By David N. Leff

To cure a morning-after hangover, sip a little of the same alcoholic beverage that made you drunk the night before.

Clinical trials now beginning in the U.S. and Europe seem to be applying this hair-of-the-dog-that-bit-you formula to the treatment of cancer.

The first such Phase I study has just ended in Germany. It enrolled 14 patients, research immunologist Primod Srivastava told BioWorld Today, "most of them with cancers of the gastrointestinal tract — stomach, colon and so on — but also some other tumors. They were an assorted bunch."

The participants were all vaccinated with proteins purified from their own tumors.

"We took tumor tissue from each patient," Srivastava recounted, "either by biopsy or surgical excision, ground it up, and purified the resulting homogenate by column chromatography to extract the heat shock proteins."

Srivastava, who directs the Center for Immunotherapy of Cancer and Infectious Diseases at the University of Connecticut, in Farmington, explained:

"We made customized vaccine for each patient, and determined that it produced no major toxicity or danger of generating autoimmunity." Although a Phase I trial doesn't gauge therapeutic efficacy, he pointed out, "we showed that more than half of the 14 patients immunized generated very clear and demonstrable T cell responses against their tumors after being vaccinated, while before vaccination they had no response."

Heat shock proteins (HSPs), the active agents of Srivastava's vaccine, are misnomers. To be sure, these busy-body molecules multiply their intracellular populations many times over when stressed by rising temperature. But many other stimuli also expand their number, so they can handle various jobs.

For one thing, HSPs act as molecular chaperones, helping new proteins fold into their proper three-dimensional shapes. For another, they sop up the minced remnants of peptides floating around in the cell as leftovers from the degradation of old, redundant proteins.

For over a decade, Srivastava has been "pursuing the phenomenon that you can vaccinate against tumors by using the whole tumor cell." He irradiated such malignant tissue to destroy its ability to form new tumors, and injected these neutered cells into mice and rats. They proved to be immune to a subsequent challenge with live tumors.

"So I asked myself the question," he recalled: "If the whole tumor cell can confer immunity, must there not be in it some component that I can purify, reflecting the immunizing ability of the whole cell?"

He began to fractionate tumors into various protein components, and vaccinated animals with these individual aliquots. "Over the course of years," he continued, "I identified three different molecules, each of which could immunize against tumors. They turned out to be heat shock proteins."

There are on the order of half a hundred HSP genes scattered all over the human genome, Srivastava observed, with no two alike. "They express proteins," he said, "that fall into several families defined largely by size, say from 10,000 Daltons on up to 110,000 and so on."

Animal Experiments Justify Human Trials

In today's Science, Srivastava reports his latest animal experiments, which validate the move into early human studies. His paper bears the title: "Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations." He and his co-authors seeded the footpads of mice with a highly aggressive murine carcinoma, which spreads rapidly to the lungs. From each individual animal's tumor, the team prepared and administered an HSP vaccine. A few weeks later, they counted an average of 18 metastases in the lungs of vaccinated mice. Unvaccinated control animals had 200.

A separate experiment sought to ascertain the effect of the vaccine on minimal residual cells after ablation of the primary tumor — a major concern in human oncology. Excision of tumor-bearing footpads left the rodents apparently disease-free. Yet 45 days after surgery, the entire unvaccinated cohort was dead of lung metastases.

But 80 percent of another group, vaccinated four times at weekly intervals following tumor removal, survived free of disease for 36 weeks or more.

Other in vivo experiments tested the efficacy of the HSP vaccine in a broad range of human cancers, including melanoma, spindle-cell and colon carcinoma and fibrosarcoma. These demonstrated prolonged survival and slower tumor growth in animals vaccinated with their own autologous HSPs, but not with vaccines from tumor cells other than their own.

Vaccine Hits Tumor Antigen Bull's Eye, Sight Unseen

Srivastava explicated the occult connection between heat shock proteins and the immune system:

"We have supposed for a long time that tumors carry a large repertoire of mutations, some of which are important for a normal cell to become a tumor. And protein gene mutations happen randomly throughout the genome. Thus, they differ from tumor to tumor, imparting to each an antigenic individuality. And because heat shock proteins have the ability to sop up peptides, when we isolate these we are able to isolate along with them the entire antigenic repertoire of their tumor — without having to identify the mutations in each tumor antigen."

His findings in Science emphasize that besides activating cytotoxic T cells to attack tumor cells, "there is a very important role for non-specific natural killer cells, which are equally important."

Srivastava is scientific co-founder of Antigenics L.L.C., a three-year-old, privately held company based in New York City. He chairs the firm's scientific advisory board, and sits on its board of directors.

Antigenics holds exclusive worldwide rights to ten or more of Srivastava's patented inventions, of which the latest, he mentioned, "was awarded in the U.S. about two or three weeks ago."

Following the human trial in Germany, he said, "Right now we are in a Phase I trial at Memorial Sloan-Kettering Cancer Center, in New York. It began in June, and is enrolling 10 patients with pancreatic cancer. It should be done in eight or 10 months."

He expects two more such clinical studies, of melanoma and renal cell carcinoma, respectively, to begin next month at M. D. Anderson Cancer Center, in Houston. "There," Srivastava observed, "we'll be testing against different doses, routes of immunization and immunological effects. Essentially," he concluded, "these studies will set the stage for us to do more significant human trials in viral infections as well as cancers." *