5
Predictions on Vaccine Development
THE NEED FOR PREDICTIONS
Predictions about specific vaccines and the processes used to develop them are an integral part of the selection scheme outlined in Chapter 3. These predictions are required to calculate the health benefits expected from each new vaccine and the associated costs. The characteristics of a vaccine (e.g., live attenuated virus versus subunit) may affect its efficacy, and the complexity of the development process determines the costs associated with producing health benefits and the time at which they would be achieved.
This chapter describes the types of predictions in the analysis. Predictions were developed separately for each vaccine/disease combination, based on the available literature and various other sources. The final predictions were made after extensive discussions within the committee and consultations with many individuals in academic institutions, government, and industry.
SELECTION OF CANDIDATES
The committee defined candidates for accelerated development as those vaccines for which development was foreseeable within the next decade. The criterion for inclusion was whether a reasonable consensus could be identified on the nature of potential vaccine components (protective antigens). (The selection process is described further in Appendix A.)
The diseases and vaccine candidates chosen for the ranking process are shown in Tables 5.1 and 5.2 and described in detail in Appendixes D-1 through D-19. Some marginal candidates were excluded (because of
Much of the material in this chapter is from New Vaccine Development: Establishing Priorities, Volume I (Institute of Medicine, 1985). However, some sections, like the one on predictions of vaccine efficacy, have been changed to reflect the special requirements of health care delivery in developing countries.
TABLE 5.1 Predictions Table—Primary
Pathogen (Target Population) |
Type of Vaccine |
Cost of Development ($ millions) |
Probability of Successful Development |
Dengue virus (Infants and children in endemic areas; travelers to endemic areas) |
Attenuated live vector virus containing gene for broadly cross-reacting protective antigen |
25 |
0.75 |
Escherichia coli (enterotoxigenic) (Infants<6 months) |
A combination of purified colonization factor antigens and possibly other antigens |
25 |
0.50 |
Genetically engineered attenuated strains |
25–50 |
0.70 |
|
Hemophilus influenzae type b (Infants) |
Conjugated polysaccharide |
15 |
0.90 |
Hepatitis A virus (Susceptibles of all ages; routine for preschool children) |
Attenuated live virus |
15 |
0.95 |
Polypeptide recombinant vaccine produced in yeast |
25 |
0.95 |
|
Hepatitis B virus (Areas with high perinatal infection: all infants at birth (if possible). Other areas: all infants, simultaneous with other vaccinations, at earliest possible age) |
Polypeptide produced by recombinant DNA technology |
5 |
0.99 |
Japanese encephalitis virus (Children in epidemic and endemic areas; foreign visitors to epidemic regions) |
Inactivated virus produced in cell culture |
50 |
0.50 |
Mycobacterium leprae (Immunoprophylactic: all children in endemic areas. Immunotherapeutic: all recently infected individuals) |
Armadillo-derived M. leprae |
25 |
0.50 |
Neisseria meningitidis (Infants, 3 to 6 months) |
Conjugated capsular polysaccharides. Groups A,C,Y, and W135 |
30 |
0.50 (dependent upon success in developing conjugation procedures with other vaccines) |
Parainfluenza viruses (Infants) |
Trivalent, subunit vaccine (which must contain fusion proteins) |
25 |
0.80 |
Plasmodium spp. (All infants at risk, military personnel, travelers) |
Plasmodium falciparum, synthetic or recombinant sporozoite antigen preparation |
25 |
0.50 |
Multivalent synthetic or recombinant sporozoite antigen preparation (P. falciparum, P. vivax, P. ovale, P. malariae) |
35 |
0.50 |
Time to Licensure (years) |
Time to Adoption (years) |
Efficacy |
Number of doses |
Cost per dose (dollars) |
10 |
2 |
0.85 |
1 |
12 |
10 |
2 |
0.75 (against strains in vaccine) |
Approx. 3 |
5–10 |
10 |
2 |
0.80 (against strains in vaccine) |
1–2 |
1–2 |
3 |
5 |
0.90 |
2 (with DTP) |
5–10 |
4 |
5 (as part of combination) |
0.90 |
1 |
15 |
5 |
5 |
0.90 |
3 |
20 |
1 |
2 (dependent on price) |
0.90 |
3 |
30 |
6–8 |
