This post is a part of our Bioethics in the News series. For more information, click here.
By Tom Tomlinson, PhD
Why are so much money and effort put into research aimed at curing cancer, and so little devoted to preventing it?
Three professors think they have at least part of the answer, as explained in a recent New York Times article (Why Preventing Cancer Is Not the Priority in Drug Development). They report that between 1971 and 2011, there were 12,000 research trials for drugs to treat patients with later stage cancer and a 90% chance of dying in five years; but only 6,000 for earlier stage patients with a 30% chance of dying. Even more starkly, 17,000 trials enrolled patients with recurrent cancers and low chances of survival, compared to 500 studying cancer prevention, where success would potentially yield much more benefit for more people (Budish, Roin and Williams).
In other words, the most money is being spent in pursuit of the least benefit for the fewest people. Why is this?
At great risk of over-simplifying, their answer is pretty simple: the commercialization lag. This is the time span between obtaining a patent for a new drug, and getting FDA approval to market it. The 20-year patent clock starts ticking when the patent is granted, but a company doesn’t begin to make money until the FDA allows sales. The shorter the commercialization lag, the more time available to amass a profit before your blockbuster goes generic.
So how does this affect cancer drug development? If the measure of success (and FDA approval) is increased cancer survival, success can be determined most quickly in those already likely to die soon. Compared to a group of cancer patients with a 1-year life expectancy, it will take 10 times as long to detect success in a group with an average life expectancy of 10 years. This is why the commercialization lag adversely affects the profitability of research to prevent cancer or to treat it in its early stages, and so discourages such research.
Understandably, the authors of the report then discuss remedies that mitigate the effects of the commercialization lag—e.g., identifying “surrogate endpoints” that can be detected much sooner, are strongly associated with increased survival, and allow for earlier FDA approval.
I’m sure they would admit that there are other factors discouraging early-stage and preventive cancer research. Here are several that occurred to me, some of which have an ethical component.
1. Enrolling early stage patients in a trial of a new, as yet unproven cancer drug could be ethically treacherous, if there is already a standard treatment regimen supported by evidence. The cleanest design would assign some patients to the new drug and others to the standard of care, and this would be especially necessary if the surrogate end point also occurred among patients receiving conventional treatment. But this would mean that the proven treatment was being deliberately withheld from the experimental group, or delayed until the surrogate end point window closed. The interests—even the vital interests—of this group would be sacrificed for the advancement of medical knowledge. Of course, we would require that they give their informed consent to being exposed to that risk. But the quality of such consent would be highly suspect, given that it is so clearly at odds with their self-interest. These are the concerns that underlie the principle of “equipoise”—that the arms of an experiment should be roughly comparable in terms of their potential risks and benefits. (See Declaration of Helsinki, #30.)
This doesn’t mean that it is always ethically impossible to conduct research on early-stage cancer, just that it will probably be ethically more challenging to design such trials.
2. Another challenge is enrolling an adequate number of participants, especially for research on agents to prevent cancer. Studies of potential preventive drugs for cancers that occur later in life and at relatively low population rates will require large numbers to be enrolled for long periods of time to detect a statistically significant effect. If the study agent carries side effects, or burdens of compliance with the research regimen, it may discourage enrollment and encourage later drop outs.
We might address this problem by enrolling people already at high risk for cancer, which will reduce the numbers required to detect an effect. This strategy probably raises no special issues for people whose increased risk is outside their control, like women and men with one of the BRCA genetic variants carrying much higher risk of breast cancer.
But how about a study focusing on preventing lung cancer, which enrolls heavy smokers to more easily achieve scientific feasibility? Its feasibility in part depends on whether the subjects continue to smoke. And the goal of the study—to determine whether the new drug reduces the risk of smoking—may indirectly encourage participants to continue smoking. This would be due to the “therapeutic misconception,” the well-documented belief among many research participants that the experimental intervention offers a much higher likelihood of benefit than the evidence suggests (Lidz et al.).
