Recent Events of Significance

Outbreaks due to revertant Sabin vaccine strains. The discovery in October 2000 of a paralytic disease outbreak due to a revertant Sabin vaccine strain is a watershed event with regard to longer-term policies for polio control. Analysis of an outbreak of ã16 cases of paralytic disease on Hispaniola (the island shared by the Dominican Republic and Haiti) has revealed that the virus responsible was a vaccine-derived type 1 strain of poliovirus, whose genetic divergence from the original vaccine strain indicates that it had been circulating for ∼24 months before causing illness [6]. A recent retrospective study of poliovirus isolates from Egypt has also revealed that ∼32 cases of paralytic disease occurred from 1988 through 1992 were associated with type 2 vaccine–derived isolates [7]. Each of these strains is readily distinguished from naturally occurring ("wild") strains of poliovirus. Therefore, their occurrence does not call into question WHO's primary goal of eradicating polio; that is, interrupting the transmission of wild polioviruses.

The findings in Hispaniola and Egypt were unexpected. During the ã40 years that OPV has been in use—during which time billions of vaccinations have been given—it had never before been documented that a Sabin polio vaccine strain reverted and assumed the transmissibility characteristics of a wild-type strain [8]. It was known since before licensure of the vaccine that the different polio vaccine strains regularly showed some reversion to virulence (as demonstrated in tests in monkeys) when they passed through the intestinal tract [9]. Soon after licensure in 1962, it was discovered that paralytic disease could occasionally occur in vaccine recipients and their close contacts [10]. All cases were believed to be isolated occurrences, however. It is possible that outbreaks due to vaccine-derived strains of poliovirus may have occurred, but in the absence of techniques for nucleotide sequence analysis, they were not differentiated from naturally occurring outbreaks of wild poliovirus infection. Because the occurrence of cases attributable to OPV was sufficiently rare and the characteristics of the vaccine were sufficiently superior to those of IPV, OPV continued to be recommended as the vaccine of choice in most countries.

Paul Fine, in commenting on the strategy option described by Wood et al. [5] before the most recent revelations from Hispaniola and Egypt, expressed concern about the reversion of OPV to virulence: "Even with much more information than is now available, we are unlikely to be fully confident that OPV viruses cannot persist where there are high numbers of susceptible [persons] living in poor hygienic conditions" ([11], p. 358). If the virus can circulate undetected for 24 months in Hispaniola, where surveillance, although not ideal, is generally better than in many developing countries, it is plausible that, in some countries, the virus could persist undetected for even longer periods before causing an outbreak of paralytic disease. However, many questions about these revertant strains remain to be defined: specifically, the frequency and conditions under which such strains might arise, as well as their contagiousness and propensity to cause paralytic illness.

Long-term carriage and excretion of polio vaccine viruses. Possible additional troublesome sources for revertant vaccine viruses are persons with certain hereditary immune deficiency disorders. Nine persons have been found to excrete vaccine viruses for periods of time ranging from a few months to nearly 10 years [5]. Few studies have provided indications about the frequency of this phenomenon but, because the hereditary risk factors may occur in as many as 1 of every 10,000 persons [8], it is probable that there have been and are a number of other persons who can and have carried and excreted the virus.

Problems in strife-torn areas. For countries to entertain the possibility of stopping vaccination, they must first have full confidence that virus transmission has been interrupted worldwide. Understandably, a special concern at this time are strife-torn areas where government authority does not extend throughout the country, as is the case in Liberia, Somalia, Sudan, Angola, Afghanistan, and the Democratic Republic of the Congo. Although polio control programs are in place in each of these countries, there are large areas in each country that are inaccessible or difficult for health personnel to access. The conflicts in most of these countries are long-standing and severe; it seems unlikely that they will be settled sufficiently any time soon to permit the field work necessary to interrupt polio virus transmission and to conduct careful surveillance for 3 years to ensure that transmission has ceased.