2–3 |
0.80 |
2 (boosters required with all current; none gives life-long immunity) |
10–20 |
10 |
5 |
0.75 |
1 |
25 |
4–6 |
5 (in moderate to high risk areas) |
0.80 |
2 |
10 |
5 |
5 |
0.80 (against severe disease in young children) |
2+boosters |
15 |
5–8 |
2 |
0.80 (assuming immunity is long-lasting) |
3 |
10–15 |
8–10 |
2 |
0.80 (assuming immunity is long-lasting) |
3 |
10–15 |
Pathogen (Target Population) |
Type of Vaccine |
Cost of Development ($ millions) |
Probability of Successful Development |
Rabies virus (Individuals at high risk, plus post-exposure prophylaxis) |
Vero cell |
5 |
0.90 |
(Same) |
Glycoprotein produced by rDNA technology in mammalian cells |
3–5 |
0.85 |
(Birth cohort in areas of high risk) |
Attenuated live vector virus containing gene for protective glycoprotein antigen |
10–20 |
0.50 |
Respiratory syncytial virus (Infants at earliest possible age) |
Polypeptides produced by recombinant DNA technology |
25 |
0.80 |
Attenuated live virus |
25 |
0.80 |
|
Rotavirus (Infants at earliest possible age, preferably with oral polio vaccine) |
Attenuated high passage bovine rotavirus |
10 |
0.90 |
Attenuated low passage bovine rotavirus |
30 |
0.80 |
|
Rhesus monkey rotavirus |
30 |
0.80 |
|
Salmonella typhi (Children; young adults at risk; travelers from developed countries to endemic areas) |
Attenuated ga1E mutant S. typhi strain TY21a |
2 |
0.90 |
Aromatic amino acid dependent strains of S. typhi |
2 |
0.50 |
|
Shigella spp. (Infants at birth or earliest possible age; elderly for epidemic strains) |
Probably plasmid mediated outer membrane protein invasion determinant (small number of promising options need investigation to determine best approach) |
25–50 |
0.70 for polyvalent vaccine (0.85 for targeted S. dysenteriae and S. flexneri strains) |
Streptococcus A (Children,<3 to 4 years) |
Synthetic M protein segment (excluding portions cross-reacting with human tissue) |
50 |
0.80 |
Streptococcus pneumoniae (Infants) |
Conjugated polysaccharides, polyvalent |
30 |
0.80 |
Vibrio cholera (Children, especially <2 years) |
Genetically defined live mutant V. cholerae (A−B+ or A−B− with respect to toxin subunit synthesis |
25 |
0.75 |
Inactivated antigens |
10 |
0.65 |
|
Yellow fever virus (Young children) |
Attenuated live virus produced in cell culture |
15 |
0.95 |
Time to Licensure (years) |
Time to Adoption (years) |
Efficacy |
Number of doses |
Cost per dose (dollars) |
3 |
2–5 (dependent on price) |
0.99 |
3–5 |
10 |
3 |
2–5 (dependent on price) |
0.95 |
3–5 |
10 |
3 |
2 |
0.95 |
1 |
1 |
5 |
2 |
0.80 (against severe disease in young children) |
2+booster |
15 |
5 |
2 |
0.80 (against severe disease in young children) |
1–2 |
15 |
2 |
2 |
0.80 (against severe disease in young children) |
1 |
10 |
5 |
2 |
0.90 (against severe disease in young children) |
1 |
10 |
5 |
2 |
0.90 (against severe disease in young children) |
1 |
10 |
1 |
2 (endemic areas) |
0.80 |
2–3 |
2 |
5–8 |
|
0.80–0.90 |
2–3 |
2 |
approx. 10 |
2 |
0.80–0.90 (for a multivalent vaccine) |
1–2 |
2 |
6–8 |
2–5 |
0.80 (depends on adjuvant or carrier development) |
2 |
5 |
5 |
2 |
0.80 |
1–2 |
20 |
5–7 |
2 |
0.90 |
1–2 |
2 |
3–5 |
2 |
0.65 |
2–3 |
2 |
2–4 |
2 (endemic areas) |
0.90 |
1 |
5 |
TABLE 5.2 Predictions Table—Secondary
Pathogen (Target Population) |
Type of Vaccine |
Clinical Trial Difficulty |
Dengue virus (Infants and children in endemic areas; travelers to endemic areas) |
Attenuated live vector virus containing gene for broadly cross-reacting protective antigen |
Phase I trials must be in adults, in nonendemic areas. Some apprehension over possible enhancement effects for dengue and with new approach |
Escherichia coli (enterotoxigenic) (Infants < 6 months) |
A combination of purified colonization factor antigens and possibly other antigens |
Moderate. High attack rate in children and travelers makes evaluation possible in relatively small population. But may need to evaluate protection against certain serotypes or CFA types |
Genetically engineered attenuated strains |
Needs careful monitoring for reversion to virulence |
|
Hemophilus influenzae type b (Infants) |
Conjugated polysaccharide |
Need to be carried out in very young children |
Hepatitis A virus (Susceptibles of all ages; routine for preschool children) |
