Anthrax vaccine: short-term safety experience in humans.

Bacillus anthracis is the major terrorist and biological warfare agent of concern to civilian and military medical planners. The licensed anthrax vaccine, adsorbed (AVA) is believed to be an effective prophylactic medical countermeasure against this threat. Our objective in this report was to expand the safety database for this vaccine by assessing data on self-reported, short-term safety of AVA during more than 25 years of use, measured by local and systemic adverse events temporally associated with the administration of AVA. A minority of AVA recipients reported systemic and injection site reactions. Females reported a higher incidence of injection site and systemic adverse events than males. Data show a difference in incidence of local reactions between lots. A prospective, randomized, placebo-controlled study to actively examine reactogenicity is needed to more completely define the extent and nature of reactions associated with receipt of AVA in humans as well as to confirm the gender lot differences in local reaction rates.

Efficacy of a human anthrax vaccine in guinea pigs, rabbits, and rhesus macaques against challenge by Bacillus anthracis isolates of diverse geographical origin.

The efficacy of a licensed human anthrax vaccine (Anthrax Vaccine Adsorbed (AVA)) was tested in guinea pigs, rabbits, and rhesus macaques against spore challenge by Bacillus anthracis isolates of diverse geographical origin. Initially, groups of Hartley guinea pigs were vaccinated at 0 and 4 weeks with AVA, then challenged intramuscularly at 10 weeks with spores from 33 isolates of B. anthracis. Survival among the vaccinated groups varied from 6 to 100%, although there were no differences in mean time to death among the groups. There was no correlation between isolate virulence and variable number tandem repeat category or protective antigen genotype identified. New Zealand white rabbits were then vaccinated with AVA at 0 and 4 weeks, and challenged at 10 weeks by aerosol with spores from six of the isolates that were highly virulent in vaccinated guinea pigs. AVA completely protected the rabbits from four of the isolates, and protected 90% of the animals from the other two isolates. Subsequently, two of these six isolates were then used to challenge rhesus macaques, previously vaccinated with AVA at 0 and 4 weeks, and challenged at 10 weeks by aerosol. AVA protected 80 and 100% of the animals from these two isolates. These studies demonstrated that, although AVA confers variable protection against different B. anthracis isolates in guinea pigs, it is highly protective against these same isolates in both rabbits and rhesus macaques.

Anthrax vaccine: increasing intervals between the first two doses enhances antibody response in humans.

The influence of dosing interval on the human antibody response to anthrax vaccine adsorbed (AVA) was evaluated in two retrospective serological studies. In both studies, the interval between the first two doses was 2, 3 or 4 weeks. In the first study, banked sera were selected from 89 at-risk individuals at a mean time of 13 days after the second dose of vaccine. In the second study, banked sera were selected from 51 at-risk individuals at a mean time of 48 days following the first dose of AVA. In both studies, the geometric mean anti-protective antigen IgG antibody titer increased significantly as the interval between the two doses increased from 2 to 4 weeks (p=0.0005-0.029). In the first study, the seroconversion rate also increased as the interval between the first two doses increased (p=0. 0034). A prospective, randomized study has been completed and is being analyzed to confirm these findings.

Immunogenicity of an inactivated Rift Valley fever vaccine in humans: a 12-year experience.

Rift Valley fever (RVF) virus causes serious and fatal disease in animals and man. To protect personnel who work with RVF virus in the laboratory, or troops who may be exposed to this virus, the US Army successfully developed an improved version of inactivated RVF vaccine, TSI-GSD-200. From early 1986 to late 1997, 598 at-risk workers at the US Army Medical Research Institute of Infectious Diseases (USAMRIID) were vaccinated as part of an occupational safety and health program. The subjects of this study received three subcutaneous doses (0, 7 and 28 days) of 0.5 ml of TSI-GSD-200. A total of 540 vaccinees (90.3%) initially responded (group A) with an 80% plaque-reduction neutralization antibody titer (PRNT80) of > or =1:40; whereas 58 subjects (9.7%) were initial nonresponders (group B) failing to achieve this titer. Volunteers who either failed to respond or who achieved a titer of > or =1:40 but whose titer waned below 1:40 were boosted 1-4 times with the same vaccine. Among 247 group A subjects who received the first recall injection, 242 (98%) were successfully boosted, achieving a PRNT80 > or =1:40. Thirty-three of 44 (75%) initial nonresponders were converted to responder status after the first booster, which is a lower rate than that of group A (P < 0.001). After the primary series and the first booster, Kaplan-Meier analysis showed 50% probability of group A members maintaining a titer of > or =1:40 for approximately eight years; whereas group B had a 50% probability of maintaining a titer for only 204 days. Group A immune response rates to boosts 1-4 ranged from 87 to 100% with geometric mean titers (GMTs) ranging from 80 to 916. Boosts 1-4 immune response rates of group B volunteers ranged from 67 to 79% with GMTs ranging from 90 to 177. Minor side effects to TSI-GSD-200 were noted in 2.7% of all vaccinees after primaries and 3.5% of all vaccinees who had primaries and up to four boosters. We conclude that the use of TSI-GSD-200 is safe and provides good long-term immunity in humans when the primary series and one boost are administered.

