Development of an improved vaccine for anthrax.

Bacillus anthracis are aerobic or facultatively anaerobic Gram-positive, nonmotile rods measuring 1.0 μm wide by 3.0–5.0 μm long. Under adverse conditions, B. anthracis form highly resistant endospores (Figure 1). These are found in soil at sites where infected animals previously died. Interest in the pathogenesis, immunity, and vaccine development for anthrax was heightened by the deliberate contamination of mail with B. anthracis spores soon after the September 11 attacks. At that time, the only US-licensed human vaccine (anthrax vaccine adsorbed, or AVA) was not available because the manufacturer, BioPort Corp., had not received FDA certification of its new manufacturing process. Although Pasteur had already demonstrated protection of sheep by injection of heat-attenuated B. anthracis cultures in 1881, our current knowledge of immunity to anthrax in humans remains limited. Widespread vaccination of domesticated animals with attenuated strains such as the Sterne strain began in the 1930s and has virtually abolished anthrax in industrialized countries. In the US, the licensed human vaccine (AVA, newly renamed BioThrax) is an aluminum hydroxide–adsorbed, formalin-treated culture supernatant of a toxigenic, noncapsulated, nonproteolytic B. anthracis strain, V770-NP1-R, derived from the Sterne strain (1). AVA was developed in the early 1950s, when purified components of B. anthracis were not available. Its only demonstrable protective component is the protective antigen (PA) protein (2). A similar culture supernatant–derived human vaccine is produced in the United Kingdom. Data from a 1950s trial of wool-sorters immunized with a vaccine similar to AVA, coupled with long experience with AVA and the United Kingdom vaccine, have shown that a critical level of serum antibodies to the B. anthracis PA confers immunity to anthrax (3, 4). As early as 1959, a British Ministry of Labour report noted that, following the introduction of regular immunization the previous year, the staff of the Government Wool Disinfection Station in Liverpool were free of the disease “despite the high risk to which they are exposed” (5). AVA also protects laboratory animals and cattle from both cutaneous and inhalational challenge with B. anthracis (1, 6, 7). Although safe and efficacious (8), AVA has limitations that justify the widespread interest in developing improved vaccines consisting solely of well-characterized components. First, standardization of AVA is based on the manufacturing process and a potency assay involving protection of guinea pigs challenged intracutaneously with B. anthracis spores (7, 9). PA is not measured in the vaccine, and there is no standardized assay of PA antibodies in animals or humans vaccinated with AVA. These factors probably explain why it has been difficult to maintain consistency of AVA. Second, this vaccine contains other cellular elements that probably contribute to the relatively high rate of local and systemic reactions (8). Finally, the schedule of AVA administration (subcutaneous injections at 0, 2, and 4 weeks and 6, 12, and 18 months with subsequent yearly boosters) is probably not optimal. This schedule, introduced in the 1950s, was designed for rapid induction of immunity (10), but it was recently shown that increasing the interval between the first two injections enhances the level of AVA-induced antibodies to PA (11). Moreover, there is no experimental support for including the injections given at 6, 12, and 18 months.