystis carinii pneumonia (PCP) in the Los Angeles area. Other examples of cluster investigations that have resulted in public health breakthroughs include John Snow's historic research on cholera in London in the 1850s, which led to the identification of a contaminated water pump as a source of the urban outbreak. Study of a group of patients with pneumonia at the Bellevue Stratford Hotel in Philadelphia led to the identification of the Legionella pneumophila organism in 1976. Investigation of the 1993 cluster of unexplained respiratory deaths in the Four Corners region of the United States identified hantavirus pulmonary syndrome. The cost of each of these public health breakthroughs should not be underestimated. When the Minnesota State health department reviewed the results of more than 500 cluster evaluations, 6 of which involved full-scale investigations, only one had resulted in documentation of an outcome of public health significance.
Identifying the causative infectious agent is only one step toward the control of its transmission. Public health threats have been minimized through identification of, and education regarding, risk associations, for example, the association of Reye's syndrome in children with aspirin consumption. Often
public health control measures can only decrease, not eliminate, risk. In the relatively controlled xenotransplant setting where exposure is known to have occurred and where there is prospective monitoring, adequate surveillance may be able to detect new xenogeneic infections in recipients prior to significant spread into the general population. Such early detection would allow suspension of the specific procedure, source animal species, or factors associated with risk, pending the development of adequate methods for prevention and control, thereby reducing the risk of additional infections. However, as noted previously, infectious agents such as HIV, which produce clinical disease only after long latency periods, may spread widely in the population before they are detected.
To answer the four questions posed at the beginning of this chapter:
- Is there a basis for concern regarding xenogeneic infection for individual human recipients? Yes.
- Do these concerns constitute a threat to the general public health, rather than being only a complication in the risk–benefit calculation for individual xenogeneic tissue recipients? Although there is considerable debate about the degree of risk, most infectious disease experts agree that some level of risk to the general public health exists.
- What options are available for the prevention and control of infectious public health risks associated with xenografting in humans? Preplanned transplantation screening of the animal that is the source of the graft for known zoonotic pathogens and posttransplantation surveillance of the recipient for adverse health events possibly associated with xenogeneic infections are the best tools available.
- What approaches are most appropriate for xenotransplantation? A national xenotransplant registry would permit continuing surveillance of recipients for the occurrence of unusual illnesses. Archiving of appropriate biologic specimens from both the animal source and the recipient would be the key to retrospective investigation of the occurrence of such events.