National Editor

Genentech Inc.'s optimistic earnings forecast March 14 sent its stock on an upward journey that was cut short last week, apparently by an analyst's report casting doubt on Avastin, in a Phase III trial against colorectal cancer.

The company's stock (NYSE:DNA) had reached almost $36.50 on the earnings guidance, and closed at $35.15 last Thursday. That was the day analyst Mark Augustine, with Credit Suisse First Boston in New York, downgraded Genentech from "neutral" to "underperform," based on data, review of literature and "conversations with key clinicians and researchers," he wrote.

Analyst Winton Gibbons, with William Blair & Co. in Chicago, followed with a research report Thursday noting the stock's weakness as a result of Augustine's report and recommending Genentech as a buy.

"We reiterated our outperform' rating," which was made in an upgrade at the beginning of the week, Gibbons told BioWorld Today.

"In the end, [Genentech's stock slide] wasn't very much, only about 3 percent, but they were down about 6 percent, trading around $33," he said. "It's worth talking about."

Gibbons said he doesn't read competitors' research but had heard from an associate about the Credit Suisse First Boston note by Augustine, who could not be reached Friday.

"Ironically, in this case I probably agree with him [about Avastin]," he said. "Talking to people we know in angiogenesis, we believe it works but it's best used in very early stage disease. In most cancer trials, it's hard to get informed consent for early stage because you're not the standard of care."

Data from the colorectal cancer trial are expected mid-year this year. A Phase III study in metastatic renal cell carcinoma is ongoing, and this is the indication in which Gibbons holds the most hope for Avastin (bevacizumab).

The drug, an antibody directed at vascular endothelial growth factor, missed its primary efficacy endpoint of progression-free survival in relapsed metastatic breast cancer in September, but a secondary endpoint - overall response rate - achieved statistical significance in the 462-patient trial. That secondary endpoint still didn't translate into benefit with regard to progression-free survival or 12-month survival, however. (See BioWorld Today, Sept. 11, 2002.)

"We're bearish on Avastin, but that's not the point," Gibbons added, pointing to the company's remaining pipeline. "If both Avastin and Tarceva fail, I'm not going to say, Who cares?' because that's too flip, but they've still got 12 [more potential products]."

Among these are Xolair (omalizumab), for asthma, and Raptiva (efalizumab), for psoriasis, both awaiting FDA approval.

"Obviously, clinical trials and regulatory approvals are issues, but we're not talking about a company that lives or dies on one product," Gibbons said. "If they did, it would be Rituxan."

Rituxan (rituximab), a monoclonal antibody, was approved in 1997, initially for relapsed or refractory, low-grade or follicular, CD20-positive B-cell non-Hodgkin's lymphoma.

South San Francisco-based Genentech, Gibbons said, has "about a 50 percent better hit rate in the clinic than the drug industry on average in total. You've got a classic pharmaceutical approach, where they've developed, for the most part, antibodies with mechanisms that allow you to keep expanding your indications" - an approach that investors and regulators find appealing, since "the devil you know is better than the devil you don't."

The company had Avastin-related news that might have been taken as negative several days before its upbeat earnings forecast - a letter to doctors warning about bowel perforations in some Avastin trial patients. That didn't bother Gibbons, either.

"It seemed like old news to us," Gibbons said. "These people have late-stage colorectal cancer, so it was probably caused by the disease and may be caused by the chemotherapy. The last thing you would suspect is the antibody."

Genentech's earnings forecast included a prediction that the company would beat analysts' estimates and repeated the January guidance calling for at least 20 percent earnings-per-share growth in 2003 and 2004, an outlook that's even better than it seems, Gibbons said. (See BioWorld Today, Jan. 17, 2003.)

"When they say 20 percent, it isn't even what you would call a best estimate," he said. "The guidance they use accounts for failures."

Genentech's stock rose 12 cents Friday to close at $35.27.