Comparative efficacy of experimental anthrax vaccine candidates against inhalation anthrax in rhesus macaques.

The authors examined the efficacy of Bacillus anthracis protective antigen (PA) combined with adjuvants as vaccines against an aerosol challenge of virulent anthrax spores in rhesus macaques. Adjuvants tested included i) aluminum hydroxide (Alhydrogel), ii) saponin QS-21 and iii) monophosphoryl lipid A (MPL) in squalene/lecithin/Tween 80 emulsion (SLT). Animals were immunized once with either 50 micrograms of recombinant PA plus adjuvant, or with Anthrax Vaccine Adsorbed (AVA), the licensed human anthrax vaccine. The serological response to PA was measured by enzyme linked immunosorbent assay. Lymphocyte proliferation and serum neutralization of in vitro lethal toxin cytotoxicity were also assayed. In all vaccine groups, anti-PA IgM and IgG titers peaked at 2 weeks and 4-5 weeks postimmunization, respectively. Five weeks postimmunization, animals in all vaccine groups demonstrated PA-specific lymphocyte proliferation and sera that neutralized in vitro cytotoxicity. Six weeks after immunization, the animals were challenged by aerosol with approximately 93 LD50 of virulent anthrax spores. Animals were bled daily for 1 week to monitor bacteremia, and deaths were recorded. Anti-PA ELISA titers in all groups of immunized animals were substantially increased 2 weeks after challenge. One dose of each vaccine provided significant protection (> 90%) against inhalation anthrax in the rhesus macaques.

Protective efficacy of a live attenuated vaccine against Argentine hemorrhagic fever. AHF Study Group.

Argentine hemorrhagic fever (AHF), caused by the arenavirus Junin, is a major public health problem among agricultural workers in Argentina. A prospective, randomized, double-blind, placebo-controlled, efficacy trial of Candid 1, a live attenuated Junin virus vaccine, was conducted over two consecutive epidemic seasons among 6500 male agricultural workers in the AHF-endemic region. Twenty-three men developed laboratory-confirmed AHF during the study; 22 received placebo and 1 received vaccine (vaccine efficacy 95%; 95% confidence interval [CI], 82%-99%). Three additional subjects in each group developed laboratory-confirmed Junin virus infection associated with mild illnesses that did not fulfill the clinical case definition for AHF, yielding a protective efficacy for prevention of any illness associated with Junin virus infection of 84% (95% CI, 60%-94%). No serious adverse events were attributed to vaccination. Candid 1, the first vaccine for the prevention of illness caused by an arenavirus, is safe and highly efficacious.

Steady-state minute volume determination by body-only plethysmography in juvenile rhesus monkeys.

When one is using nonhuman primates for studying the inhalation of infective or toxic agents, a respiratory minute volume (MV) range of +/- 50 ml is desirable to ensure the accurate delivery of calculated doses of the aerosolized agent. When one is working with highly infective or toxic agents, it is desirable to anesthetize the animals and to separate the plethysmograph, used to measure MV, from the aerosol chamber, used to administer agents, in order to minimize decontamination procedures and to maximize safety. In our laboratory the sequential completion of these procedures requires at least 20 min. Therefore it is necessary to find an anesthetic that achieves a +/- 50 ml steady-state MV for at least 20 min and that does not change when an animal is transported from one apparatus to another. Using 2.6- to 4.0-kg, 14- to 18-month-old rhesus macaques, we determined that tiletamine/zolazepam induced a steady-state MV of 48 +/- 17.8 min, beginning 21.5 +/- 4.7 min after injection of the anesthetic agent. This MV did not significantly change when animals were transported. The use of ketamine and ketamine/acepromazine resulted in a steady-state MV period of 11.5 +/- 4.5 and 22.0 +/- 7.9 min respectively. When we compared these findings with previously reported mathematical estimations of MV based on functions of weight or respiratory rate, we further determined that the accurate measurement of MV before each aerosol exposure was critical for calculating inhaled doses of the agent.

