It hasn't been a swell summer for Roche, and now a bit of near-term pipeline news that could have put the company on a more positive trajectory has fallen flat. FDA refused to file Roche's BLA for breast cancer agent trastuzumab-DM1 (T-DM1), a conjugated version of Herceptin , for accelerated approval, the company announced Aug. 27.

Roche has already faced two significant pipeline setbacks this summer, a disappointing advisory committee review for Avastin (bevacizumab) in breast cancer and safety issues with taspoglutide for type 2 diabetes. But given the early Phase II data used for the filing, FDA's decision on T-DM1 wasn't entirely unexpected either.

The filing was based solely on the results of a single-arm 110-patient Phase II study, which showed T-DM1 shrank tumors in one-third of women with advanced HER2-positive breast cancer who had received an average of seven prior medicines, including two HER2-targeted agents. While the data were positive, it's not surprising that FDA would want to see more, especially given that the agency is taking a harder look at progression-free survival as an endpoint for breast cancer and how it relates to clinical outcomes.

The regulatory action pushes a potential re-filing and launch out two years, pending the results of an ongoing Phase III trial. Dubbed EMILIA, the trial compares T-DM1 to GlaxoSmithKline's Tykerb (lapatinib) in combination with capecitabine in people with advanced HER2 positive breast cancer whose disease has worsened after initial treatment. Roche said it expects to resubmit the BLA in mid-2012, which would position the drug for a launch in 2013.

During Roche's investor meeting in March, the company admitted the filing strategy had a lower probability of success than one based on significantly more data. "If we file early, we file on early data and the risk of not succeeding there is higher," acknowledged Jean-Jacques Garaud, head of research and early development.

FDA decided not to file the application because it did not meet the standard for accelerated approval, Roche said, because all the available treatment choices approved for metastatic breast cancer had not been exhausted in the study population.

"We firmly believe in the potential of T-DM1 as a novel HER2 targeted option and remain fully committed to its ongoing development," the firm said in a statement.

Still, Roche did consult with FDA prior to the submission, which suggests the agency offered some positive feedback at that time, and it's not clear why, if at all, the agency then changed its thinking. But certainly FDA must be feeling cautious about moving toward an accelerated approval after its Oncology Drugs Advisory Committee voted against granting full approval of Avastin for first-line treatment of metastatic breast cancer in July, basically nixing FDA's accelerated approval for the indication in 2008. That decision was based largely on the committee's feelings that the drug did not delay disease progression long enough to be clinically meaningful (¹ Also see "Avastin Breast Cancer Approval Debate Pits Clinical Benefit Versus Statistics" - Pink Sheet, 26 July, 2010.).

The vote in favor of removing the breast cancer indication from Avastin's labeling was a controversial one. If FDA ultimately agrees with ODAC, it will put a pinch on Roche's top-line this year (² Also see "Clinical Development Setbacks, Pricing Pressures Hit Roche" - Pink Sheet, 26 July, 2010.).

"We can't speculate if the Avastin decision had anything to do with this decision," a Genentech representative said. "But FDA continues to make it clear that they want randomized data and overall survival data for approval of new breast cancer drugs."

The case of another problematic accelerated approval, Pfizer's Mylotarg (gemtuzumab) for acute myeloid leukemia, is also fresh in the agency's mind. Mylotarg was voluntarily pulled from the market by Pfizer in June, 10 years after it received accelerated approval, when it failed in a confirmatory trial (³ Also see "Pfizer Pulls Mylotarg For Safety 10 Years After Accelerated Approval" - Pink Sheet, 28 June, 2010.).

Roche has a lot riding on T-DM1, an antibody-drug conjugate, or so-called "armed antibody," that attaches trastuzumuab and the chemotherapy DM1 together using a stable linker and is delivered directly to the cancer cells. DM1 and the linker technology are licensed from Waltham, Mass.-based ImmunoGen, which is also banking on the product's success. The approach is expected to offer additional efficacy benefits over Herceptin and also greater tolerability given the targeted delivery of the cytotoxic agent.

Genentech is looking to T-DM1 to drive continued growth of its blockbuster Herceptin franchise and eventually replace Herceptin as the standard of care. Roche is also looking to develop T-DM1 in combination with another investigational agent, pertuzumab, with the aim of eliminating the need for chemotherapy. Pertuzumab is an HER2 blocker that is believed to work synergistically with Herceptin.

The FDA decision is also a road bump for ImmunoGen, delaying the royalties and milestones the company would receive upon approval and launch. The approval milestones would be in the low double-digits in the millions of dollars, said ImmunoGen CEO David Junius. Until now, the positive Phase II data and praise from Roche have helped validate the company's platform technology and put the company on an aggressive growth track (⁴ Also see "ImmunoGen Gains Momentum On Success Of T-DM1, The Herceptin Conjugate In Development By Roche/Genentech" - Pink Sheet, 8 June, 2010.).

"We continue to believe in the potential for T-DM1 and don't see FDA's decision as a reflection on that," Junius said during a same-day conference call. "This appears to be a technical issue, not a fundamental one."

Nonetheless, the company acknowledged the financial impact and said it would provide revised financial guidance shortly. The company does not have any immediate plans to reduce its cash burn, Junius added. ImmunoGen had approximately $110.3 million in cash and marketable securities as of June 30.

FDA's refusal to file may dampen general investor enthusiasm for conjugated therapies, but it seems like the decision has more to do with the accelerated approval than the technology itself.

Seattle Genetics, another drug developer working on antibody-drug conjugates, is also partnered with Genentech. But Seattle Genetics said it isn't concerned that the latest action will derail such programs. "We do not believe the FDA's decision on T-DM1 has broader implications for antibody-drug conjugates," the firm said. "In our view, the decision appears to relate to the treatment landscape for metastatic breast cancer rather than more generally applicable principles."

The company's lead program, SGN-35, which combines a CD30 antibody with its ADC technology, is in Phase III for relapsed and refractory Hodgkin's lymphoma patients under a Special Protocol Assessment with FDA. The firm said FDA has agreed the trial design meets the requirements for accelerated approval.

- Jessica Merrill ( ⁵ j.merrill@elsevier.com )

Roche's T-DM1 Early Filing Strategy Backfires As FDA Refuses To File BLA

More from Archive

More from Pink Sheet