Attenuated live virus |
Large number of subjects needed. Initial trials in adults may give false concepts of immunogenicity and reactogenicity for children |
Polypeptide recombinant vaccine produced in yeast |
Large number of subjects needed |
|
Hepatitis B virus (Areas with high perinatal infection: all infants at birth (if possible). Other areas: all infants, simultaneous with other vaccinations, at earliest possible age) |
Polypeptide produced by recombinant DNA technology |
Relatively simple |
Japanese encephalitis virus (Children in epidemic and endemic areas; foreign visitors to epidemic regions) |
Inactivated virus produced in cell culture |
Difficult. Low clinical attack rate requires very large number of subjects. |
Mycobacterium leprae (Immunoprophylactic: all children in endemic areas. Immuno therapeutic: all recently infected individuals) |
Armadillo-derived M. leprae |
Low incidence and long incubation period requires many subjects and long time for trials |
Neisseria meningitidis (Infants, 3 to 6 months) |
Conjugated capsular polysaccharides, Groups A,C,Y, and W135 |
Difficult because epidemic disease is unpredictable |
Parainfluenza viruses (Infants) |
Trivalent, subunit vaccine (which must contain fusion proteins) |
|
Plasmodium spp. (All infants at risk, military personnel, travelers) |
Plasmodium falciparum, synthetic or recombinant sporozoite antigen preparation |
Mosquito challenge to volunteers |
Multivalent synthetic or recombinant sporozoite antigen preparation (P. falciparum, P. vivax, P. ovale, P. malariae) |
Mosquito challenge to volunteers |
Route of Administration |
Adverse Reactions |
Delivery Requirements |
Incorporation into EPI |
Intradermal |
Low grade fever, soreness, muscle aches |
Cold chain required; possible freeze-drying in future |
Yes |
Oral |
None |
Adjuvant use may reduce number of doses |
Yes |
Oral |
Possibly mild diarrhea in 20% |
Cold chain for lyophilized bacteria; adjuvant may be needed |
Yes |
Intramuscular |
5% local |
Refrigeration |
If in a polyvalent vaccine |
Parenteral, subcutaneous, or intramuscular |
Minimal |
Refrigeration of lyophilized preparation |
As a combination with IPV and other antigens |
Subcutaneous or intramuscular |
Minimal |
Refrigeration |
Might be combined with other nonliving vaccines |
Intramuscular or subcutaneous |
Negligible |
Refrigeration |
Could be incorporated at present; efficacy much improved if delivery possible at birth, i.e., with modified EPI schedules |
Subcutaneous |
Some possibility of life-threatening effects associated with current vaccines: allergic encephalomyelitis; acute viral encephalitis |
Nothing unusual |
Yes |
|
Feasible |
||
Intramuscular |
Minor local |
Refrigeration |
Feasible |
Subcutaneous |
None |
Nothing unusual |
|
Subcutaneous or intramuscular |
No data—unknown |
Adjuvant required, probably alum |
Probably |
Subcutaneous or intramuscular |
No data—unknown |
Nothing unusual |
Probably |
Pathogen (Target Population) |
Type of Vaccine |
Clinical Trial Difficulty |
Rabies virus (Individuals at high risk, plus post-exposure prophylaxis) |
Vero cell |
Little; depends on antibody response |
(Same) |
Glycoprotein produced by rDNA technology in mammalian cells |
Fatal natural disease and current availability of effective vaccine require rigorous proof of likely efficacy prior to field trials |
(Birth cohort in areas of high risk) |
Attenuated live vector virus containing gene for protective glycoprotein antigen |
Some possible apprehension over new approach |
Respiratory syncytial virus (Infants) |
Polypeptides produced by recombinant DNA technology |
Difficult. Needed very early in life; need rapid response. Vaccines won’t take in persons with antibodies |
Attenuated live virus |
|
|
Rotavirus (Infants, 0–6 months) |
Attenuated high passage bovine rotavirus |
Relatively easy. Pathogen present everywhere in world. Can do trial in children<1 year |
Attenuated low passage bovine rotavirus |
||
Rhesus monkey rotavirus |
||
Salmonella typhi (Children; young adults at risk; travelers from developed countries to endemic areas) |
Attenuated ga1E mutant S. typhi strain TY21a |
Trials largely completed; further work needed to optimize vaccine |
Aromatic amino acid dependent strains of S. typhi |
||