Long-term duration of detectable neutralizing antibodies after administration of live-attenuated VEE vaccine and following booster vaccination with inactivated VEE vaccine.

The US Army successfully developed a live-attenuated Venezuelan Equine Encephalitis (VEE) vaccine, TC-83, in 1961, and subsequently developed a formalin-inactivated vaccine, C-84, in 1974. Initial evaluation of both vaccines was promising, but no long-term safety and immunogenicity data have been reported. This study is the first analysis of the long-term safety and immunogenicity of TC-83 and C-84. From January 1976 to December 1990, 821 laboratory workers at the USAMRIID were vaccinated with a single 0.5 ml subcutaneous (s.c.) dose of TC-83; 128 were boosted with a single 0.5 ml s.c. dose of C-84. Eighty-two per cent of vaccinees responded to TC-83 with an 80% plaque reduction neutralization titer (PRNT80) of > or = 1:20. Minor side-effects were noted in 23% of vaccinees. No long-term sequelae were recorded. Kaplan-Meier analysis showed a 60% probability of vaccinees maintaining a PRNT80 of > or = 1:20 for 5.5-8 years. C-84 was given to two groups: 76 initial nonresponders to TC-83, Group A, and 52 initial responders to TC-83 whose PRNT80 became < 1:20 over time, Group B. C-84 successfully boosted 76% of Group A and 100% of Group B to a PRNT80 > or = 1:20 Kaplan-Meier analysis showed 100% probability of Group B members maintaining a titer of > or = 1:20 for the duration of follow-up, which, in some cases, exceeded 10 years; while Group A had only a 60% probability of maintaining a titer for 1-2 years. Only minor local reactions to C-84 were noted in 6.3% of vaccinees. We conclude that, although TC-83 is reactogenic, when administered as the primary vaccine and C-84 is administered as a boost, these vaccines provide good long-term immunity and are safe in humans. However, a single dose vaccine that is more immunogenic and less reactogenic is needed.

Exposure variables in bancroftian filariasis in the Nile Delta.

To demonstrate focality of filariasis within endemic rural areas and to define exposure variables which may influence this phenomenon, the population of an agrarian endemic village, of 12,500 individuals, in the Nile Delta of Egypt was censused. A sequential sample of individuals residing in every fifth house was tested for microfilaremia (239 households with 8.6 +/- 3.5 individuals per household (HHD). Three areas of the village were tested simultaneously and a questionnaire was filled out for each sampled HHD with special emphasis given to the entomological and environmental factors that might affect filarial infection. One area (area A) had a higher intensity of larvae and biting adults of the main filarial vector, Culex pipiens, than the other two areas (areas B and C). Of the 1488 persons who agreed to be tested in the three areas 181 (12.2%) were microfilaremic. Microfilaremia prevalences were the same in males and females and microfilariae were present in all age groups. Filarial infection was most prevalent in area "A" (1.16 +/- 0.14 infected people per HHD) than in area "B" (0.44 +/- 0.11) or "C" (0.72 +/- 0.10) (ANOVA; p = 0.0003). several possible predictor variables were analyzed by logistic regression with the presence of infection as the response variable. Among individuals residing around the main Cx. pipiens development sites, those living in houses facing vacant land are exposed to more mosquito bites and had a greater chance of having filarial infection (relative risk [RR] = 1.5; logistic regression, P = 0.0089). People residing in large households had a reduced chance of having filarial infection (RR = 0.87; logistic regression, p = 0.0015). These data show that the distribution of microfilaremic individuals is uneven within the study village and suggest that small HHD and houses that bordered open areas containing mosquito development sites are potential risk factors for acquiring filarial infection.

Non-toxigenic derivatives of the Ames strain of Bacillus anthracis are fully virulent for mice: role of plasmid pX02 and chromosome in strain-dependent virulence.

