Recombinant BCG vaccines for the prevention of cancer

Abstract

The anti-tuberculosis vaccine bacille Calmette-Guerin (BCG) has been administered to more than 3 billion people worldwide. Aside from its use in tuberculosis and leprosy control, BCG has been well studied for its stimulation of immune responses against tumors. While studies of BCG as an anti-tumor agent have yielded variable results, BCG has proved a reliable modern treatment for superficial bladder cancer.;Concurrent with the first identification of a human tumor antigen, introduction of foreign genetic material into mycobacteria, such as BCG, became possible. These discoveries occurred only after research activity on the use of BCG for cancer had waned. We investigated whether recombinant expression of tumor antigens could enhance the anti-tumor activity of BCG. While most recent work on cancer vaccines has focused on immunotherapy rather than secondary prevention, we studied BCG prevention of tumorigenesis and tumor growth for reasons discussed in the text.;We compared BCG and recombinant BCG in several mouse models of cancer. (1) These strains, were observed for effects on tumor incidence and latency in transgenic mouse model of spontaneous melanoma. (2) In a transplantable melanoma model, expression of the recombinant fusion product specifically inhibited BCG growth in vivo, and multiple doses of this attenuated BCG recombinant did not stimulate significant immune responses against the tumor antigen. (3) In a transplantable tumor system using a model tumor-associated antigen, the effects of BCG immunization on tumor growth depended on time of BCG inoculation and tumor challenge. Immune responses to vaccine and tumor in this model were characterized. At no time in this study did expression of the tumor-associated antigen exert a significant effect on the anti-tumor activity of BCG.;This work has identified factors limiting the use of recombinant BCG vaccines for the prevention of cancer: (1) unstable expression of antigen by presently available multicopy plasmids, (2) insufficient antigen expression by stable, single-copy, integrating plasmids, (3) weak antigenicity of wild-type tumor antigens, (4) importance of timing of immunization on anti-tumor activity, and (5) compartmentalization of antigen expression within the bacteria. Based on these findings, we suggest future directions for the study of recombinant BCG and cancer.