Shigella spp. (Infants at birth or earliest possible age; elderly for epidemic strains) |
Probably plasmid mediated outer membrane protein invasion determinant (small number of promising options need investigation to determine best approach) |
Moderate to difficult |
Streptococcus A (Children,<3 to 4 years) |
Synthetic M protein segment (excluding portions cross-reacting with human tissue) |
Moderate to very difficult |
Streptococcus pneumoniae (Infants) |
Conjugated polysaccharides, polyvalent |
Requires high degree of patient and physician cooperation. Multitude of bacterial types creates problems of accurately determining vaccine efficacy |
Vibrio cholera (Children, especially <2 years) |
Genetically defined live mutant V. cholerae (A−B+ or A−B−) with respect to toxin subunit synthesis |
Difficult. Need large populations in hyperendemic areas. Screening possible in volunteers |
Inactivated antigens |
Difficult. Same problems as live |
|
Yellow fever virus (Young children) |
Attenuated live virus produced in cell culture |
Ethical problems in field testing an improved vaccine when an effective one already exists |
Route of Administration |
Adverse Reactions |
Delivery Requirements |
Incorporation into EPI |
Intramuscular or subcutaneous |
Slight/none |
Refrigeration |
Probably not warranted. Duration of immunity not certain and post-exposure prophylaxis strategy preferred |
Same as at present (with alum adjuvant) |
Fewer reactions than current vaccine |
Nothing unusual |
Same |
Probably intradermal |
Unknown |
Probably nothing unusual |
Probably, depending on duration of immunity |
Subcutaneous |
None |
Nothing unusual |
Feasible |
Intranasal |
None |
Nothing unusual |
Feasible |
Oral |
Minimal or none |
Refrigeration and must be administered with food or milk |
Yes |
Oral (an alternative to enteric-coated capsules, i.e., a liquid vaccine formula, may be needed for infants and young children) |
None seen |
Refrigeration and moisture control for enteric-coated capsules |
Feasible (if new vaccine formulation is developed) |
None |
|
|
|
Oral |
None |
Probably lyophilization; possibly enteric-coated capsules |
Feasible |
Intramuscular |
|
|
Readily incorporated |
Parenteral |
Local soreness, low grade fever |
Refrigeration (4 C) |
Feasible, dependent on cost |
Oral |
Present prototypes cause mild diarrhea in 20% |
Cold chain for lyophilized bacteria |
Feasible |
Oral |
None expected |
Lyophilized probably will withstand moderate ambient temperatures |
Probably promptly in endemic areas |
Subcutaneous |
Minimal |
Refrigeration |
Feasible |
staff resource constraints), and the prospects for their development are discussed briefly in the supplement to this volume (see Appendix I). This supplement also includes information on pathogens that cause major global disease problems but that were considered unsuitable for accelerated vaccine development at this time.
VACCINE CANDIDATES AND TARGET POPULATIONS
One or more vaccine candidates for accelerated development have been identified for each disease. Vaccine descriptions usually are based on current research in specific areas. in some cases, however, the number of vaccine possibilities led the committee to base predictions on a combination of research findings and general knowledge about probable requirements for licensure.
To identify an appropriate vaccine target population, the committee considered the age distribution of the disease consequences (particularly of those conditions considered most desirable to avoid); the relative risk of illness in various geographic population groups; and accessibility to the health care system. For reasons described in the next chapter, the committee assumed that most of the vaccine candidates would be delivered through the World Health Organization Expanded Program on Immunization (WHO-EPI). The effects of this assumption on determinations of vaccine efficacy are outlined below.
PREDICTIONS ON VACCINE DEVELOPMENT
Predictions on vaccine development are an attempt to foresee events from 1985 until the time at which vaccine licensure might occur. Predictions are based solely on technical feasibility and not on judgments about the desirability of particular courses of action; no distinction has been made between public and private sector developmental resources.