The toxin-encoding plasmid pX01 and capsule-associated plasmid pX02 are required for full virulence of Bacillus anthracis in some animals. However, the non-toxigenic pX01-cured derivatives of certain anthrax strains are not completely attenuated for mice, and their virulence is strain-dependent. The strain-related differences were partially associated with plasmid pX02 as demonstrated by pX02 transductants of the attenuated vaccine strain UM23-1 cured of pX01. To determine the virulence of non-toxigenic variants of virulent strains, we isolated pX01- derivatives of the Vollum 1B strain and the more 'vaccine-resistant' Ames strain which carried pX02 from either Ames or Vollum 1B. The 50% lethal dose (LD50) values of the derivatives of both strains which carried the Ames pX02 were not significantly different from the LD50s of the pX01+ pX02+ strains (and were lower than those of pX01+ pX02- strains). pX02+ derivatives of strain UM23-1 were less virulent than the comparable Ames and Vollum 1B strain derivatives, emphasizing a role for chromosomal loci in the virulence of the latter two strains. Non-toxigenic isolates which carried the Ames pX02 were more virulent for CBA/J mice than those with Vollum 1B pX02, and the differences were mouse strain-dependent. The pX01- pX02+ strains multiplied and achieved high concentrations systemically.

Pathogenesis of Rift Valley fever in rhesus monkeys: role of interferon response.

Rhesus monkeys inoculated intravenously with Rift Valley fever (RVF) virus presented clinical disease syndromes similar to human cases of RVF. All 17 infected monkeys had high-titered viremias but disease ranged from clinically inapparent to death. Three (18%) RVF virus-infected monkeys developed signs of hemorrhagic fever characterized by epistaxis, petechial to purpuric cutaneous lesions, anorexia, and vomiting prior to death. The 14 remaining monkeys survived RVF viral infection but, 7 showed clinical signs of illness characterized by diminished food intake, cutaneous petechiae, and occasional vomiting. The other 7 monkeys showed no evidence of clinical disease. All monkeys had detectable serum interferon 24-30 h after infection, but 4 of 7 monkeys that did not develop clinical illness had serum interferon titers within 12 h after infection. In lethally infected macaques, indices of hepatic function and blood coagulation were abnormal within 2 days, implicating early pathogenetic events as critical determinants of survival. Serum transferase values were elevated in proportion to severity of clinical disease and outcome of infection. Both myocardial damage and laboratory evidence consistent with disseminated intravascular coagulation were present in fatal infections. All surviving monkeys developed neutralizing antibodies to RVF virus 4-7 days after infection, and this coincided with termination of viremia. Two fatally infected monkeys were viremic until death on days 6 and 8, and the third cleared viremia on day 5 and developed antibody on day 6 but died on day 15. There was a significant correlation between a delayed interferon response and mortality, suggesting that the early appearance of interferon was influential in limiting the severity of disease.

Prevention of Rift Valley fever in rhesus monkeys with interferon-alpha.

Prophylactic and therapeutic efficacy of recombinant leukocyte A interferon (rIFN-alpha A) and Sendai virus-induced human leukocyte interferon (HuIFN-alpha) administered intramuscularly to Rift Valley fever virus (RVFV)-infected rhesus monkeys was studied. Clinical, virologic, immunologic, and hemostatic parameters were monitored. Five daily inoculations of 5 X 10(5) units of either interferon product per kilogram of body weight, initiated 24 hours before or 6 hours after RVFV infection, prevented or greatly suppressed viremia. No clinical signs of disease or laboratory evidence of impaired hemostasis was observed. Serum neutralizing antibody to RVFV was detected within 6 days of virus inoculation. Prophylactic administration of 5 X 10(4) or 5 X 10(3) units of rIFN-alpha A per kilogram also limited viremia, hepatocellular damage, and hemostatic derangement. Untreated, RVFV-infected, control monkeys developed high-titered viremia, clinical disease, and impaired hemostasis. These data suggest that rIFN-alpha A and HuIFN-alpha are effective in protecting RVFV-infected rhesus monkeys from viremia and hepatocellular damage and may be beneficial in human RVF infection.

Hemostatic derangement produced by Rift Valley fever virus in rhesus monkeys.

