DSAI-05 GABAPENTIN USE IS ASSOCIATED WITH REDUCED RISK OF DEVELOPING BRAIN METASTASES IN MEDICARE BENEFICIARIES WITH LIMITED STAGE SMALL CELL LUNG CANCER: A RETROSPECTIVE COHORT STUDY

Abstract

Brain metastasis (BrM) in small-cell lung cancer (SCLC) is common and presents a major clinical challenge, with limited therapies and poor prognosis. We previously showed that gabapentin, a commonly prescribed medication for pain control, is associated with improved survival in glioblastoma. To assess whether gabapentin may also have clinical benefit in secondary CNS malignancies, we assessed whether gabapentin reduces the risk of developing BrM in limited stage SCLC (LS-SCLC). We used the NCI’s Surveillance, Epidemiology, and End Results data linked to traditional Medicare to identify LS-SCLC patients diagnosed between 2007-2019 (followed through 2020) who received chemotherapy. Patients were excluded if ≤ 65 years at diagnosis, not enrolled in Medicare Parts A/B/D for equal time, missing follow-up, or diagnosed by autopsy/death certificate. To ascertain the true effect of gabapentin, we also compared the effects of duloxetine and pregabalin, which are prescribed for similar indications. We utilized an intent-to-treat protocol (≥2 claims considered “on treatment”, with ≥1 claim within 6 months of diagnosis). Covariates were balanced between treatment groups using propensity score matching with inverse-probability of treatment weights. Hazard Ratios (HR) of BrM adjusted for competing risk of death, demographic and clinical covariates was estimated using Fine-Gray regression modelling. We identified 3,050 individuals with LS-SCLC at diagnosis. Of these, 889 (29%) developed BrM by the end of the study. There were 240 (7.9%), 65 (2.1%), and 44 (1.4%) individuals treated with gabapentin, duloxetine, and pregabalin, respectively. Individuals receiving gabapentin had a 51% lower rate of developing BrM (95% confidence interval [95%CI]=0.37-0.67, pvalue <0.001) compared to no treatment. This was lower than both duloxetine (HR = 0.58, 95%CI=0.35-0.97, pvalue:0.03) and pregabalin (HR = 0.63, 95%CI=0.34-1.17, pvalue=0.15). Gabapentin use within 6 months of LS-SCLC diagnosis was associated with meaningful reduction in BrMs risk. A prospective clinical trial is warranted to further evaluate this association.