CHICAGO – With the failure of Avastin to improve overall survival in a pair of studies in first-line glioblastoma, Roche/Genentech Inc. is again in the position of taking discordant confirmatory data to FDA in support of an accelerated approval, only this time they may have support from quality-of-life data.

Two separate studies looking at Avastin (bevacizumab) in first-line glioblastoma multiforme were presented at the American Society of Clinical Oncology annual meeting May 31-June 4.

Genentech carried on the tradition of Avastin being in the plenary spotlight at ASCO, where the top four studies are presented to the majority of the 25,000+ attendees. Genentech has been featured the last six years running, and Avastin has been a plenary feature nine times in the last 10 years. Avastin was the subject of two of the plenary trials, both conducted independently by cooperative groups, one a success (in cervical cancer) and one a failure (in first-line glioblastoma). The failed brain cancer trial was replicated in a separate study from Roche.

The company’s AVAglio study was the confirmatory trial for the 2009 accelerated approval to treat recurrent GBM; the follow-up study moved to the first-line setting. The second study was run by the Radiation Therapy Oncology Group and sponsored by the National Cancer Institute. There was a progression-free survival benefit in each trial (though the RTOG trial did not reach statistical significance), but no effect on overall survival.

Roche will be submitting the data to FDA, which might be an uncomfortable position for a company still recovering from the 2011 withdrawal of Avastin’s accelerated approval for metastatic breast cancer. FDA rescinded the approval, which was based on a strong PFS benefit, after confirmatory trials failed to match the original PFS result and failed to show a benefit in overall survival. As part of the proceedings, the agency indicated that positive quality of life data could have supported the claim Also see "Quality-Of-Life Data Take On Added Significance With Avastin Debate" - Pink Sheet, 18 July, 2011..

Picking Avastin’s Brain (Data)

The 637-patient RTOG study compared standard radiotherapy plus temozolomide to bevacizumab. While there was no survival difference (16.1 months for standard therapy compared to 15.7 months for Avastin), median PFS was 7.3 months for the placebo arm versus 10.7 months for bevacizumab. However, that did not reach the established level of benefit for PFS in the trial.

The AVAglio study had a similar design, using temozolomide and radiotherapy. Median survival was similar between the arms, but the bevacizumab arm had a median PFS of 10.6 months versus 6.2 months for the radiotherapy/temozolomide control – a statistically significant effect with a hazard ratio of 0.64 (p<>

“What differentiates these two trials and what we feel very good about is if you look at the AVAglio study, we did a very robust assessment of the quality of life with various measures,” because the impact of tumors on brain function make it very susceptible to loss of QoL with tumor progression, Hal Barron, head of Roche pharma global product development, explained during the firm’s June 2 on-site ASCO analyst event. “We really thought that reducing the time it takes patients tumors to progress would translate into improvements in quality of life. That’s in fact what we observed and we believe is really meaningful for this population,” he said.

[Editor’s note: For further coverage of the quality of life findings in glioblastoma, see the next issue of “The Pink Sheet.”]

Roche recently filed for first-line use in Europe based on this data. The company will take the AVAglio results to FDA in the second half of the year, Roche Pharmaceuticals COO Daniel O’Day told the analyst event, adding that it will be “discussing what that means relative to our accelerated approval filing with the very strong feeling from our side [that] this still has a very important play for patients with GBM around the world.”

Still, he pointed out that GBM sales in the U.S. (predominantly in the second-line setting) accounted for $179 million CHF in sales out of a franchise that was close to $6 billion CHF in 2012, “just to put it in perspective a little bit.”

Avastin’s Promise In Cervical Cancer

Roche will also be meeting with regulators to discuss Avastin’s use in cervical cancer.

Bevacizumab was shown to increase overall survival in combination with chemotherapy in the Gynecologic Oncology Group’s study 240 – which UCLA’s Gottfried Konecny described as a “practice-changing study” in his discussion of the findings during the plenary.

Lead investigator Krishnansu Tewari characterized the results as a “triple-header” – with statistically significant improvements in overall survival, progression-free survival and response rate. In addition, there was no deterioration in quality of life for patients who added Avastin to the chemo regimen compared to those on chemo alone.

There was an increase in known bevacizumab-related adverse events, namely gastrointestinal fistula and thromboembolic events, but the occurrence was “relatively low” at under 10%, Tewari said. No new adverse events were detected.

Standard treatment for cervical cancer has been platinum-based chemotherapy with a combination of paclitaxel and cisplatin, which yield median overall survival of less than 12 months. Concerns about growing platinum resistance prompted the use of one arm of the GOG 240 trial to test an alternative chemo regimen of topotecan plus paclitaxel, though that was discontinued part-way through the study after it became clear that it was not better than the cisplatin combo.

Median overall survival for Avastin plus chemo was 17.0 months versus 13.3 months for chemo alone, a reduction in risk of 29%. There was a 33% reduction in median PFS, at 8.2 months versus 5.9 months. And the response rate was 48% versus 36% for chemo-alone, with significant more complete responses on the Avastin arm.

The National Comprehensive Cancer Network is already considering adding the GOG 240 findings to its cervical cancer guidelines, Tewari reported. Genentech Senior VP-Global Hematology/Oncology Product Development Sandra Horning said in an interview that the company was discussing the results with regulators. Incorporation into the NCCN compendia, however, should be enough to trigger adoption and coverage.

As cervical cancer levies a heavy burden in the developing world, both Tewari and Konecny commented on the need for cost-effectiveness research. “We urgently need an anti-angiogenic strategy that can benefit the global population,” Konecny concluded.

They each noted the lack of predictive factors that could be used to select appropriate patients. Genentech has an extensive program looking for VEGF biomarkers, Horning confirmed. Its Phase III program in metastatic breast cancer is examining plasma VEGF – an outgrowth of its negotiations with FDA over the now-rescinded accelerated approval in that setting.

The lack of biomarkers was also bemoaned by the RTOG investigators, who were hopeful that prognostic markers could be found to identify a responsive subgroup.

Konecny commented on the strength of the cervical cancer findings relative to the other settings, both approved and not, where Avastin has been studied. In many of those settings, the monoclonal antibody has demonstrated an advantage in PFS but not OS. “In contrast, [GOG] 240 saw relative reductions in both, independent of chemo,” he observed. “It’s fair to say GOG demonstrated very favorable data compared to what bevacizumab does in other cancers in the first-line.”

Part of that is attributable to the study design, which, he noted, did not allow for cross-over. That was a negotiated feature, Konecny said. Many oncology trials provide for patients to switch over to the investigational arm upon progression, though this undermines the study’s ability to provide a read on overall survival.

Roche is still pursuing Avastin’s use in another gynecological cancer. The firm is awaiting final survival data from its ovarian cancer program for U.S. filing, though GlaxoSmithKline PLCannounced at ASCO that it intended to move forward with a submission for Votrient (pazopanib) in ovarian cancer based on PFS data ([A#14130601001]).

Avastin is already approved for ovarian cancer in Europe and is in the NCCN compendia listings in the U.S. Cervical cancer adoption is likely to follow the ASCO data release.

Avastin Takes One Step Forward, One Step Back At ASCO

More from Clinical Trials

More from R&D