There has been considerable discussion within the oncology literature during the past several years regarding the level of evidence required to consider a new antineoplastic agent an acceptable "standard-of-care" in routine disease management.
Editor-in-Chief of OncologyLive
Senior vice president for Clinical Affairs and National Director for Medical Oncology Cancer Treatment Centers of America, Eastern Regional Medical Center
It took five years to recruit 13 patients for a multicenter clinical trial investigating imatinib in -mutated metastatic melanoma. If a phase III trial were needed to state that imatinib is an acceptable standard-of-care therapeutic option for such patients, how many decades would be required to obtain the answer?
There has been considerable discussion within the oncology literature during the past several years regarding the level of evidence required to consider a new antineoplastic agent an acceptable “standard-of-care” in routine disease management. In the not-so-distant past, the mantra in oncology was that a randomized trial should be conducted whenever there was a discussion of defining management paradigms. But with the rapidly evolving understanding of the molecular basis of cancer, it has become clear to most observers that this mandate is simply no longer tenable
Objectively, 20% of non-small cell lung cancer patients (NSCLC) (ie, documented EGFR mutation) represent a sufficient patient population in which to undertake and complete in a timely manner a phase III randomized trial, as would 50% of individuals with metastatic melanoma (ie, documented BRAF mutation). But is it rational to adhere to that view if one were contemplating a registration trial involving 2%-4% of patients with metastatic gastric or ovarian cancer?
Consider, for example, several highly provocative recent examples from the peer-reviewed oncology literature suggesting substantial clinical activity for an FDA-approved antineoplastic agent in an uncommon/ rare molecularly defined patient subset.
In a multicenter phase II trial examining the activity of imatinib in patients with metastatic melanoma whose cancers contained an abnormality (mutation or amplification) in the KIT proto-oncogene, Hodi et al noted a striking objective response rate of 54% (n = 7) among the 13 patients with a KIT mutation but no responses among the similarly small group of individuals with amplifications.1 Within the mutation-positive patient population, the overall disease control rate was reported to be 77%.1
KIT mutations in melanoma represent a small piece of the mutation pie as described in current research, according to My Cancer Genome, an online resource that Vanderbilt-Ingram Cancer Center manages at www.mycancergenome.org. Snapshot of Somatic Gene Mutation Frequency in Melanoma 22761 INFI Percentages are approximate and do not add up to 100%.
However, it must be noted here that this multicenter study required five years to complete accrual of this limited sample size. So, the specific question to be addressed is as follows: If one were to declare that a phase III trial will be needed to state that imatinib is an acceptable standard-of-care therapeutic option in patients with KIT mutation-positive metastatic melanoma, how many decades would be required to obtain the answer?
Returning to the previously noted lung cancer example, what if the subpopulation being considered was only 1.7% of the entire NSCLC population rather than 20%? In fact, a recent report noted that, of the cancers of 3800 patients with NSCLC, 1.7% (n = 65) were found to demonstrate mutations in HER2.2 And, of the 16 previously treated patients with a HER2 mutation who received a total of 22 lines of anti-HER2-targeting drugs, an objective response rate of 50% (11 partial responses) was observed. If one considers the patients who achieved evidence of disease stabilization, an overall disease control rate of 82% was noted in this limited patient sample.2 Finally, the disease control rate associated with chemotherapy plus trastuzumab-based treatment (n = 15) was 93%, and of the 3 patients who received single-agent afatinib, all (100%) achieved evidence of response or stabilization of the malignant process.
These encouraging data involving a limited number of pretreated patients with NSCLC are consistent with the new paradigm of “N of 1” experience. Should it be required that a randomized trial that examines the clinical utility of anti-HER2 therapy in HER2 mutation-positive NSCLC be initiated, completed, and “favorable” results reported before a clinician is permitted to treat a patient with such therapy “off study” in this clinical setting? If the answer to this question continues to be “Yes,” it is critical to appreciate that the likelihood of such a trial ever being conducted is extremely low and, even assuming the scenario actually occurred, the wait for the results would be painfully long.
Further, as we continue to learn more about molecularly defined resistance mechanisms in lung and other cancers, it is highly likely that multiple unique subsets will be defined, each requiring specific antineoplastic strategies that have the potential to meaningfully delay the time to subsequent disease progression and favorably impact survival.3
Is it rational to suggest that the only way a given strategy can be employed as a component of “standard-of-care” would be to initiate, complete, and report the results of a randomized clinical trial? Acknowledging the reality of the current and future revolutionary developments in our understanding of the molecular basis of cancer, it is imperative that the cancer research establishment finds alternative approaches to define “acceptable evidence.” Our patients deserve no less.
Maurie Markman, MD, editor-in-chief, is senior vice president for Clinical Affairs and national director for Medical Oncology at Cancer Treatment Centers of America. firstname.lastname@example.org