A new article published in the Journal of the American Medical Association (JAMA) details the preapproval and confirmatory clinical trials of drugs under the US Food and Drug Administration's (FDA) accelerated approval programme and brings several issues to light in the process. Some of the key findings of the report were:
- Between 2009 and 2013, the FDA granted accelerated approval to 22 drugs covering 24 indications. Of these, 19 were oncology indications. At a minimum of three years of follow-up, only half had completed confirmatory studies.
- The median number of patients enrolled in the preapproval studies for these drugs was 132, with a range of 89-224. Of the studies, 27% included fewer than 100 participants.
- In 42% of indications, post-approval requirements were satisfactorily completed, which means the trial was completed and clinical benefit from preapproval study was confirmed. Among the remaining 58% of indications where trials had not been completed, 8% failed to demonstrate clinical benefit, 8% were terminated, 13% were delayed for more than one year, and 29% were progressing according to target timelines.
- Clinical benefit has not yet been confirmed for 33% of indications that had been initially approved five years or more prior.
The findings of this JAMA study provide data to illustrate the pitfalls of the accelerated approval programme that critics have previously observed. The low number of patients, in what are often single-arm trials that lead to accelerated approval, has been one cause for concern. Critics argue that the programme allows drugs that have shown efficacy in a very limited and highly selective patient pools to enter the market without demonstrating clinical benefit in a real-world patient population that is likely much more heterogeneous. There has also been objection that many confirmatory trials are either never completed or significantly delayed, which allows drugs with unconfirmed clinical benefit to remain on the market for a prolonged period of time.
Roche’s Avastin received accelerated approval in 2008, in combination with docetaxel, based on progression-free survival (PFS) benefit in metastatic breast cancer. The combination remained a treatment option for breast cancer patients until the approval was rescinded three years later after multiple post-marketing trials failed to show improvement in overall survival. It also reported considerable toxicity issues associated with Avastin usage.
Another example is AstraZeneca’s lung cancer drug Iressa. Originally approved via the accelerated pathway in 2003, Iressa’s approval was later rescinded in 2012 based on lack of confirmatory survival benefit. This case differed from Avastin because Iressa was able to demonstrate PFS benefit in a smaller group of patients in 2015 with a specific mutation in lung cancer. This data supported its full FDA approval as a frontline therapy in this specific patient population, and facilitated the relaunch of Iressa.
The FDA has rarely withdrawn approval for an indication where confirmatory data have been negative in the 25 years since the introduction of the accelerated approval pathway. Even when such actions are taken, there is significant lag time between confirmatory trial failure and the revocation of approval. In May 2017, Roche’s Tecentriq failed its confirmatory Phase III trial in bladder cancer, but the FDA has not acted to rescind its accelerated approval, which was granted in May 2016.
Though not a fool-proof system, the accelerated approval programme remains an important part of the drug development process, as it allows early access to potentially life-saving drugs. Some of the strongest proponents of the system are patients and their families, as pointed out by former head of the FDA Robert Califf in an editorial accompanying the JAMA report. He said it is unlikely that there will be an overhaul of the system.
The concerns highlighted in the report will hopefully be taken into consideration as the FDA mulls its future policies on accelerated approvals.