3. And this example takes me to my last question. Would developing a whole armamentarium of drugs to prevent cancer be an unalloyed good? Of course, it’s almost certain that these drugs will carry their own side-effects and risks, as the pharmaceutical company TV ads constantly remind us; and these risks will be experienced by large numbers of people who never would have gotten cancer in any event. And for this reason, preventing cancer in this way could end up costing much more in total than treating it, as Louise Russell pointed out back at the dawn of time (Russell).
But even setting aside these complications, I would still worry about the effects that ready availability of such drugs would have on other approaches to preventing cancer, like behavior change and environmental clean-up.
If we could all just take our pills as directed, would we in the process be eroding our sense of our personal and political responsibility for the causes of cancer, and if so would that be too high a price? I’m not sure how to answer those questions. I hope you might have some ideas.
Tom Tomlinson, PhD,
is Director of the Center for Ethics and Humanities in the Life Sciences and a Professor in the Department of Philosophy at Michigan State University.
Join the discussion! Your comments and responses to this commentary are welcomed. The author will respond to all comments made by Thursday, February 4, 2016. With your participation, we hope to create discussions rich with insights from diverse perspectives.
You must provide your name and email address to leave a comment. Your email address will not be made public.
References
- Budish E, Roin BN, Williams H. 2015. Do Firms Underinvest in Long-Term Research ?Evidence from Cancer Clinical Trials. American Economic Review. 2015; 105(7): 2044–2085. http://dx.doi.org/10.1257/aer.20131176.
- Frakt, Austin. Why Preventing Cancer Is Not the Priority in Drug Development. New York Times. Dec 28 2015. Retrieved from: http://nyti.ms/1me2V7W.
- Gemain, Denicoff, Dimond, et al. Use of the National Cancer Institute Community Cancer Centers Program Screening and Accrual Log to Address Cancer Clinical Trial Accrual. Journal of Oncology Practice. 2014 Mar; 10(2): e73–e80. http://dx.doi.org/10.1200/JOP.2013.001194
- Lidz, Appelbaum, Grisso, Renaud. Therapeutic misconception and the appreciation of risks in clinical trials. Social Science & Medicine. 2004 May; 58(9): 1689-97.
- Murthy, Krumholtz, and Gross. Participation in Cancer Clinical Trials: Race-, Sex-, and Age-Based Disparities. Journal of the American Medical Association. 2004; 291(22): 2720-2726. http://dx.doi.org/10.001/jama.291.22.2720.
- Russell, Louise. Is Prevention Better than Cure? Brookings Institution Press; 1986.
MSU Bioethics 
Thanks much for this post! I’d like to focus on one of your last points, which is that there are other approaches for preventing cancer, such as behavior change and environmental clean-up. When I saw the beginning of your post, I thought the focus would be on why so much more money is spent on developing drugs as opposed to investigating behavioral and environmental ways to prevent cancer. Thus, it was striking to me that even with the approach of drug development, the focus is primarily on late-stage, really deadly cancer. So not only do we focus primarily on drugs rather than more preventive solutions, but even the drugs we develop are very non-preventative.
It would be interesting to me to hear more of your thoughts on why there’s so much focus on drug development as opposed to the behavioral and environmental factors. I assume a major factor is that it’s much more profitable to develop drugs than to focus on other solutions. Presumably, many people also prefer to take a pill than to engage in more complicated preventive measures. In general, we probably also have a tendency to ignore problems until they manifest themselves rather than preventing them. But if you have other thoughts on the forces that push toward the emphasis on drugs, I’d welcome them.
Thanks, Kevin. There is one other factor that I left out of the post– the imperative of rescue. It feels so much more ethically compelling to rescue specific individuals from certain death than to prevent even a much larger number of deaths among a population of people. The former is an act of mercy; the latter is good public health practice. Remember the parable of the Good Samaritan? The moral would have been completely different if he’d endowed a highway patrol rather than tended to the poor man’s wounds. It doesn’t have to be “rather than,” because of course we can both rescue and put preventive measures in place. But when we are deciding how to prioritize our limited resources, the imperative of rescue will exert its pull.
