'Think Different' About Clinical Cancer Research

David L. Graham, MD


June 20, 2016

The American Society of Clinical Oncology (ASCO) Annual Meeting ended 2 weeks ago, so I have had time to digest and internalize the sessions I attended and the data presented. There were a number of interesting presentations in the genitourinary (GU) arena, and certainly the data on atezolizumab in bladder cancer were impressive. In fact, they were impressive enough for the US Food and Drug Administration (FDA) to approve its use before they were even presented.

What was just as impressive to me were the data presented on therapies after their FDA approvals. Some of these trials, particularly in therapies for prostate cancer, were mandated by the FDA upon approval; another significant report, in advanced renal cell carcinoma, simply extended the survival data in patients treated with nivolumab to 4-5 years. Seeing a 30%-plus survival rate in these patients was a surprise, particularly because a number of these folks had progressed through that line of therapy. The end result is that we have an ever enlarging armamentarium of therapies to fight GU malignancies, and now we have to decide how to choose the optimal therapies.

we have an ever enlarging armamentarium of therapies to fight GU malignancies, and now we have to decide how to choose the optimal therapies.

These ideas were mulling around in my head as I attended the hallmark presentation by Vice President Joe Biden. In his speech, he emphasized the need to work collaboratively and to think differently about how to find the answers. As I thought about his exhortation to us as cancer doctors, I started to apply his approach to the problem of second-line therapies in renal cell cancer.

We have upwards of eight approved therapies for the second-line therapy of hypernephroma. Which one do we use? Should cost be a consideration? Do we fall back on the drug that we are most comfortable with? These are ideas that have been used in practice before, for good or bad. The purists would say that we need randomized phase 3 trial data to know which is the best. This would require the equivalent of a "World Cup" tournament of trials that would take years and millions of dollars to complete, by which time, given the rate of therapies being developed, we would probably have newer therapies and still be facing the same question. This is just one line of therapy in one malignancy. Expand this idea out to the full range of cancers and the task becomes truly daunting.

So, what other options are there to find these answers? Maybe part of the solution was suggested by the long-term follow-up of nivolumab I mentioned earlier. The use of guidelines and pathways is often encouraged as a sign of following evidence-based medicine, but do they really go far enough? Most pathways exist as a "read-only" document. Physicians find the appropriate situation and follow that suggestion, and then document that in their note. There is no registration of that patient in the guidelines, and because of this, the opportunity to follow how these patients do in aggregate does not exist. At Levine Cancer Institute, where I practice, we have initiated a pathways mechanism that allows for patient registration and the ability to follow outcomes over time. We're just one institution, though, and getting an adequate number of patients to answer large questions will be limited.

In 1997, Apple Computer started an ad campaign with the slogan "Think Different." It is an idea that we may wish to strongly consider here. If pathways and guidelines were expanded to include tracking of outcomes data, or, alternatively, if the ASCO CancerLinQ initiative was broadly adopted, we could practice along our usual current patterns and probably get significant outcomes data in the time that it takes to get one clinical trial developed, accepted, accrued to, and analyzed. In 2009, the Mayo Clinic reported a statistical evaluation showing that the RECIST 2 criteria regarding the number of lesions to be measured were unnecessarily rigorous. It would be interesting to have a statistician analyze whether prospective collection of unselected outcomes data could potentially be "good enough" to help limit broad lists of potential therapies.

It would take an open mind to even consider the question. It would take collaboration between institutions and organizations to create. It would take some willingness by physicians to forego trials, and it would take willingness on the part of the National Institutes of Health and the National Cancer Institute to consider a change in funding ideas. In other words, it would take exactly the ideas that Vice President Biden urged us to consider.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: