Formalin and glutaraldehyde-treated toxoids, administered parenterally, also have been evaluated and found to be relatively ineffective (Finkelstein, 1984; Levine et al., 1983).

Studies in American volunteers and in endemic areas have established conclusively that the disease itself is an immunizing process (Finkelstein, 1984; Levine et al., 1983). The human host, presented with all the products of the cholera vibrio at the local level, undergoes maximal stimulation of secretory antibody. This finding has contributed to the notion that the best way to produce immunity against cholera would be to use a living, attenuated strain of V. cholerae that would colonize the gut and stimulate immunity, but not cause cholera.

The causative agent of epidemic cholera is Vibrio cholerae, serogroup 01. This group includes the El Tor biotype (the cause of the present pandemic) and the classical biotype, both of which occur as two serotypes, Inaba and Ogawa. The bacterium is gram-negative, comma-shaped, oxidase positive, and motile by means of a single polar flagellum. It is identified by its agglutinability in 0 group 1 and type-specific, Inaba or Ogawa, antisera. Other ancillary characteristics, such as growth and reaction on selective media (e.g., TCBS agar), and particular biochemical reactions (e.g., fermentations) can be useful in suggesting its identity, but confirmation depends on serologic reactions. The definition of the Vibrio cholerae species recently has been expanded to include a diverse variety of other vibrios that are known collectively as non-0 group 1 V. cholerae (because they do not agglutinate in O group 1 antisera). Occasional rare strains of non-O group 1 vibrios, formerly called NAG (nonagglutinable) or NCV (noncholera vibrios) produce a cholera-related enterotoxin and have been associated with diarrheal disease.

Despite the noninvasive nature of cholera, a vigorous immune response is induced in the host. This response is manifested both by circulating IgM and IgG antibodies and by secretory IgA antibodies against the lipopolysaccharide (LPS) somatic antigen, the enterotoxin, and other less well studied antigens. The antibacterial antibodies are vibriocidal (in the presence of complement) and agglutinating, and the anti-enterotoxin is neutralizing. Either or both are protective in animal models, and there is some evidence that they may interact synergistically. The disease is an immunizing process, and it has been demonstrated that volunteer convalescents are resistant to rechallenge for at least 3 years. In endemic areas, cholera is less frequent in adults who have circulating antibody.