Staff Writer

Between the negative Phase III news on Oxford BioMedica plc's TroVax and the positive Phase III news on Biovest International Inc.'s BiovaxID, it's been a big month for cancer vaccines.

Oxford saw its stock tumble 60 percent after an interim analysis revealed that the TroVax trial was unlikely to meet its endpoints - adding one more to the long and painful list of setbacks in the cancer vaccine field. And while Biovest may turn out to have the first statistically significant survival data in the intent-to-treat population in a randomized, controlled cancer vaccine trial, the results released this month were plagued by statistical questions that left many investors skeptical.

But despite all the failures and amid all the skepticism, 10 cancer vaccines are in Phase III trials, and some are approaching the finish line. Canaccord Adams Inc. analyst Joseph Pantginis predicted the field will be "defined or broken" by events playing out over the next 12 to 18 months.

Hard Lessons Learned

Investors tend to lump cancer vaccines into a single dubious-at-best category, but there are a lot of different approaches, Pantginis told BioWorld Financial Watch.

Perhaps the most important variable is vaccine design. Is it autologous (personalized) or allogeneic (off-the-shelf)? Does it target a single antigen or multiple antigens? Which antigens? And what about other approaches like whole cell, dendritic cell, DNA or idiotype?

There are arguments for and against almost every one of those strategies. Pantginis favors allogeneic products because of the "significantly lower" cost of goods. Yet David Miller, CEO of independent research firm Biotech Stock Research LLC, argued that the manufacturing process for some personalized products, such as Dendreon Corp.'s Provenge (sipuleucel-T), is "not as complicated as people think."

In terms of antigen selection, Rodman & Renshaw analyst Ren Benjamin wrote in a recent report that "many in the field believe that exposure to a broad array of tumor antigens is important to elicit a long-lasting, augmented, immune response."

Biovest used an argument along those lines to explain why it believes BiovaxID will succeed in non-Hodgkin's lymphoma where Favrille Inc.'s Specifid and Genitope Corp.'s MyVax both failed in Phase III. Although all three products are idiotype vaccines, BiovaxID is hybridoma-produced rather than recombinant, which Biovest said allows production of the entire idiotype rather than just a segment, potentially resulting in a better immune response.

Beyond vaccine design, adjuvant choice also can factor into the success or failure of a program. Pantginis said that even though CancerVax Corp.'s melanoma vaccine Canvaxin used a whole-cell design, the choice of BCG as an adjuvant led to a "suboptimal" product and contributed to its Phase III failure.

Then there's the question of which tumors to target. Some companies "might actually have a vaccine that is well-designed scientifically but pick a cancer population that is not alive long enough to show the immune response," Pantginis said. Targeting aggressive tumors like melanoma might have contributed to the Phase III failures of Antigenics Inc.'s Oncophage and Progenics Pharmaceuticals Inc.'s GMK. Ditto for the Phase III failure of Therion Biologics Corp.'s Panvac-VF in pancreatic cancer.

And even if a company manages to design a promising product targeting a potentially receptive patient population, a poorly designed trial could spoil the chances for success. Miller said many cancer vaccine failures have been caused by trial design issues, pointing to Provenge as an example. Time to progression might have been a valid endpoint when Dendreon originally designed its Phase III prostate cancer trials for Provenge, but the emphasis has since shifted to survival, he said.

The administration schedule for the vaccine also can be a thorny issue, as Oxford found when its latest interim analysis indicated that fewer doses actually may be needed to achieve optimal efficacy.

Deciding whether or not to give the drug in combination with chemotherapy, which chemotherapy to pair with and how to manage the administration schedule also can affect a trial's outcome. Although many cancer vaccines are given after chemotherapy, Benjamin noted that "chemotherapeutic regimens are often immunosuppressive and may decrease or delay the immunotherapeutic efficacy."

Into the Backstretch

The good news is that today's cancer vaccine companies seem to have learned from the mistakes of the past and "have designed better mouse traps," Benjamin said, adding that he expects to see the first therapeutic cancer vaccine achieve FDA approval possibly as early as 2009.

Thanks to its recent Phase III data, Biovest's BiovaxID is in the running. The drug met the primary endpoint of significantly improving disease-free survival in the multicenter, randomized, controlled Phase III trial in indolent follicular non-Hodgkin's lymphoma (NHL). Patients receiving the vaccine had a median progression-free survival of 33.8 months compared to 21.2 months in the control arm (p = 0.047).

Yet Miller cautioned that while a "p" value of less than 0.05 normally would be sufficient, the "p" value specified in the trial's statistical analysis plan was 0.01. Add to that the fact that the analysis wasn't quite final and started with treatment rather than randomization, and Miller said BiovaxID is "not likely to get approved," although falling victim to trial design issues "doesn't mean the drug doesn't work."

In Miller's opinion, Dendreon's autologous dendritic cell vaccine Provenge "clearly has a shot to be first across the finish line." Interim data from an ongoing, confirmatory Phase III prostate cancer trial are expected in October, and positive findings could pave the way to approval. Otherwise, investors will have to wait for the final analysis, likely in the second half of 2009.

Also expected in the second half of 2009 are data from Cell Genesys Inc.'s ongoing Phase III trial with prostate cancer vaccine GVAX. Pantginis noted that GVAX's allogeneic, whole-cell design has generated better median survival data than Provenge, but Miller questioned the Phase II trials supporting the drug.

Behind those front runners, several other cancer vaccines are coming down the pipeline. AVAX Technologies Inc. expects data in 2010 from its Phase III melanoma trial with M-Vax, while Bioniche Life Sciences Inc. is enrolling a Phase III bladder cancer trial with Urocidin. Oncothyreon Inc. (formerly Biomira Inc.) is enrolling patients in a Phase III non-small-cell lung cancer (NSCLC) trial with Stimuvax, and GlaxoSmithKline plc is in Phase III for NSCLC with its MAGE-A3 ASCI. Vical Inc. is enrolling a Phase III melanoma trial with Allovectin-7, while Receptor BioLogix Inc. is in Phase III for pancreatic cancer with Insegia and The Vaccine Co.'s Proteinase 3 PR1 peptide vaccine is in Phase III in acute myeloid leukemia.

To Profitability and Beyond

In addition to the 10 cancer vaccines in Phase III, Rodman & Renshaw reports that there are more than 50 other companies with cancer vaccines in development. With so many shots on goal, it's likely that some will eventually gain approval.

But will they make money?

While expensive to make (and most likely expensive to take, as well), Benjamin predicts that cancer vaccines will "represent a paradigm shift in the treatment and chronic management of cancer" thanks to their potential to increase survival with minimal side effects. Although most are being tested in late-stage cancers, Benjamin said their use after approval will shift to earlier-stage patients with healthier immune systems, providing "multibillion dollar opportunities."

Miller also thinks cancer vaccines will find a multi-billion-dollar market, as physicians likely will use them as a first-line option before moving on to chemotherapy, given their relatively benign side-effect profile.

Pantginis agreed that the market potential is significant, drawing a comparison to monoclonal antibodies, another class of biologics that are expensive to make but have proven extremely profitable.