Rift Valley fever (RVF) is an important cause of disease in animals and humans in sub-Saharan Africa. In a small percentage of human cases, the disease is complicated by hemorrhage, which often is associated with a fatal outcome. Inoculation of rhesus monkeys with the Zagazig Hospital strain of RVF virus produced a clinical picture similar to illness in humans. Ten of 17 monkeys developed clinical evidence of hemostatic impairment. When coagulation tests were performed, this group of monkeys had significant abnormalities, including evidence for disseminated intravascular coagulation. These abnormalities were much less pronounced in the remaining seven monkeys-whose only sign of illness was transient fever-and, in general, they paralleled the level of viremia and the degree of elevation in levels of serum hepatic enzymes. Autopsy of the three monkeys with severe disease revealed hepatic necrosis.

Reduced survival of adult Culex pipiens infected with Rift Valley fever virus.

The effect of Rift Valley fever (RVF) viral infection on the survival of female Culex pipiens was examined. In 3 experiments in which mosquitoes ingested RVF virus, there was a 44% decrease in survival to days 14-16 for transmitting vs. nontransmitting mosquitoes, and a 48% decrease in survival for individuals with disseminated vs. nondisseminated infections. These results were corroborated by other experiments in which survival of mosquitoes intrathoracically inoculated with RVF virus was compared with that of those inoculated with diluent. In both the per os and inoculation tests, uninfected mosquitoes survived significantly longer than infected mosquitoes. Even though mosquitoes with disseminated infections had a lower survival rate than did uninfected mosquitoes, dissemination and transmission rates were similar at days 7 and 14-18 after the infectious bloodmeal. This suggests that nondisseminated individuals were developing disseminated infections and becoming capable of transmitting virus between days 7 and 14-18 at approximately the same rate older transmitters were dying. The decreased survival associated with RVF viral infection should be considered in predictive models of this disease.

Evaluation of a new Rift Valley fever vaccine: safety and immunogenicity trials.

A formalin-inactivated Rift Valley fever vaccine prepared in primary monkey kidney cells has been used to protect laboratory workers from disease since 1967. A similar but improved vaccine was prepared in 1978-1979 using well characterized diploid fetal rhesus lung cells. In initial clinical trials reported here, the new vaccine elicited high levels of plaque neutralizing antibodies and caused only minimal local reactions at the injection site. Significant variability was observed in the geometric mean titre evoked by various vaccine lots. This variability had not been predicted by conventional pre-filtration or pre-inactivation virus infectivity assays, or the results of animal potency tests. These findings emphasize the need for statistically valid human potency testing and the development of accurate predictive preclinical measurements for this and other vaccines.

Differences in susceptibility of inbred mice to Bacillus anthracis.

Animal species differ in their resistance both to infection by Bacillus anthracis and to anthrax toxin. A mouse model was developed to study the basis of the host differences and the pathogenesis of infection. When mice were infected with the virulent B. anthracis strain Vollum 1B, low 50% lethal dose (LD50) values (5 to 30 spores) were found for all 10 strains of inbred mice tested. However, analysis of time-to-death data revealed significant differences among the strains, which could be divided into three groups: most susceptible (A/J and DBA/2J); least susceptible (CBA/J, BALB/cJ, and C57BR/cdJ); and intermediate (the remaining five strains). In contrast, the mice were distinctly susceptible or resistant to lethal infection by the toxigenic, nonencapsulated Sterne vaccine strain. The LD50 for the susceptible A/J and DBA/2J mice was approximately 10(3) spores of the Sterne strain, whereas the remaining eight relatively resistant strains were killed only by 10(6) or more spores. F1 hybrid and backcross studies suggested that resistance to the Sterne strain is determined by a single dominant gene or gene complex. Mice lethally infected with B. anthracis showed an acute course of infection, characterized by extensive gelatinous edema and large concentrations of bacilli in the blood and organs (e.g., 10(9) CFU/g of spleen). The susceptibility of A/J and CBA/J mice to intravenously injected anthrax toxin components appeared to differ from their susceptibility to infection. The toxin LD50 values for both strains were similar. However, CBA/J mice died sooner than did A/J mice, with mean time to death of 0.9 and 3.7 days, respectively, in mice given 4 LD50 of toxin. The mouse model appears to be useful in studies on host resistance to anthrax and on the pathogenesis of the infection.