Concerns regarding use of patient-reported outcomes in biomarker studies of chemotherapy-induced peripheral neuropathy

Abstract

Clinician-rated toxicity data has been systematically collected within oncology clinical research using the National Cancer Institute’s CTCAE scale, providing estimates of the occurrence and severity of toxicity from cancer treatment. CTCAE is being supplemented by collection of patient-reported outcome (PRO) toxicity within clinical research and clinical practice, where PRO has demonstrable benefits. There is general agreement that PRO data is more sensitive and reliable than CTCAE data, particularly for subjective adverse effects. Based on this premise, researchers have begun to use PRO toxicity data collected within prospective clinical trials as the primary endpoint to discover pharmacogenetic and other predictive biomarkers of treatment-related toxicity. This perspective raises caution about the superiority of PRO data to CTCAE data for biomarker research, particularly in regards to chemotherapy-induced peripheral neuropathy (PN). The reader is provided an introduction to PRO and their integration into clinical research and practice, comparisons of PN data collected by PRO and CTCAE, examples of attempts to use PRO PN data for biomarker discovery, and evidence suggesting that PRO may not be superior to CTCAE for PN biomarker studies. The perspective concludes with a proposed approach for empirically testing whether PRO or CTCAE data is the better option for use in PN biomarker research, which can serve as a model for similar comparisons within other treatment-related toxicities.