Abstract
‘What is a hero? Now, my definition is completely different. I think a hero is an ordinary individual who finds the strength to persevere and endure in spite of overwhelming obstacles’ Citation. Christopher Reeve
A true superhero
To many, as a 6 foot 4 inch, 215-pound Superman, he was a hero. On 27 May 1995, Christopher Reeve’s life changed very dramatically. His horse, Eastern Express (Buck), stopped cold before a jump and Christopher Reeve flew over his head and hit the top of the fence, overextending his neck and suffering a C1, C2 vertebral fracture. He survived to become a wheelchair-ridden, ventilator-dependent, quadriplegic. His choice was to live and to make the lives of others better through his personal experience and his influence. He worked to promote scientific research. He lobbied for increased funding for the development of novel therapies. He strove to influence bureaucracies to act expediently in the review and approval process. He was a strong voice and a relentless advocate for patients. All of this was not just for the benefit of those with spinal cord injuries, but on behalf of all who are not being heard as they plead for a better future, for hope through the expedited availability of new promising therapies for their implacable diseases. As a quadriplegic, he became a true superhero. He was simply amazing (a phrase used by Christopher Reeve when speaking about others and that I believe best applied to him) (Box 1). Christopher Reeve passed away on 10 October 2004, but his courage and his example have left a positive and indelible mark on many of us. Again, he was a true superhero. Superheroes live forever.
In a previous editorial, I advocated the more frequent use of the accelerated approval mechanism as a way to expedite drug develop-ment, review and approval Citation[2]. I suggested that showing some clinical activity in Phase II trials should suffice for the approval of a new anticancer agent and that this is, in fact, possible today under current regulations without the need to enact new legislation or make any changes to our bureaucratic pro-cesses. I stressed that the duty of our regulatory agencies should be the evaluation and approval of new anticancer agents as single agents (developed by or under the sponsorship of pharmaceutical companies) and that the evaluation of combination therapies is the responsibility of the oncology community (e.g., via the cooperative study groups, consortia and academic sites). Importantly, I emphasized that the optimal development of potentially curative combination therapies takes a significant amount of time and effort, that overseeing this (and much less requiring this) for approval purposes is not the job of our regulatory agencies, and that patients should not be forced to wait that long for promising new therapies to be approved.
Simply stated, single-agent development is the responsibility of the pharmaceutical industry, single-agent evaluation and approval is the job of the regulatory agencies, and combination therapy development and optimization is the task entrusted to our oncology community . Regulatory agencies have no business intervening with, requiring or regulating the developmen-t of combination therapies.
Why not? Why should we not provide the cancer patient with the earliest possible access to promising new agents, particularly those that actually show a response rate in the cancer in question? Why should cancer patients die without the benefit of existing new therapies because the lengthy development and approval process precluded their use during the patient’s lifetime? It should not take a rocket scientist or a brain surgeon to answer these questions. The answer is simple and intuitive; yes, we should be developing, approving and providing these new therapies as fast as possible. How can we make it happen? Even within our bureaucratic processes, the possibility exists in the form of the accelerated approval mechanism. Let us be optimistic for a moment and consider a situation where the proposal I have described is accepted and new anticancer agents would be approved based on some clinical activity in Phase II trials.
Hurdles pre accelerated approval
Wasn’t it Shakespeare who once said ‘first we kill all the lawyers’? We have no lawyers to kill, but there are many hurdles to jump over. Before accelerated approval can proceed as suggested, the regulatory agencies have to accept that their job is really the evaluation and approval of new anticancer agents as single agents and the accelerated approval is the best mechanism (at least for now). Then we would have to agree that response rate suffices to show clinical activity and that data on duration is desirable but not really necessary at this stage . Furthermore, the acceptable response rate should be the lowest reasonable rate, not to be compared with available therapies (much less with combinations), and statistics (other than number of responders divided by number enrolled in the study) are not necessary. Development would be simple, efficient, rapid and economical. We would save on Phase III trials (the most expensive and time-consuming part of development). We would save on regulatory evaluation and approval (less bureaucracy). Pharma-ceutical companies and regulatory agencies would need fewer resources, equipment and physical facilities. These savings should be passed on to the patient by the pharmaceutical company in the form of lower drug prices, and to the taxpayer by the regulatory agency (the government) in the form of lower taxes or better use of these surplus monies by increasing research funding.
