Staff Writer

Shares of GenVec Inc. fell 43.5 percent Wednesday after the company released what appeared to be fairly positive interim data from its pivotal Phase III study of gene therapy TNFerade for pancreatic cancer.

In an analysis triggered by 92 deaths - one-third of the total expected in the trial - patients receiving TNFerade plus standard of care had a 25 percent lower risk of death than those receiving the standard of care alone, based on a hazard ratio of 0.753 and a 95 percent confidence interval of 0.494 to 1.15.

While the upper limit of the confidence interval would need to fall below 1.0 to achieve statistical significance, GenVec's senior vice president of product development, Mark Thornton, pointed to positive early trends indicating that the 1.0 target is "where we are headed." Median survival was 9.9 months in both arms.

"You may be wondering why, with the 25 percent lower risk of death, median survival . . . was the same in both groups," GenVec's President and CEO Paul Fischer said during a conference call with investors. "It is because the apparent benefit seen in the TNFerade arm occurred after that point."

According to a Kaplan-Meier analysis, overall survival at 12 months was 39.9 percent for the TNFerade arm vs. 22.5 percent for the control arm, a benefit occurring after the 9.9-month median mark had been passed. The two arms further diverged by 18 months, with overall survival of 30.5 percent for the TNFerade arm vs. 11.3 percent for the control arm.

By 24 months, the two arms converged, with overall survival of 10.6 percent for the TNFerade arm vs. 11.3 percent for the control arm.

Fischer said the benefit around the 12-month mark and "flattening of the curve" after 18 months is consistent with what GenVec has seen in previous studies. Thornton added that the 24-month data are the least reliable part of the analysis, given the relatively small number of patients to reach that point before the interim look.

Debjit Chattopadhyay, analyst with Boenning & Scattergood Inc., wasn't concerned about the long-term convergence of the TNFerade and control arms. "Pancreatic cancer is a death sentence - all of the patients are eventually going to die," he told BioWorld Today. While long-term separation between the two arms would indicate that patients were being cured, the primary endpoint of the trial is not a cure but overall survival across the duration of the trial as measured by the hazard ratio, he said.

As a point of reference, he pointed to data supporting the approval of Tarceva (erlotinib, Genentech Inc. and OSI Pharmaceuticals Inc.) in pancreatic cancer. Median survival was 6.4 months for patients receiving Tarceva plus Gemzar (gemcitabine, Eli Lilly and Co.) vs. six months for Gemzar alone. Overall survival at 12 months was 23.8 percent for Tarceva vs. 19.4 percent for Gemzar. The hazard ratio was 0.81 with a 95 percent confidence interval of 0.68 to 0.97.

Chattopadhyay said he is "convinced" the interim TNFerade data are good.

Analyst Ren Benjamin, of Rodman & Renshaw agreed that the data look "promising" but said it is "still too early to make a definitive conclusion."

Yet shares of Gaithersburg, Md.-based GenVec (NASDAQ:GNVC) dropped to new 52-week lows on Wednesday, shedding 50 cents to close at 65 cents. Both Chattopadhyay and Benjamin said concern over the company's cash position may have contributed to the sell-off.

GenVec reported $22.5 million in cash and investments at the end of the third quarter, which it said will last into third-quarter 2009. That means the company will have to either raise money - no easy task in this market - or find a partner for TNFerade before then.

Fischer said securing a partner is a "high priority" for GenVec and indicated that the company is optimistic about showing the interim data to prospective partners.

But in the absence of a partnership, GenVec's next major milestone for TNFerade isn't expected until late next year, when the company conducts the second interim analysis after 184 deaths occur in the ongoing trial. About 300 of the 330 patients in the trial are expected to be enrolled by the end of next year.

"The data look promising, but it's not getting approved tomorrow," Chattopadhyay said. And since hedge funds are struggling to deliver returns every quarter, there is no need for them to hold the stock now in the absence of near-term, stock-moving milestones. He predicted that some investors might have been hoping TNFerade would get approval based on the interim analysis, and they likely jumped ship when that didn't happen.

The independent data safety monitoring board recommended that the trial continue as planned.

TNFerade is an adenovector carrying the gene for tumor necrosis factor (TNF)-alpha and is also in trials for rectal cancer, head and neck cancer and metastatic melanoma. GenVec also has programs for macular degeneration and hearing loss as well as government-funded vaccine programs targeting malaria, respiratory syncytial virus, foot-and-mouth disease, HIV and other diseases.