There is a provocative article in this week’s New England Journal of Medicine asking whether the accelerated approval process should be used for more cancer drugs:
“The striking results of recent phase 1 trials of targeted cancer drugs have provoked serious discussion about shortening the road to drug approval.”
The main thrust of the argument was that it takes on average seven years from entering human trials to approval if phase III trials are included in the oncology drug development process.
“Of the 23 oncologic drugs given accelerated approval between 1993 and 2008, two were ultimately withdrawn from the U.S. market — gemtuzumab because of toxicity and gefitinib because of lack of efficacy.”
The author argues that:
“Phase 3 trials are expensive and time-consuming, usually taking at least 2 to 3 years to reach survival end points. The news of a highly successful new compound in phase 1 or 2 rapidly reaches physicians and patients, creating demand for early access.”
While we have seen some successes with the Accelerated Review process with imatinib (Gleevec), erlotinib (Tarceva), cetuximab (Erbitux), bevacizumab (Avastin) and others, there has also been a flood of promising phase II agents that duly flopped in randomised phase III trials, with Pfizer’s figitumumab, Novartis/Antisoma’s ASA404 and sanofi-aventis’ iniparib to name a few off the top of my head.
One of the challenges here is that companies often take a targeted therapy but strangely test it in an unselected patient population, which will increase the chances of failure in a more rigorous randomised phase III trial. In an ideal world, several carefully designed adaptive phase II trials would help develop logical combinations and markers of response, thus increasing the chances of success in phase III studies.
The big problem as I see it then, is that while we have exciting new agents likely to be approved in 2011 such as:
- crizotinib in lung cancer
- ipilimumab (Yervoy) and PLX4032 (vemurafenib) in metastatic melanoma
- brentuximab vedotin in Hodgkin Lymphoma
- abiraterone acetate (Zytiga) in prostate cancer
we don’t always know how to carefully select patients to enable treatment based on the underlying molecular basis of the disease.
Of those mentioned above, with crizotinib (ALK), vemurafenib (V600E BRAF) and brentuximab (CD30) we clearly do, but with abiraterone and ipilimumab the issue of patient selection seems less clear at the moment. Sadly there are not any biomarkers available to tell us which patients are most likely to benefit from treatment in targeted therapies already approved such as bevacizumab.
The fate of bevacizumab in breast cancer has yet to be determined. It was approved under the FDA Accelerated Review process based on the initial phase II data, with the assumption that the phase III trials would show an improvement in overall survival. The AVADO and RIBBON1 trials showed a benefit in progression free survival or PFS (ie a surrogate marker of event free survival), but unfortunately were not positive for overall survival, which is the ultimate measure of disease progression and the condition required to be met under the fast track process.
The FDA are therefore recommending withdrawal based on the lack of overall survival as per the accelerated agreement and Roche offering the counter argument that there was evidence of patient benefit. That issue will no doubt continue to be debated for much of 2011 until the public hearing later this year. There is an excellent analysis of the impact of the FDA recommendation on bevacizumab uptake in the US in Oncology Business Review for anyone interested in trends.
Ultimately, we need have a better understanding of the molecular basis of the cancer types and drugs developed to target that aberration in a more carefully selected patient population. The arguments for and against accelerated review will run and run – probably for longer than the debate about how to pronounce some of the new names!
My position on the accelerated review process?
When it works, it works well. However, problems can arise when you get phase III trials that do not support the full approval due to a lack of a proven overall survival advantage in the population evaluated. The FDA can find themselves in an impossible position especially given the high emotions that run in breast cancer, for example. The onus should be on the company to do further research or trials better defining the patients who are most likely to respond rather than risk exposing thousands of patients to the systemic effects of a drug that may not offer meaningful benefit to the majority.
Chabner, B. (2011). Early Accelerated Approval for Highly Targeted Cancer Drugs New England Journal of Medicine, 364 (12), 1087-1089 DOI: 10.1056/NEJMp1100548