By David N. Leff

Oncologists are recruiting a soil bacterium that feeds on the flesh of dead animals to devour solid tumors in humans instead. The germ in question, Clostridium novyi, is best known for causing gas gangrene in fingers, toes and deep wounds, including the necrotic liver ? putrifying for lack of an oxygen-carrying blood supply.

C. novyi itself couldn?t care less. As an anaerobic microorganism, it thrives in oxygen-free (anoxic) environments, such as occur in large solid tumors.

?The general idea is very old,? observed research and clinical oncologist Bert Vogelstein at The Johns Hopkins School of Medicine in Baltimore. ?It?s been noted by pathologists for more than a century that solid tumors generally have relatively large regions of poor vascularization. In the 1940s, 50s and 60s, there were a few approaches to try and use bacteria that could grow in anoxic environments to target those regions of tumors. I was always intrigued by those early studies,? Vogelstein continued. ?They were eventually abandoned because of the toxicities associated with these anaerobic bacteria, or because the bacteria were not very effective. But the idea was intriguing in principle, so we decided to revisit it, using technologies now available for engineering bacteria that didn?t exist back then.?

Vogelstein is senior author of a paper in the Proceedings of the National Academy of Sciences (PNAS), released online Nov. 26, 2001. Its title: ?Combination bacteriolytic therapy for the treatment of experimental tumors.?

?Our overall purpose,? Vogelstein told BioWorld Today, ?was to design a way to target the poorly vascularized pockets in tumors, which are the most resistant to conventional anticancer chemotherapy and radiation. It?s one of the reasons that conventional therapies often fail.?

Therapeutics Need Oxygen ? From Where?

When advanced neoplastic cells are growing out of control, some areas of malignant tissue expand so rapidly that they outgrow angiogenesis. Their oxygen-freighted blood supply network never has the time to catch up. These anoxic areas short-change conventional cancer therapy, since ionizing radiation needs oxygen in order to kill cells, and chemotherapy drugs need blood vessels along which to travel to their tumor target.

?Our goal,? Vogelstein explained, ?was to target these malignant, blood-starved regions with a new anaerobic bacterium, and then combine those bacteria with conventional agents in order to more effectively treat transplanted solid tumors. We began by systematically screening for bacteria that had particular properties,? he recounted. ?Not only should they localize to poorly vascularized regions of tumors, but would also kill the surrounding viable tumor cells. We didn?t know whether we could find any bacteria that behaved that way, but in fact we were very fortunate. Our screen was able, finally, to single out one promising strain of Clostridia ? called C. novyi.

?This strain differed from the 26 strains in three bacterial species that we tested,? he recalled. The team injected C. novyi spores directly into human colorectal cancer tumors, implanted and growing robustly in experimental mice. ?Within eight to 12 hours we could clearly see that they were not simply growing within the dead mass inside tumors. Large tumor cells were being destroyed.

?There was just one small problem,? Vogelstein added. ?The problem with them was that 16 to 18 hours after treatment began, all the mice died. So this obviously could not be a viable therapeutic strategy.?

The co-authors laid the animals? demise at the door of bacterial spore toxin ? similar to the lethal toxin spores of the anthrax and tetanus soil bacteria, which Clostridia resemble. ?We suspected that the cause of death was the release of potent lethal toxins from the bacteria germinating within the tumors,? Vogelstein surmised.

?We then took one of those strains,? he said, ?namely, C. novyi, and were able to get rid of its secreted toxic genes, using genomic knowledge of Clostridia that had been published in the literature. Once we did that, the animals no longer died. We could inject the spores into normal mice, or those with tumors no longer secreting the toxin, but retaining their ability to kill viable tumor cells that include necrotic regions.?

So far, so good, but not quite far enough, yet.

?The spores still left a viable tumor rim of very well-vascularized cells at the margin of the lesion,? Vogelstein pointed out. ?That led to our strategy of attacking the tumors from the inside with bacteria, and from the outside with conventional chemotherapeutic agents. We concentrated on two classes ? DNA-damaging drugs, such as mitomycin and cytoxan, and dolastatin, which appears to partially collapse tumor vasculature.

?When we added these conventional chemotherapeutics to the spores in a construct we call combination bacteriolytic therapy? ? COBALT for short ? we were very excited and encouraged by the fact that in many cases the tumors simply dissolved within 24 hours into just a hemorrhagic, necrotic mass of tissue, leaving eight mice permanently tumor-free.?

COBALT?s Backlash Needs More Work

?However,? Vogelstein went on, ?the dramatic antineoplastic effects of this combination were associated with significant toxicity. Approximately 15 percent of animals with tumors of 350 cubic millimeters in size died within 24 to 72 hours of receiving COBALT; those carrying twice that size numbered 45 percent. These deaths may have been due to tumor lysis syndrome, when large tumor burdens are rapidly destroyed by antineoplastic agents.?

To see whether COBALT would affect other tumor types, the co-authors treated mice carrying melanoma tumors. Weekly COBALT boosters were required to keep their tumors from regrowing, whereas a single treatment cured about half the colorectal tumor-bearing animals.

?Clinical trials are not planned at this time,? Vogelstein observed, ?as it will take several years to determine which chemotherapy agents make the best combinations with COBALT, and to develop strategies that avoid the toxicity associated with rapid destruction of large tumor masses. We hope that this research will add a new dimension to cancer therapy, but we realize that the way tumors respond to treatment in mice can be different than in humans.?