[Skip to Navigation]
Sign In
Comment & Response
December 2018

Moving Precision Oncology Forward Amid Myths and Misconceptions

Author Affiliations
  • 1Division of Cancer Medicine, Department of Investigational Cancer Therapeutics (Phase I Clinical Trials Program), The University of Texas MD Anderson Cancer Center, Houston
  • 2Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston
JAMA Oncol. 2018;4(12):1788-1789. doi:10.1001/jamaoncol.2018.5064

To the Editor In the article by Marquart et al,1 the proportion of patients with advanced cancers eligible for and benefiting from US Food and Drug Administration–approved genome-driven treatments between 2006 and 2018 is estimated. A key finding is that such strategies would only benefit 6.62% of patients in 2018. These results need to be interpreted with caution, especially because they are approximated using multiple retrospective data assumptions and sensitivity analyses that do not consider key clinical trial details. For example, the 16% response rate of trastuzumab was based on a trial that included 23% of patients with moderate (2+) human epidermal growth factor receptor 2 (ERBB2/HER2) expression breast cancers who are unlikely to respond to treatment unless they have HER2 amplification, and the trial assessed trastuzumab monotherapy rather than chemotherapy combinations where trastuzumab provides the most benefit, with response rates ranging from 42% to 72%.2-4 Progress in cancer medicine is achieved by building on Food and Drug Administration–approved drugs through improved patient selection and combination strategies. A major constraint of the study by Marquart et al1 is that such evolving advances are not considered (eg, a 10% response rate of gefitinib in non–small cell lung cancer was based on an unselected patient population5). These historical statistics do not reflect today’s clinical practice and are potentially misleading.

Add or change institution