Almost one-half of patients treated with radiotherapy (RT) for oropharyngeal cancer require chronic opioid use to manage oral mucositis (OM) pain. 1 Given the paucity of level 1 evidence for OM pain management, current guidelines allow consideration of prophylactic gabapentin. 2,3 However, the optimal dose remains unclear. In this comparative effectiveness study, we performed a secondary analysis of 2 clinical trials conducted by this research team, hypothesizing that a higher dose of gabapentin would be associated with less opioid use and feeding tube (FT) placement.
MethodsEligible patients in both studies received concurrent chemoradiotherapy for nonmetastatic squamous cell carcinoma of the head and neck and prophylactic oral gabapentin (titrated to 900 mg vs 2700 mg daily in 1 study 4 and 3600 mg daily in the other study). Using the 3600 mg cohort as the reference group, we performed a multivariable competing risks analysis to evaluate time to first use of opioids and a logistic regression analysis to assess FT placement, adjusted for baseline characteristics. This study followed the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) reporting guideline. Additional information is available in eMethods in the Supplement.
ResultsAmong 92 patients, the median age was 62.1 years (IQR, 55.4-66.4 years); 82 patients were male (89.1%), and 10 (10.9%) were female. Baseline cohort characteristics were well balanced (Table ).Most patients tolerated gabapentin per protocol. For example, adverse effects from gabapentin, such as dizziness and drowsiness, were only reported in 1 patient (3.1%) in the 3600 mg cohort (grade 3 dizziness) and 2 patients (6.3%) in the 3600 mg cohort (gabapentin was discontinued because of nausea and vomiting partly associated with RT and low adherence to oral hygiene).The multivariable competing risks model revealed the time to first opioid use for additional pain control was greatest in the 3600 mg cohort (Figure). The proportion of patients requiring opioids during RT was also smallest in the 3600 mg cohort (12 patients [37.5%]) vs the 900 mg cohort (27 patients [93.1%]) and the 2700 mg cohort (19 patients [61.3%]; P < .001). Multivariable logistic regression analysis revealed that, compared with the 3600 mg cohort (3 patients [9.4%]), the 2700 mg cohort (14 patients [45.2%]) had significantly greater odds of FT placement during RT (adjusted odds ratio, 9.9; 95% CI, 2.1-75.7; P = .009); however, the odds were not significantly greater in the 900 mg cohort (6 patients [20.7%]; adjusted odds ratio, 3.6; 95% CI, 0.7-28.1; P = .15).
DiscussionThis study is, to our knowledge, the first evaluation of prospective data to suggest that higher doses of gabapentin (900-3600 mg daily) are well tolerated and associated with delayed time to first