Hemoglobin, Body Mass Index, and Age as Risk Factors for Paclitaxel- and Oxaliplatin-Induced Peripheral Neuropathy

Abstract

Importance: Chemotherapy-induced peripheral neuropathy (CIPN) is a debilitating adverse effect of neurotoxic cancer treatments including taxanes and platinum agents. Limited knowledge exists of potential prechemotherapy factors associated with CIPN development. Objective: To identify the association of pretreatment blood-based and clinical factors with CIPN persistence in patients who received paclitaxel or oxaliplatin. Design, Setting, and Participants: This cohort study assessed pretreatment blood-based clinical factors and demographic characteristics of 333 patients treated with paclitaxel and oxaliplatin chemotherapy at urban multicenter cancer clinics and academic institutions in Australia between September 2015 and February 2020. Comprehensive neuropathy assessments were undertaken 3 to 12 months posttreatment. Posttreatment CIPN severity was compared with blood-based factors within 30 days prior to commencing chemotherapy. Data were analyzed between March and December 2020. Exposures: Paclitaxel or oxaliplatin chemotherapy. Main Outcomes and Measures: CIPN was measured using composite neurological grading scales, nerve conduction studies, and assessments of fine motor skills (grooved pegboard test), sensory function (grating orientation test and 2-point discrimination), and patient-reported outcomes. Independent samples t tests and Mann-Whitney U tests with post hoc Bonferroni correction were used to compare CIPN between patients according to blood-based factor normative ranges. Linear regression was used to identify blood-based and clinical associations with CIPN development. Results: The study included 333 participants (266 [79.9%] women; median [interquartile range] age, 58 [18] years) who were consecutively recruited and referred (228 treated with paclitaxel, 105 treated with oxaliplatin; 138 [41.4%] with breast cancer, 83 [24.9%] with colorectal cancer). Most participants had grade 1 CIPN or higher (238 [71.5%] participants). Participants with low hemoglobin pretreatment had worse CIPN posttreatment (median [IQR] composite neurological grading scale score, 5 [2-8] vs 4 [1-6]; P =.002; grooved pegboard mean [SD] time, 84.2 [28.7] vs 72.9 [21.1] seconds; P =.002; grating orientation task, 4.8 [2.8] vs 3.9 [1.8] mm; P =.03; 2-point discrimination, 45% vs 28%; P =.01), with no other impairments outside normative ranges associated with CIPN. In the multivariable model, several factors were associated with worse CIPN (F4,315= 18.6; P