Probability of Successful Development
The likelihood of bringing a specific vaccine to licensure within the time allotted, and with the predicted efficacy and other characteristics, is described as the probability of successful development. This probability is based on the state of current research, the complexity of the problem (e.g., the number of known serologic types), and characteristics of the natural immune response. The committee assumed that vaccine candidates would have to comply with safety and efficacy standards similar to those required by U.S. licensing regulations and the WHO.
Cost of Development
The estimate for the cost of development includes all future costs needed to bring the vaccine to licensure, irrespective of the funding source. Factors considered in estimating this amount were the current state of vaccine development, the complexity of the problem (e.g., difficulties encountered in culturing the pathogen), the availability of animal models, the number of alternatives to be tested in human clinical trials, and possible difficulties in conducting clinical trials or in establishing efficacy and safety in the target population.
Time to Licensure
The time to licensure is defined as the shortest time in which a vaccine could be licensed, if all developmental stages are completed without major delays. Factors considered in determining this time were similar to those for estimating the cost of development.
The committee also considered interrelationships among the probability of success, the cost of development, and the time to licensure; for example, the extent to which extra funding could significantly reduce the time to licensure for a particular vaccine.
PREDICTIONS ON VACCINE CHARACTERISTICS
The committee based its predictions on the characteristics of individual vaccines primarily on known characteristics of existing vaccines of similar type, for example, live attenuated virus, polysaccharide, or subunit vaccine. These predictions also incorporate assumptions about likely licensure requirements.
Efficacy
The prediction of a vaccine’s efficacy represents a population-based measure of protection rather than a measure of antibody production in an individual and is given by
Factors considered in estimating the efficacy were the type of pathogen and number of serotypes involved in the disease, the nature of the vaccine candidate, and the extent of immunity from natural infection.
Vaccine efficacy predictions also incorporate the assumption that vaccines will be administered at ages compatible with delivery through EPI (see Appendixes D-1 through D-19 for specific details). unfortunately, the EPI delivery schedule may not be ideal for some
candidates. Efficacy would be affected if recipients were unable to respond fully or to maintain immunity for a sufficient length of time.
Adverse Reactions
Adverse side effects of a vaccine, especially those likely to occur at very low frequency, are extremely difficult to predict but can seriously affect vaccine acceptance. Predictions about side effects are based on the nature of the new vaccine and its purity, and on observations of similar existing vaccines. To facilitate calculations, predictions concerning the incidence of adverse reactions are best expressed “per dose” rather than “per vaccinee.”
In this analysis, anticipated adverse reactions were judged, for the vaccines considered, to be negligible for the purposes of calculating potential vaccine benefits (see Chapter 7). For new contenders the likely reactions and their frequency should be evaluated to determine if an adjustment of the potential health benefits, which accounts for adverse reactions, is warranted.
Production Technology, Delivery Requirements, and Cost per Dose
The technical difficulty of producing a vaccine and the delivery (storage) requirements affect vaccine cost. The committee based its predictions in these areas on the nature of each vaccine candidate and on requirements for existing similar vaccines. Production technology and delivery requirements also affect the cost per dose, which, in turn, may affect vaccine acceptance.
Number of Doses and Route of Administration
The number of doses necessary to achieve a vaccine’s predicted protective efficacy and the route of administration also may affect the ease of integrating the vaccine into existing immunization programs. Predictions of these characteristics are based largely on the nature (including the probable antigenicity) of each vaccine candidate.
CONCLUSIONS
The predictions in Tables 5.1 and 5.2 resulted from extensive deliberations by the full committee on estimates made by a subgroup, with suggested revisions by many outside consultants. The predictions were designed to reflect relative differences in vaccine candidates’ prospects for development; they are not intended to be precise descriptions of future events. Predicting vaccine development is complicated by many factors, including the rapid pace of new advances in biotechnology. Some candidates excluded from the current analysis
(but discussed in the supplement to this volume) may soon need to be reassessed.
Although the outcome of scientific investigations cannot be predicted, the committee believes that the estimates and probabilities in Tables 5.1 and 5.2 are reasonable because they are based more on developmental than basic research investigations. The factors considered in arriving at each prediction have been stated in as much detail as possible, in the belief that regular reappraisal of these factors is essential. The flexibility of the model described in this report makes it easy to substitute alternative or updated predictions as they become available.
REFERENCE
Institute of Medicine. 1985. New Vaccine Development: Establishing Priorities, Volume I. Diseases of Importance in the United States. Washington, D.C.: National Academy Press.