One week after the European Commission approved the use of Avastin (bevacizumab) in combination with standard chemotherapy (carboplatin and paclitaxel) as a front-line treatment for women with advanced ovarian cancer, findings from two Phase III studies published in the New England Journal of Medicine suggested the drug offered a modest but noticeable benefit over standard chemotherapy in terms of progression-free survival (PFS).

But with overall survival (OS) now the key yardstick – some would say the only yardstick – used by the FDA to measure cancer drugs, Avastin developer Roche AG/Genentech Inc. is backing away from the battle to seek approval for the indication in the U.S.

'We are pleased that women with ovarian cancer in Europe will have access to a new therapy with Avastin, the first major treatment advance for women with newly diagnosed advanced ovarian cancer in 15 years,' Genentech spokeswoman Charlotte Arnold told BioWorld Today.

Reiterating guidance provided during Roche's third-quarter earnings call, however, 'we have not submitted the data for a new indication in the United States,' Arnold said, adding there was 'a low likelihood' the company ultimately would pursue an FDA application for Avastin in ovarian cancer.

Arnold said a final decision would be made after Roche analyzed complete data from the Phase III trials in ovarian cancer. The company is awaiting final OS data from several Phase III studies.

In the Phase III ICON7 trial, conducted by the Gynecologic Cancer InterGroup, 1,528 women at 263 sites in 11 countries were randomized into carboplatin and paclitaxel every three weeks for six cycles, or to that regimen plus Avastin every three weeks for five or six cycles and continued for 12 additional cycles or until progression of disease.

Ninety percent of the women had epithelial ovarian cancer, 69 percent had a serous histologic type, 9 percent had high-risk early stage disease, 30 percent were at high risk for progression and 70 percent had Stage IIIC or Stage IV ovarian cancer. In initial analyses, PFS at 36 months was 20.3 months with standard therapy and 21 .8 months with the addition of Avastin, which also was associated with more toxic effects – most often, hypertension of grade 2 or higher.

PFS was largely unchanged in the analysis at 42 months: 22.4 months with standard therapy compared to 24.1 months with Avastin. In patients at high risk for progression, however, Avastin showed greater benefit, with PFS of 18.1 months at 42 months compared to 14.5 months with standard therapy, and median OS of 36.6 months compared to 28.8 months with standard therapy.

The second NEJM report – the double-blind, placebo-controlled Gynecologic Oncology Group (GOG-0218) study – enrolled 1,873 women with newly diagnosed Stage III or Stage IV epithelial ovarian cancer who had undergone surgery into one of three arms. All of the patients received paclitaxel and carboplatin. One-third also received Avastin for the first six three-week cycles, and another third received Avastin through 22 three-week cycles.

The primary endpoint was PFS, with a median PFS of 10.3 months in the control group, 11.2 months in the Avastin-initiation group and 14.1 months in the group that received Avastin throughout the study. At the time of analysis, 76.3 percent of patients remained alive, with no significant differences in OS among the three groups. The rate of hypertension requiring medical therapy was higher in the Avastin-initiation group (16.5 percent) and the group that received Avastin throughout the study (22.9 percent) compared to the control group (7.2 percent).

Roche originally had reported data from both studies in 2010. Last year, the company also reported data from OCEANS, its Phase III study evaluating Avastin in combination with gemcitabine and carboplatin followed by continued use of Avastin in women with previously treated platinum-sensitive ovarian cancer.

The data suggested that women who received Avastin experienced a 52 percent reduction in the risk of their disease progressing compared to women who received chemotherapy alone. The results translated into median PFS of 12.4 months in the Avastin arm compared to 8.4 months in women who received chemotherapy alone.

With the drama of last June's unprecedented two-day hearing before the FDA's Oncologic Drugs Advisory Committee and the agency's subsequent revocation of the drug in breast cancer, Avastin provided one of biotech's more interesting story lines last year. (See BioWorld Today, June 29, 2011, June 30, 2011, and Nov. 21, 2011.)

The question now is what comes next for the drug – at least in the U.S. Avastin remains on the market as an approved treatment for certain types of colon, lung, kidney and brain cancers. However, its use in HER2-negative metastatic breast cancer declined to about 15 percent to 20 percent of eligible patients at the time of the FDA commissioner's decision in November, according to Arnold, from a peak of 60 percent prior to the advisory committee meeting.

The prospects for increasing Avastin's availability in the U.S. cancer market aren't likely to change without solid OS data, according to David Miller, CEO of the independent research firm Biotech Stock Research LLC, of Seattle. The Office of Hematology and Oncology Products – and Director Richard Pazdur, specifically – have made it clear that oncology drugs won't move forward in the U.S. on promising PFS data.

'This is a bias that Pazdur has had and continues to enforce,' Miller told BioWorld Today. 'We can argue about cost vs. benefit – which is an argument that our health care system has to have – but that's not an argument the FDA is supposed to be involved in.'

For the time being, Avastin remains in use for certain patients with metastatic breast cancer, and insurers are reimbursing it, Miller added.

Pursuing an off-label route in ovarian cancer is a logical next step, he said.

'Avastin is used off-label often enough because of compendia listings that I don't think Roche needs to push for FDA approval,' Miller said.

Roche does not promote the off-label use of its compounds or present data to national compendia, such as the National Comprehensive Cancer Network, for off-label uses, according to Arnold.

But Miller pointed out that having published data in medical journals might serve the same purpose, not only for Roche but for other oncology drug developers.

'Companies are going to find some place to get their drug approved, sometimes even in last-line indications where they can use PFS or response rates,' he said. 'Then, while they're running the clinical trials to convert from accelerated to full approval, they'll put out Phase I and Phase II studies necessary to get compendia listing in earlier stages of the disease. This strategy is not new, but we're going to see it a lot more often.'