Abstract

Introduction: Previous research has suggested thiazolidinediones (TZDs) may be associated with certain solid tumors. We examined incidence rates over time for patients with solid tumors who received TZD treatment for diabetes. Methods: We identified medication use and diagnosis codes that were aggregated to phecode disease classifications derived from the nationwide Veterans Administration Health Record System from 2000 to 2021. We identified 148,139 patients who had ≥2 diabetes diagnoses and had no previous cancer diagnosis. Among these, 8981 subjects had ≥4 years of TZD exposure. We then identified subjects with ≥2 diagnosis codes for solid tumors including bladder (n = 3987), breast (n = 632), colorectal (n = 5139), esophageal (n = 482), glioma (n = 591), lung (n = 5142), melanoma (n = 1896), pancreatic (n = 726), prostate (n = 11,884), renal (n = 3145), testicular (n = 369), and thyroid (n = 513). We used multivariable Cox proportional hazards regressions to measure associations between TZD use and cancer incidence. TZD use was modeled as a time-varying covariate from the first to last prescription of TZD medication, and analyses were adjusted for age at diabetes diagnosis, self-reported race, self-reported ethnicity, sex, body mass index, and cancer site-specific polygenic risk scores. Results: Long-term (≥4 years) exposure to TZDs was significantly associated with increased risk of developing prostate cancer (HR = 1.24, p < 0.001) and decreased risk of developing lung (HR = 0.58, p < 0.001), bladder (HR = 0.51, p < 0.001), and renal cancer (HR = 0.75, p = 0.003). Conclusions: Decreased risk of developing several solid tumors (lung, bladder, and renal) indicate that TZDs may be strong candidates for drug repurposing strategies to manage these types of cancer. These results warrant replication attempts in external datasets.

IPC Classification

Keywords