Maurie Markman, MD
It is difficult to overstate the potential for improvements in cancer treatment outcomes based on the stunning advances in our understanding of the molecular basis of the development and progression of malignant disease and the increasing insights into the mechanisms of drug resistance.
The large number of novel antineoplastic drugs in the development pipeline is also testimony to the excitement in the biotechnology community regarding the expanding role of innovative strategies in cancer management.
Unfortunately, the current status of clinical cancer research in the United States falls far short of what is necessary to effectively and efficiently change this amazing opportunity to improve both the quantity and quality of the lives of patients with cancer into an objective reality.
Consider the simple and quite disturbing observation that less than 4% of patients with cancer participate in a clinical trial.1
This most distressing fact needs to be repeated so that its implications can be fully absorbed: the care and outcomes of fewer than 1 in every 25 patients with cancer currently contribute to the generation of the new knowledge that informs the development of fresh management strategies, defines the anticipated short-term and long-term adverse event profiles of novel approaches, and serve as the population base for oncologists employ in conversations with prospective patients and their families regarding the probability of benefit versus risk of harm associated with various treatment options.
Moreover, the 1 in 25 patients who participate in clinical trials fall short of depicting the real world of individuals who require management of their cancers, as multiple analyses reveal a lack of representation of the elderly and those with common comorbidities who have the potential to meaningfully alter the risk-to-benefit ratio for a given strategy. Further, in the absence of such essential data there should surely be serious questions regarding the optimal starting doses or schedules for novel agents and concerns about important contraindications in a given setting such as pre-existing mild/moderate renal, cardiac, or pulmonary dysfunction that remain unknown in the absence of prior clinical trial experience.
And if this picture is not distressing enough, one must add the documented, often inane bureaucracy associated with the current clinical trials paradigm. 2,3
Does the incredible paperwork required to document why a complete blood count was obtained on day 10 instead of day 8 really add to the validity of the results of a clinical trial? Does the information obtained justify the added costs? This abbreviated list of the multiple serious concerns inherent in the current clinical cancer research domain and its negative impact on the efficient evaluation and development of useful therapeutics have been extensively discussed in the oncology literature—but little has changed.
Although many reasons can be advanced to explain this unsatisfactory situation, there is a stunning lack of alignment among interested participants in the research domain regarding even the foundation of the problem requiring resolution.
Overall Survival Poses Problems as Endpoint
Consider, for example, that some members of the purist academic/university collective continue to believe that the only valid endpoint for approval of a new pharmaceutical agent or strategy is the finding of a statistically significant improvement in overall survival in a phase III randomized trial.
Yet, for several well-established reasons, it should be appreciated that having this endpoint as the sole measure of clinical benefit in an individual clinical trial is highly problematic.
First, advanced cancers are increasingly being recognized as chronic disease processes, where survival after progression on a given trial may legitimately be measured in years rather than a few months. Beneficial treatments administered after this date may substantially impact a patient’s ultimate survival completely independent of treatment on the study in question.
Second, as the clinical behavior of individual cancers originating from the same organ are based on unique biological characteristics often revealed by the presence of specific molecular markers—for example, EGFR
rearrangements, and KRAS
mutations in lung cancer—much smaller patient populations will be available to define clinical utility. Under these circumstances, how many years will those desiring the answer to the question of risk versus benefit, including patients, physicians, and society, be required to wait to determine if a novel strategy improves survival in a randomized trial?