Assistant Managing Editor

It looks like $95 per share was a pretty good deal after all for Genentech Inc. shareholders.

The South San Francisco-based firm, bought out by majority shareholder Roche Holdings AG earlier this year for $46.8 billion, reported much-anticipated adjuvant data showing that its blockbuster cancer drug Avastin (bevacizumab) failed to improve disease-free survival in early-stage colon cancer patients when added to chemotherapy.

While those disappointing data are not expected to have any effect on Avastin's use in currently approved indications in metastatic disease, analysts and investors had been looking for the C-08 trial data to serve as an early prognosticator of the drug's potential in the overall multibillion-dollar adjuvant cancer market.

In fact, many analysts believed Roche was pushing to close the deal prior to release of the C-08 data, which, if positive, likely would have boosted Genentech's asking price to well above $100 per share. Conversely, negative data likely would have sent Genentech shares falling, possibly even below the $89-per-share price offered by Roche at the start of negotiations last summer. (See BioWorld Today, July 22, 2008, and March 13, 2009.)

"Obviously, if Genentech were still independent, we would expect the stock to sell off sharply," since "hopes for this trial were high with significant read-through with respect to Avastin adjuvant programs in lung and breast cancers," analyst Christopher Raymond, of Robert W. Baird & Co., wrote in a research note.

So it might be lucky for Genentech shareholders that the deal went through when it did, though the company said it believes Avastin still could end up being an effective treatment in early-stage cancer.

Detailed data were not released - Genentech and Roche said those will be reported at the upcoming American Society of Clinical Oncology in Orlando, Fla. - but Hal Barron, senior vice president of development and chief medical officer, stated in a press release that initial reviews of the data indicated that "Avastin may be active in patients with early stage colon cancer," and other adjuvant programs in colon cancer, breast cancer and lung cancer are expected to continue.

Study C-08 was designed to test Avastin, an anti-VEGF antibody, in combination with FOLFOX (oxaliplatin, 5-FU and leucovorin) chemotherapy vs. FOLFOX chemotherapy alone in 2,710 patients with resected Stage II or Stage III adenocarcinoma of the colon. The primary endpoint was disease-free survival, while overall survival was the secondary endpoint.

Next year, Genentech and Roche expect data from another Phase III trial, designated AVANT, testing Avastin in combination with the XELOX (capecitabine and oxaliplatin) or FOLFOX chemotherapy regimens vs. FOLFOX alone.

Additional adjuvant trials are ongoing testing Avastin in HER2-negative breast cancer, HER2-positive breast cancer and nonsquamous, non-small-cell lung cancer.

Avastin, which is approved in first- and second-line colorectal cancer, first-line NSCLC and in metastatic breast cancer, made up $2.7 billion of Genentech's $9.5 billion 2008 product revenue. Earlier this year, the drug got a thumbs-up review from the FDA advisory committee, which recommended accelerated approval of Avastin in glioblastoma multiforme, marking the first time agency reviewers have used objective tumor response as approval criteria. (See BioWorld Today, April 1, 2009.)

The C-08 miss is not anticipated to have any effect on existing approvals of Avastin, but Raymond said it could cause a "temporary drag on sentiment" in terms of other commercial, oncology-focused companies, particularly Tarrytown, N.Y.-based Regeneron Inc. The company, along with partner Paris-based Sanofi-Aventis Group, is testing Aflibercept (its VEGF Trap) in a Phase III program in combination with standard chemotherapy regimens in prostate, NSCLC, colorectal and pancreatic cancer.

But even a Regeneron "read-through may be a stretch," he added, since Aflibercept is being tested in metastatic settings, and "Avastin's adjuvant failure should mean very little" to that program.