West Coast Editor

Genentech Inc.'s third-quarter earnings win with the colorectal cancer drug Avastin helped the firm narrowly beat consensus, though analysts remain watchful of the off-label tussle for patient share with Lucentis, the same firm's compound for age-related macular degeneration.

Shares of Genentech (NYSE:DNA) closed Tuesday at $75.04, down $2.46.

Non-GAAP net income rose 22 percent over the same period last year to $778 million, or 73 cents per share, sliding past the forecast by Thomson Financial analysts, who expected 72 cents per share and revenue of $2.93 billion.

Genentech reported about $2.91 billion in revenue, not counting $3 million in deferred royalties from the buyout of Tanox Inc., disclosed late last year. (See BioWorld Today, Nov. 13, 2006.)

Total product sales of $2.32 billion fell below Lazard Capital Markets' $2.43 billion estimate, but royalties of $521 million handily beat the forecast of $424 million, offsetting other results.

Avastin (bevacizumab) sales for the first time surged past those of Rituxan (rituximab, which targets the CD20 antigen, for non-Hodgkin's lymphoma and rheumatoid arthritis), despite patients switching to lower-dose Avastin. But Phase III trials could yield more value for Rituxan, which is being tested against primary progressive multiple sclerosis (PPMS) and lupus.

In the latter, Susan Desmond-Hellman, president of product development for South San Francisco-based Genentech, pointed to a "growing belief among the thought leaders that we may have an important new medicine. The caution would need to be that this is a very heterogeneous patient population," she added, with "a long history of disappointments. We certainly don't want to overreach until we see the data, despite our optimism for our approach."

The ongoing Phase III trial known as LUNAR, testing Rituxan in lupus nephritis, is expected to finish enrolling in the fourth quarter of this year, and Genentech likely will start a Phase III trial with a second-generation anti-CD20 drug for systemic lupus erythematosus around the same time.

Rituxan is partnered with Cambridge, Mass.-based Biogen Idec Inc., which recently said it would entertain offers from buyers.

"We can make a bid on the outstanding rights, or the rights we don't have, on Rituxan and related molecules," if Biogen is taken over, said David Ebersman, chief financial officer for Genentech.

Meanwhile, Genentech remains in arbitration with Biogen over the rights to second-generation CD20s, a process that has a "reasonable probability" of finishing by the end of next year.

The dispute is over "decision-making rights, and whether or not we're allowed to proceed with certain studies if Biogen Idec doesn't think we should," he said.

In PPMS - another potential Rituxan use - no approved therapies exist. "There are some who question whether the biology of PPMS is similar to that in [relapsing-remitting MS]," Desmond-Hellman said. "We think this is a more risky study area."

Though Avastin performed well during the quarter, the breast cancer therapy Herceptin (trastuzumab) as well as Lucentis (ranibizumab) fell short of hopes.

Specifically, Avastin sold $597 million, surpassing consensus expectations of $586 million and well ahead of the $435 million in sales during the same period a year ago. Rituxan sold $572 million vs. the consensus of $588 million, compared to $509 million a year ago, and Lucentis hit $198 million, having sold $153 million in the same period last year. Herceptin sold $320 million, compared to $302 million a year ago, missing the consensus number of $339 million.

Genentech said last week it would no longer sell Avastin to compounding pharmacies, which repackage the compound for off-label use in AMD - a move that the firm said was made for the sake of safety. Lazard analyst Joel Sendek wrote in a research report on earnings that the AMD market likely has become evenly split between Avastin and Lucentis, which will be tested in a head-to-head trial sponsored by the National Institutes of Health, starting by the end of this year.

The Avastin threat aside, Lucentis will be tried in a Phase III program against a potentially larger market than AMD: diabetic macular edema (DME), and data are expected in the third quarter of next year from ongoing trials in retina vein occlusion (RVO). It is possible that cost-conscious regulators and curious clinicians could test Avastin in DME and RVO as well, Sendek wrote.

Among the near-term news items due from Genentech is word from an FDA advisory panel in December on Avastin for metastatic HER2-negative breast cancer, taking up the issue of whether progression-free survival (PFS, in which Avastin plus paclitaxel proved strong in a trial known as ECOG 2100) is acceptable as an endpoint to win approval. Overall survival is the customary endpoint in such trials.

PFS "should be an approvable endpoint because we think the way Avastin performed in this trial was consistent with clinical benefit," Desmond-Hellman said, noting "a long history in first-line metastatic breast cancer of approvals based on PFS."

The PDUFA date for Avastin in breast cancer is Feb. 24, 2008.

Genentech's earnings showed that sales of the lung cancer drug Tarceva (erlotinib), partnered with Melville, N.Y.-based OSI Pharmaceuticals Inc., remained discouraging for the second quarter in a row, totaling $101 million - well short of the $111 million consensus - compared to $100 million in 2006.

Overseas, Tarceva is still doing well. Basel, Switzerland-based F. Hoffmann-La Roche Ltd., OSI's partner outside the U.S., Tuesday reported stronger than expected Tarceva sales, tallying $126.9 million. Andrew Fein, analyst at Collins Steward LLC, had estimated $116 million.

Roche cited differences in physician reimbursement practices in explaining the better results.

"While we are encouraged by continued growth ex-U.S., we wonder if the clinical challenges that Tarceva faces in the U.S. will eventually manifest themselves abroad," Fein wrote in a research report.