Accelerated approval of anticancer drugs: a safe choice?
All oncologists know how anxiously cancer patients look forward to the availability in clinical practice of drugs that have yielded positive results in the registration trials. And every year, returning from the annual meeting of the ASCO (American Society of Clinical Oncology), every specialist would like to be able to use immediately that drug or that therapeutic modality that came to the fore at the most important annual congress,
demonstrating that they can significantly improve the possibilities of treatment of cancer. On the other hand, there are certainly no shortage of examples of drugs that have been put on the market without reliable evidence of efficacy and that have then proved to be devoid of significant or even unacceptably toxic therapeutic activity.
In the United States, an accelerated approval pathway has existed since 1992 for drugs with "unprecedented activity in cancers": it allows market access based on the evaluation of surrogate outcomes1 and is, at least in theory, conditional as it requires the conduct of post-approval confirmatory studies to assess their real impact (in terms of risk-benefit ratio) in daily clinical practice. What is the situation after more than a quarter of a century?
The course must be corrected
Some articles2,3,4 question the whole procedure and highlight its potential dangers, recalling the need to correct the route5. Here are the reasons for this:
- the questionable choice of surrogate endpoints: in particular, the objective response may correlate little with overall survival, as documented also recently by a systematic review4 which shows that up to 90% of studies may demonstrate a low correlation between response and survival
- the lack of post-approval confirmatory studies: according to a recent analysis3 no more than 20% of approved therapeutic indications would meet a meticulous post-marketing scrutiny using the same endpoints reported in the registration studies, while another 21% use different methodologies and therefore risk not leading to reliable results. Of the others, little is known, except that in a quarter of the cases the confirmatory studies do not appear to have started or are in progress for a long time or the results are not reported
- cases such as bevacizumab in glioblastoma progression after first-line therapy: while the registration study had seen a clear superiority in terms of objective responses of the lomustine arm plus experimental drug over the standard with lomustine alone, the confirmatory study of over 400 patients could not show any advantage for the combination in terms of strategic outcomes including overall survival and had to double toxicity. Nevertheless, the FDA approved bevacizumab in this indication...
The limited percentage of confirmatory studies that contradict the results of the registration trials is not in itself evidence of the validity of the FDA's methodology, as too often there is still room for surrogate endpoints, with the risk of maintaining the commercial availability of drugs with dubious efficacy and safety and significant costs for public and private insurance coverage, or even for the patient. It would therefore be necessary to design confirmatory studies that use endpoints (typically global survival) different from those of registration studies and that are closer to the clinical reality and the needs of patients. Such studies should be conducted promptly, in order to overcome the intolerable quota of "lost in translation" research mentioned above. Above all, maximum consideration should be given to safety issues, ready to withdraw the approval of a drug if evidence of unacceptable toxicity emerges.
An inappropriate call?
This call for greater rigour may seem out of place at a time when the superlative adjectives in the description of the positive effects of cancer therapies abound, not infrequently without reason6, and when there is a widespread feeling that treatments directed to molecular targets and especially immunotherapy are able to chronicise the majority of metastatic neoplasms. In fact, even in the field of immune checkpoint inhibitors, accelerated approval has produced quite a few failures7, not forgetting that the various multinationals have produced at least 5 very similar drugs in this class and that each tends to occupy market space through the conduct of registration studies with limited differences between them, so as to justify the pervasive role of all these revolutionary therapeutic agents.
Moving on to Italy, we often run the risk of falling into the opposite excess and having to deal with a sort of "slowed down approval of anticancer drugs", which also includes truly innovative products that are desperately needed by limited segments of patients, but for which there is a very high probability of response for treatments aimed at a well-defined molecular target: think, for example, of metastatic colorectal carcinomas that present microsatellite instability and for which there is a marked inhibitory check-point activity. In this case, AIFA has been on a line of little understandable rigidity, denying reimbursement, while many other times the Agency simply delays to express itself, making it very difficult for oncologists who have to face every day the despair of patients and their families, in the obvious search for the best treatment for their disease. If this is seen as a way to save on pharmaceutical spending... well, it is certainly not the most ethical or even the most intelligent.
What the excessive acceleration and the unacceptable slowdown in the registration of drugs have in common is the lack of confidence of health authorities in the professional and scientific capacity of oncologists, who are perfectly capable of designing and conducting independent studies (therefore, with clinically significant endpoints) that confirm (or not) the data of the registration trials and that allow prompt access to truly innovative drugs. The battle for independent clinical research, largely focused on the activity of cooperative oncology research groups, is in full swing and represents the primary interest of our patients, as well as the raison d'être of every oncologist worthy of the name.
We will see how it ends: we are here, ready for confrontation.
Director DIPO (Dipartimento Interaziendale Provinciale di Oncologia), Bergamo
- Blumenthal G, Kluetz P, et al. Oncology drug approvals: evaluating endpoints and evidence in an era of breakthrough therapies. Oncologist 2017;22:762-7. CDI
- Chen E, Raghunathan V, et al. An overview of cancer drugs approved by the US Food and Drug administration based on the surrogate endpoint of response rate. JAMA Intern Med 2019;179:915-21. CDI
- Gyawali B, Hey S, et al. Assessment of the clinical benefit of cancer drugs receiving accelerated approval. JAMA Intern Med 2019;179:906-13. CDI
- Haslam A, Hey S, et al. A systematic review of trial-level meta-analyses measuring the strength of association between surrogate endpoints and overall survival in oncology. Eur J Cancer 2019;106:196-211. CDI
- Di Magno S, Glickman A, et al. Accelerated approval of cancer drugs – Righting the ship of the US Food and Drug Administration. JAMA Intern Med 2019;179:922-3. CDI
- Abola M, Prasad V. The use of superlatives in cancer research. JAMA Oncology 2016;2:139-41. CDI
- Gill J, Prasad V. A reality check of the accelerated approval of immune-checkpoint inhibitors. Nature Rev Clin Oncol 2019;16:656-8. CDI