Preclinical studies and human trial
Preclinical efficacy studies
The efficacy of rBCG30 was evaluated in the guinea pig model of pulmonary tuberculosis, a model noteworthy for its relevance to human tuberculosis. Unlike the mouse and rat, but like the human, guinea pigs are susceptible to low doses of M. tuberculosis, develop a strong cutaneous delayed-type hypersensitivity (DTH) response to tuberculin that is characterized by a dense mononuclear cell infiltrate, and display Langhans giant cells and caseation necrosis in tuberculous lesions.
In immunogenicity studies, guinea pigs were sham-immunized or immunized with parental BCG or rBCG30 (1). Nine weeks later, the animals were tested for cutaneous DTH to the M. tuberculosis 30 kDa protein (r30). Sham-immunized animals and BCG-immunized animals displayed baseline cutaneous DTH responses to r30. In contrast, rBCG30-immunized animals had strong cutaneous DTH responses that were significantly greater than those in the sham-immunized and BCG-immunized animals.
In the first type of efficacy study, guinea pigs were sham-immunized, immunized with BCG, or immunized with rBCG30, and 10 weeks later, the animals were challenged with an aerosol dose of the highly virulent Erdman strain of M. tuberculosis (1). Ten weeks after challenge, the animals were euthanized and the bacterial burden in the lung and spleen assayed. The sham-immunized animals had the highest bacterial burden in their lungs and spleens. BCG-immunized animals had a marked reduction of ~1½ - 2 logs in the bacterial burden in their lungs and spleens. Most importantly, compared with BCG-immunized animals, rBCG30-immunized animals had a further reduction in bacterial burden averaging 0.5 logs in the lung and 1 log in the spleen, differences which were statistically significant and highly so. Statistically significant differences of comparable magnitude between BCG and rBCG30-immunized animals in colony-forming units (CFU) of M. tuberculosis in the lung and spleen after aerosol challenge have now been demonstrated in eight consecutive experiments involving large groups of animals.
At necropsy, rBCG30-immunized animals also had less organ pathology than BCG-immunized animals (1). Compared with BCG-immunized animals, rBCG30-immunized animals had ~10-fold fewer tubercles in their lungs, spleens, and livers.
In a second type of efficacy study, guinea pigs were immunized as described above and then monitored for survival (5). Sham-immunized animals died most rapidly. BCG-immunized animals survived significantly longer than sham-immunized animals. Most importantly, rBCG30-immunized animals survived significantly longer than BCG-immunized animals, and 40% of this group survived to the point at which uninfected control animals began to die.
Preclinical safety studies
Several types of safety studies were performed in preparation for human trials (5). First, rBCG30 and BCG were found to be equally avirulent in guinea pigs. Animals immunized with these two vaccines gained weight normally and displayed no evidence of illness. Second, rBCG30 was found to disseminate and multiply in guinea pig tissues and subsequently be cleared from those tissues at the same rate as BCG. Third, rBCG30 and BCG vaccines were demonstrated to be sensitive to the same antimycobacterial antibiotics, although rBCG30 was found to be slightly less sensitive to INH. Fourth, the pMTB30 plasmid was demonstrated not to be mobilizable or self-transmissible to other bacteria. Fifth, the pMTB30 plasmid was demonstrated not to be expressed in E. coli even when forced into it by electroporation. Finally, the rBCG30 vaccine was documented to have had no exposure to animal products from countries where BSE has occurred.
Phase 1 human trial of rBCG
In February 17, 2004, a Phase 1 human trial of rBCG30 was initiated. The study was designed to assess the safety of rBCG30 in tuberculin-negative adult volunteers, and to test the capacity of the vaccine to induce cell-mediated and humoral immune responses to the M. tuberculosis 30 kDa protein.