Great point! I love the comparison to the Good Samaritan!
While I concur with the authors of the original article that commercialization lag is a potential issue, another issue is simply the return on investment for prevention trials. Key to FDA approval is attribution of the benefit (and potential harm) to an intervention, thus the randomized controlled trial (RCT) “gold standard”. Sample size, data collection requirements, time horizon, relative and competing risks, etc. create a very costly study with often higher risk of “no significant difference” between control and intervention group for any prevention study. So, given a competing portfolio of interventions I will likely choose something not only with a shorter time horizon but smaller sample size and greater chance of predicting a successful outcome. One can do a small pilot study for a 6-12 month treatment and get a good sense of benefit and side effects rather quickly. Any prevention study, even a pilot, will be larger, longer and riskier. And, a challenge often faced with vaccines, for example, uptake, side effects whether actually or not actually attributable to the vaccine, …become financially risky compared to a treatment. Thus, a potentially greater role for publicly-funded clinical studies with interventions, as we see in, for example, mammography and cervical cancer screening studies.
Yep, you’re absolutely right. Running a prevention study will usually take much longer, require more participants, and in other ways be more expensive to conduct with less chance of monetizing the result. This is why, as the authors of the report document, early-stage and prevention research is largely publicly funded.
Thanks Tom! Your thoughtful article started my thinking about other aspects of health care, particularly cardiovascular disease. Certainly the makers of statin medications have benefitted greatly from the sale of preventive therapies. Likewise, anti-hypertensive medications are more of a preventive action than they are a disease therapy, and are very commonly prescribed.
Why is cardiovascular care so different from cancer care? I think in part because modern informed societies have bought into the concept of ownership of cardiovascular disease. Most people accept that their “poor” lifestyle choices contribute to atherosclerosis. They have a pragmatic understanding of what happens when a heart attack occurs. We also have a high degree of confidence that modern medicines has highly successful ways of treating heart disease once established, and hence are not paralyzed with fright at the thought of it.There is significant social pressure to “do the right thing” in terms of heart health, and many believe these changes can make a difference.
Cancer “ownership” is much less common. A few cancers, like squamous cell lung cancer or skin cancers, can be deliberately attributed to behaviors like smoking or unprotected sun exposure, and like heart disease, there is significant social pressure to avoid these high risk health behaviors. But most other cancers are considered medical “terrorists” inasmuch as they strike “innocent” people unpredictably and without provocation.
There is so much fear surrounding the topic of cancer that most people don’t feel empowered to be able to prevent it. So much so, that they don’t even want to think about it, even if a history of cancer runs in their family. Likewise, most people believe that the outcomes of current cancer therapies are much less positive than those for heart disease.
The upshot is that there is much less societal interest in cancer preventive therapies, and hence, less impetus for pharmaceutical companies to expend resources in this direction.
Thanks, Carmen. I think you’re right. There is a distinctive set of beliefs and attitudes surrounding cancer that make it more difficult to approach it in the more positive vein that “prevention” implies.
But with cardiovascular disease in particular, there was another factor that encouraged the development of preventive medications like statins– the link between high blood cholesterol and heart disease. This made it possible to use lowered cholesterol levels as a surrogate marker for lowering heart disease risk. And in turn, this permitted conducting shorter– and less expensive– clinical trials than would have been possible if we had to wait for heart disease to develop.
This is why Budish, Roin and Williams recommend identification of (and FDA approval of) similar surrogate endpoints for cancer prevention– which I suspect will not be so straightforward as it was for heart disease.
Not only unlikely to be “so straightforward as it was for heart disease” there remains much debate regarding statins and prevention of heart disease particularly around measures meaningful to patients like extended life and greater quality of life.