The approval decision would be a clinical decision based on the judgment of experts treating cancer patients every day. This is also an existing mechanism; in the USA it is the Oncologic Drug Advisory Committee (ODAC). The oncology community could then take over and conduct the appropriate Phase III trials to determine the optimal use of the new agent in combination regimens. These Phase III trials may take many years, but in the meantime the new agent would be available to those who could benefit from its use. I will be simply amazed should all of this come to pass in my lifetime. Nevertheless, one always hopes (and perseveres, and continues to insist, and insist and insist).
Are there negatives? Yes, a toxic agent with very limited or no activity could be approved. Toxicity, for most agents, is usually characterized in early development. Even in a dossier with limited patient numbers, there will be a minimum of 300 patients evaluable for safety. An unusual adverse event that occurs in, say, one of every 10,000 patients may not be seen, even in the larger Phase III trials. Long-term toxicities may take many years to characterize. Besides, I know of no other physician group that is faster at recognizing, understanding, characterizing and reporting adverse events than hematologists and oncologists (because we have the most experience with adverse events). Thus, if one considers this negative relative to the positive effects of early availability, it is reasonable to make a risk–benefit decision that favors the obvious benefits. Another possible negative is the approval of an agent with limited or no clinical activity. There will be some cases where such an agent had a reasonable response rate in Phase II trials but later proved ineffective in the larger Phase III studies. At that point, the oncology community will stop using that agent and the drug will die. It will not be necessary for the regulatory agencies to remove it from the market (more savings). Hematologists and oncologists are some of the best informed and most experienced specialists in modern medicine. We know very well what’s good and what’s not for the cancer patient.
Accelerated approval experience
The sum total of new anticancer agents approved under this mechanism since it became available in 1992 (through January 2004) is 18 Citation[3]. We should all be glad to hear this. However, the critical reader will note that only 11 of these 18 agents were approved based on response as the sole end point and based on a single-arm trial. This is the subset that truly represents the intent of accelerated approval. In the case of the other seven agents, there were multiple indications and randomized trials were available for some, or the end points included some measure of response duration. A total of 11 agents in over a decade, approximately one agent per year, is hardly a record of which to be proud. Pharmaceutical companies and regulatory agencies need to work together to improve on this.
What is not available in this publication is information on denominators. A total of 18 agents, yes, but out of how many? How many agents have been discussed with the US Food and Drug Administration (FDA) since 1992, and an accelerated approval strategy proposed? How many of these proposals were declined and a strategy for regular approval requested? How many agents since 1992 have actually followed an FDA-sanctioned accelerated approval strategy? How many failed and why? What percentage of the total number of approvals were based on accelerated approval? Many have been critical of this mechanism. Some have called it a back-door approach to approval. Accelerated approval has been made more and more difficult since it became available in 1992. It appears as if the intent is to make an example and dissuade others from utilizing this mechanism. Pharmaceutical companies have become increasingly reluctant to use this mechanism. How many of the 300-plus new agents currently under investigation are following an accelerated approval strategy?
Nevertheless, the authors of this publication state that ‘accelerated approval is commonly based on objective response rates in single-arm studies in patients with refractory disease, with confirmatory studies conducted in related populations, usually in patients with less refractory or earlier stage disease. This approach may expedite drug availability to patients with life-threatening disease because single-arm studies are completed and analyzed more rapidly than large, randomized trials’. The authors then conclude that ‘… the accelerated approval program in oncology has been successful in making 18 different products available to patients for 22 different cancer treatment indications’. Although these statements underline the approach that should be followed, they are not representative of actual experience. In this paper, the authors decry response rate as an end point and strongly suggest the use of randomized trials with duration end points. These statements imply the successful application of this mechanism, whereas, one could view the experience to date as a failure to capitalize on an opportunity to bring many more promising new agents to the patient early on.
