Monday, March 24, 2014
Concerns for Cooperative Groups
An article in the current issue of The ASCO Post (15 March 14) addresses the NCI’s cooperative group program, and includes some interesting comments by Dr. Richard Schilsky and Dr. Laurence Baker, both former heads of major cooperative groups.
Their comments reflect long-standing problems in the NCI-sponsored clinical trials program. I attended and spoke at the Institute of Medicine hearings that helped prepare for the IOM’s 2010 report to the NCI. During my day there, basically two opinions were expressed: The first was the tweaking, consolidating, and/or "more money” approach -- despite the fact that more money was never a likely prospect because of the economy and NCI’s very tight budget at that time.
The other approach was to step back and hire independent experts (like Rand and some universities) to examine why and how the current model has become so inefficient technically and scientifically. An RFP could then be sent out to develop several new models for the program that could be tested to see if they can overcome at least some of the problems.
I was apparently the only proponent of this approach. I offered that idea because the clinical trials program is a shopworn system with deeply ingrained interests accustomed to the old model, which makes substantive changes extremely difficult. But the cancer research landscape has changed dramatically in the past decade.
Furthermore, as Dr. Schilsky pointed out “…NCI’s creation of the National Clinical Trials Network will revert the power to a small number of individuals on steering committees within the NCI. They will decide what clinical trials will be done. It’s a very different model with a narrower vision, and I’m not confident [in its success].”
I agree. The track record of the NCI making dramatic changes in the cooperative groups has been short of vision and a disregard for unconsidered side effects that damage the system. The failure to harness the research creativity available across the country, in the context of a more efficient model, would be a giant step backward.