Such problems were also encountered during the smallpox vaccination program, but usually the program could be executed during periods of as little as 3–6 months of comparative tranquility in an area. However, there are 2 major differences in dealing with the control of smallpox and polio in insecure areas that are difficult to access. First, surveillance to determine whether and where the smallpox virus was present was comparatively simpler than polio surveillance and could be accomplished rapidly. The vast majority of patients had a distinctive rash; there were no asymptomatic patients and no long-term carriers. Thus, a team could rapidly search an area and determine whether the virus was present and, if it was, determine its extent and distribution without consulting a laboratory. Contrast this with the surveillance problems presented by poliomyelitis, in which there are ⩾200 asymptomatic infections for every paralyzed patient. The only way one can ascertain whether the virus continues to circulate in an area is by an extended period of surveillance, during which a great number of stool specimens are examined. Time and access to a laboratory are critical requirements.

The second difference has to do with the efficacy of vaccination. One inoculation of smallpox vaccine protects nearly 100% of those vaccinated. However, in areas where polio is endemic, at least 3 doses of OPV, and often 5 or 6 doses, are required to achieve protective levels of 90% against types I and III poliovirus, which are the predominant paralytic strains [9]. Therefore, although vaccination immunity against smallpox could be increased rapidly in areas of conflict where access was possible for only a few weeks or months, this is not possible for polio.

Vaccination. Given the practical problems of achieving and confirming eradication in strife-torn areas and the recent findings that demonstrate the potential for polio vaccine strains to cause outbreaks of paralytic disease, there would appear to be no option at this time but to plan to continue vaccination programs for the foreseeable future and perhaps indefinitely. By sustaining reasonably high levels of population immunity, protection would also be provided against the possibility of catastrophic epidemics, should poliovirus accidentally escape from a laboratory or be disseminated deliberately.

At present, most industrialized countries are using IPV and most developing countries are using OPV. There are cogent reasons for this difference. Soon after licensure of OPV in 1962, it became the vaccine of choice for almost all countries. This was primarily because it was much easier to administer, it provided substantial intestinal immunity against infection with wild poliovirus, and it spread to close contacts, thereby protecting a number of people who were not themselves vaccinated. The only drawback to the use of OPV was the occasional occurrence of a case of paralytic illness among vaccine recipients and, sometimes, their close contacts. This occurred with a frequency of about 1 case per 1 million recipients of first doses of vaccine and, overall, about 1 case per 3 million doses of vaccine distributed [9].

Over time, many industrialized countries opted to use IPV to avert this risk, despite several attributes that continued to favor the use of OPV. Specifically, IPV was substantially more costly and had to be given by injection, instead of orally. Because it is an inactivated product, it does not spread to household contacts as did OPV, and although it confers pharyngeal immunity (as does OPV), it provides little intestinal immunity. For industrialized countries with ample health care resources, such factors as cost of vaccine, ease of administration, and spread of the vaccine were marginal considerations. Moreover, because virus transmission in industrialized countries is believed to occur primarily through droplet infection from the pharynx, the lack of intestinal immunity was not considered important. For developing countries, however, each of the special characteristics of OPV is of special relevance. This is especially true for intestinal immunity, because transmission of polio in these countries is believed to occur primarily by the fecal-oral route.

Under epidemic conditions, OPV is, by far, the preferred vaccine in all countries. Within days after community-wide application, OPV provides immunity and blocks circulation of wild virus strains. IPV, by contrast, requires 6–8 weeks for a satisfactory antibody response and would appear to have little inhibitory effect, at least in developing countries, on the spread of wild polio viruses.

Given present realities, there is no practical option other than continuing polio vaccination for the foreseeable future. One proposed alternative to using IPV in some countries and OPV in others would be to use IPV in all countries, perhaps combined with diphtheria-tetanus-pertussis vaccine [5]. Although an apparently simple solution, this approach is far more problematic than it might seem. Combining antigens into a single product while retaining the antigenicity of each of the products has proved difficult [12]. This problem is illustrated by the fact that, even today, there is no vaccine product available in the United States that combines IPV with any other antigen, although such a product is available in Canada and some European countries. Because diphtheria-tetanus-pertussis vaccine is produced in many different national laboratories, each of which uses different methods and materials, the task of converting all or even most of these laboratories into effective producers of a combined product would be a daunting challenge.

A more serious problem for developing countries is the questionable effectiveness of IPV. A carefully conducted WHO collaborative study in Thailand, Oman, and The Gambia revealed that after receiving 3 doses of IPV at 6, 10, and 14 weeks of age, as many as one-third of children were unprotected against type I polio, the most common type of wild virus [13]. Why IPV performed so poorly in this study, even though it consistently provides satisfactory serological responses in industrialized countries, is under study.

A third problem would be production of IPV after eradication was achieved. The three polio strains used in the inactivated vaccine are known to be highly virulent; all 3 were recovered from patients with severe paralytic polio, 1 of whom died [9]. To ensure that, during the posteradication period, these virulent polio virus strains did not escape the vaccine production facility, the facilities would be required to be more secure than any now producing this or any other biological product. The feasibility and costs of doing so are unclear.

Finally, if OPV use were to be stopped, it is unclear what provisions could be made to ensure both vaccine-production capability and the availability of large emergency reserves of vaccine. Unlike smallpox vaccine, which remains potent almost indefinitely at -20°C, OPV has a storage life of only a few years. The costs of sustaining a large standby production facility would not be inconsequential, even if the project were feasible.

Laboratory Containment of Poliovirus

The risk of poliovirus escaping from a laboratory bears significantly on vaccination policy. The risks are reviewed by Wood et al. [5], who identify 12 instances between 1941 and 1976 when laboratory-associated cases occurred. As they point out, the risk of an escape is low but cannot be ignored. Accordingly, they propose “the formidable task of locating the many laboratories that have stocks of infectious and potentially infectious wild poliovirus and ensuring that such stocks are destroyed or adequately contained.” They add that “the bigger challenge will be to identify all laboratories with potentially infectious clinical, epidemiological, research, or environmental specimens collected for other purposes, in a geographical area and at a time of wild poliovirus endemicity.” The plan calls for all such specimens to be dealt with under biosafety laboratory level–3 security as soon as polio virus transmission is thought to have been interrupted (a point once identified as December 2002). They propose that, after eradication is confirmed, materials containing poliovirus be handled under biosafety laboratory level–4 conditions (a point once expected to be December 2005).

The magnitude of the projected plan for dealing with laboratory specimens is formidable, and the likelihood of meeting its objectives is questionable. Racaniello, a research scientist, in commenting on the WHO's proposed program, characterizes the task as “mind-boggling, particularly in light of the absence of an enforcement authority,” “simplistic,” and “unsolvable” ([14], p. 360). In the course of the smallpox eradication program, a similar plan for laboratory containment was instituted, but the magnitude of the task was far less because there were very few diagnostic laboratories for smallpox and only a handful of investigators, and the search was specifically for known specimens of smallpox only. Nevertheless, the task was difficult; a number of investigators and national authorities resisted destroying specimens or transferring specimens to 1 of 2 designated WHO reference laboratories. The proposed plan for poliovirus is many orders of magnitude more complicated.

Efforts need to be made to minimize the risk that poliovirus will escape from a laboratory after transmission of wild poliovirus is interrupted. However, if vaccination is an ongoing process, the problems of virus escape are significantly diminished. With diminished risk, use of less draconian (although still onerous) measures for control of virus specimens would be possible, and there would be a greater likelihood of cooperation of laboratories and investigators. Efforts could be focused on those laboratories engaged in diagnostic or research studies of the poliovirus itself, rather than on all laboratories with fecal or other specimens in which poliovirus was an incidental and unknown contaminant.

Polio Program Goals

The ultimate goal of polio eradication has been to stop transmission of the naturally occurring ("wild") polioviruses. As in the smallpox program, eradication of the virus itself would be desirable. However, given the many laboratories that have materials believed to contain poliovirus and the practical impossibility of verifying the destruction of all materials that might contain poliovirus, the definition of the goal has been stated in measurable terms (as was the case for smallpox eradication): stopping virus transmission in humans.

Paralytic illnesses caused by OPV have been considered to be in a special category, on the basis of the fact that viruses isolated in such cases can be clearly differentiated from the naturally occurring wild virus strains. The occurrence of a few cases of vaccine-associated paralysis each year has been properly considered to have no bearing on a country's polio-free status. For example, the region of the Americas is still considered to be polio-free, despite the outbreak of paralytic disease caused by a revertant type I vaccine strain. For the present, at least, it makes sense to consider these viruses (both the strains and the diseases they cause) in a separate category and to investigate all suspected outbreaks carefully.