LOS ANGELES - Cancer vaccine companies got a lift recently with the advisory board recommendation to approve Dendreon's Provenge. (See BioWorld Today, March 29, 2007, March 30, 2007, and April 2, 2007.)

But not everyone on the advisory committee was convinced that Provenge has truly shown efficacy. And similarly, not everyone in the research world is convinced that the cancer immunotherapy approach has proven its value.

"One cannot conclude that cancer vaccines will not work. Nor can one conclude, in 2007, that they have worked in any sort of reliable fashion," James Yang of the National Cancer Institute told the audience at a Tuesday symposium titled "Cancer Vaccine: Fantasy or Reality?"

Yang, the designated vaccine skeptic on the panel, said that personally, he would have chosen the title "Cancer vaccines - spin or substance?" because, he said, cancer vaccines are effectively a reality, owing to the sheer amount of both academic and industrial effort being expended on the approach.

Just perhaps not a useful reality. Yang showed some sobering numbers to back up his skepticism of the current state of cancer vaccine technology. The response rate to immunotherapy is around 2 percent, which is low even by oncology standards. Additionally, many cured patients have subcutaneous or lymph node tumors, which rarely kill a patient in the first place "The number is extremely small, and the problem being addressed is not always the critical one."

That's not to say, Yang said, that an immunotherapeutic approach is an exercise in futility. Quite the contrary, he said, noting that he has decades of cancer vaccine research under his belt and is "married to the concept - I just have a very tough prenup."

What it will take for cancer vaccines to be successful, he said, is threefold. The first is better epitopes. "I think we gave up on antigen discovery way too early," he said, asserting that several epitopes now in clinical trials lack underlying data that show they actually are presented by tumors. On a related note, even those epitopes that are bona fide tumor epitopes are not sufficiently immunogenic, and vaccination with those epitopes is unable to marshal sufficient numbers of T cells to actually deal with typical tumor volumes.

Yang's skeptical assessment, in fact, is that this is why immunotherapy currently does not have toxicity problems. By combining immunotherapy with various ways to combat tolerance or anergy - the third issue he sees for immunotherapy - "you can get a brisker immune response. But then you will also have toxicity," he predicted.

As for Dendreon's success, which is based on a survival benefit that, while robust, was not a primary or secondary endpoint of the study, Yang expressed his reservations succinctly. "If you're allowed to change the endpoint, it's like shooting your arrow, walking up to it and drawing the bulls' eye around it," he said. "That's not fair."

At a Tuesday session on "Immunotherapy and Cancer Vaccines," there was, of course, greater optimism about the cancer vaccine approach. Among other studies, two Phase I studies reported promising results when autologous dendritic cells (the same approach that is used by Dendreon with Provenge) were used to stimulate an immune response to MUC-1 for the treatment of pancreatic cancer and p53 for the treatment of head and neck cancer, respectively.

MUC-1 is a tumor protein expressed in about 90 percent of pancreatic cancer patients, making a vaccine based on it useful for most patients, first author Andrew Lepisto of the University of Pittsburgh said. P53 is, of course, universally expressed, and a tumor suppressor at that. Senior author Theresa Whiteside, also of the University of Pittsburgh but in a different laboratory, told reporters at a press conference that "in cancer, p53 is altered and accumulates in tumor cells."

While the head and neck cancer trial is ongoing, the MUC-1 trial has been completed. Lepisto said that despite the fact that it mainly focused on safety, as Phase I trials are wont to do, the survival rate to date has been more than twice what would be expected from historical controls.

Both Lepisto and Whiteside echoed Yang's assertion that it will be necessary to tame inhibitory components of the immune system, specifically regulatory T cells, to have a successful cancer vaccine. Whiteside said that since tumors are known to suppress the immune system, "when we are talking about therapeutic vaccines, we must think not only about how to activate an immune response, but also how to maintain this activation." Lepisto currently is studying blood and tissue samples of the surviving patients in the MUC-1 trial to see whether he can find out what allowed them to mount an immune response that led to long-term success.

The conference ended Wednesday.