Presentation Authors: Rishi R Sekar*, Wayne G Brisbane, Sarah K Holt, Brian R Winters, Evan Y Yu, John L Gore, Jonathan L Wright, George R Schade, Seattle, WA
Introduction: Recent institutional series suggest that diabetes mellitus (DM) as well as the DM medication metformin may be associated with adverse outcomes in various malignancies, including non-muscle invasive bladder cancer (NMIBC). We set out to evaluate the relationship between DM and metformin use on disease progression and disease specific survival (DSS) in a cohort of Medicare beneficiaries with NMIBC undergoing intravesical Bacillus Calmette-Guerin (BCG) therapy.
Methods: Using the Surveillance, Epidemiology, and End Results (SEER) Medicare database, we identified patients with high-risk NMIBC who received at least one dose of intravesical BCG. We defined disease progression as receipt of cystectomy, systemic chemotherapy, radiation, SEER stage progression and/or disease specific death. The Diabetic Comorbidity Severity Index was used to define DM severity as mild (0-1 comorbidity) and severe (>2 comorbidities). The relationship between DM severity, disease progression, and DSS was evaluated using the Kaplan-Meier method and multivariable Cox proportional hazards models adjusting for age, race, gender, pathologic stage, and BCG treatment (>/=5 doses vs < 5 doses). A subset analysis was then performed on DM patients with Medicare part D coverage to evaluate the prognostic impact of metformin therapy on NMIBC outcomes.
Results: Our cohort consisted of 8,774 patients, of which 2,546 had mild DM and 1,017 had severe DM with a median follow-up of 39 months. In adjusted cox proportional hazards models, patients with severe DM had significantly increased risk of disease progression (HR 1.17, 95% CI 1.00 - 1.38) and worse DSS (HR 1.21, 95% CI 1.01 - 1.45) vs. controls, while mild DM was associated with significantly improved DSS (HR 0.87, 95% CI 0.75 -1.00) but not progression (HR 1.08, 95% CI 0.95-1.22) vs. controls (see Figure). Metformin use was observed in 1,136 patients accounting for 54% of mild DM patients and 43% of severe DM patients. On sub-analyses, receipt of at least one year of metformin therapy was associated with significantly improved DSS (HR 0.43, 95% CI 0.18 - 1.00) in patients with severe DM, but patients with mild DM had no significant difference (HR 0.67, 95% CI 0.36-1.23).
Conclusions: Medicare beneficiaries with NMIBC and severe DM are at increased risk of disease progression and disease specific death compared to controls. Long-term use of metformin in this population of DM patients may help reverse the negative impact of severe DM on NMIBC outcomes. Further study into the role of DM and metformin on NMIBC outcomes is needed.
Source of Funding: Drive for the Cure Northwest