Hurdles post accelerated approval
Again, let us be optimistic for a moment and consider a situation where the proposal I have described is accepted and new anticancer agents would be approved based on some clinical activity in Phase II trials. Even then, these agents might not be available to patients depending on reimbursement issues and policies of the health authorities. A new drug and related administration expenses might not be included in a formulary or might not be reimbursed by insurance companies, other third parties, and importantly in the USA, by Medicare. These and other financial and bureaucratic hurdles may still be imposed post accelerated approval. We certainly need more heroes, as defined by Christopher Reeve, to address and help resolve all of these issues for the benefit of the cancer patient.
Then again, one might think that this is entirely the wrong time to advocate accelerated approvals. We have only recently experienced the problems described with rofecoxib (Vioxx®) and celecoxib (Celebrex®). The FDA has been taken to task by the public for not identifying these earlier, and blamed for a lapse in their safety surveillance. More recently, natalizumab (Tysabri®), a new agent for multiple sclerosis, has been removed from the market for serious adverse events including a death from progressive multifocal leukoencephalopathy. Again, the FDA has been under fire. In cancer, gefitinib (Iressa®), a relatively recent accelerated approval, now shows no activity in a large-scale randomized trial.
As discussed above, unusual, long-term and infrequent adverse events cannot be predicted and may not be seen until many patients have been treated and observed over many years. The finding of such adverse events some years after approval is no-one’s fault, it is not the fault of the pharmaceutical company and it is not the fault of the FDA. One must consider how many patients have benefited from the early availability of the agent and then weigh risk versus benefit. These unusual cases should not stand in the way of early approvals that can benefit thousands of patients.
Conclusions regarding accelerated approval
The accelerated approval mechanism was intended to make new promising anticancer agents available early on to patients who would benefit from their use. The name of the mechanism makes the intent quite clear; to accelerate approvals. We need to utilize this mechanism more and more frequently. We need to overcome the hurdles that stand in the way of its application. We need to make it easier and not more bureaucratic. We need to keep it simple and not convert the clinical trials appropriate for this purpose (Phase II trials) into large randomized studies, thus defeating the purpose of accelerated approval. We need to be pragmatic regarding end points and accept some reasonable response rate as sufficient. We need to differentiate between cancer drug development (single agent) and treatment development (combination therapies), and who is responsible for each. Above all, we must persevere, endure and insist that the steps be taken to bring new agents promptly to the cancer patient.
Christopher Reeve was not able to jump over that last hurdle. Likewise, the cancer patient has numerous hurdles to overcome. However, these hurdles are not there for sport. They are excessive, bureaucratic and often unrealistic man-made hurdles, and should not be imposed on the cancer patient and the heathcare personnel caring for the patient. There are enough challenges presented by the disease itself without more hurdles being imposed. In turn, those who do care for the welfare of the cancer patient need to endure, insist and hope that these hurdles will be removed. We need more such heroes.
‘When we have hope, we discover power within ourselves we may have never known – the power to make sacrifices, to endure, to heal and to love. Once we choose hope, everything is possible. We are all on this sea together. But the lighthouse is always there, ready to show us the way home’ Citation[4]. Christopher Reeve.
Information resources
Additional reference and background material can be found at the FDA’s website, www.FDA.gov.
Table 1. Who is responsible for what?
Table 2. Stages in the development of new anticancer therapeutics.
Box 1. To amaze.
To fill with great surprise or sudden wonder (e.g., Christopher Reeve’s courage and perseverance always amazed us).
New World Dictionary of the American Language
References
- Reeve C. Still Me. The Random House Publishing Group, NY, USA, (1998).
- Grillo-López AJ. The ODAC Chronicles: Part 2. Statistics and clinical medicine in the USA: the triumph of science over art? Expert Rev. Anticancer Ther. 4(6), 941–944 (2004).
- Dagher R, Johnson J, Williams G, Keegan P, Pazdur R. Accelerated approval of oncology products: a decade of experience. J. Natl Cancer Inst. 96, 1500–1509 (2004).
- Reeve C. Nothing is Impossible: Reflections on a New Life. Ballantine Books, NY, USA (2002).