Association Between Hypoglycemia Agents and Long-term Survival Outcomes for Patients with Non-muscle-invasive Bladder Cancer Treated with Intravesical Bacillus Calmette-Guérin Immunotherapy.

BACKGROUND AND OBJECTIVE

There is a lack of data on the impact of hypoglycemia agents, especially metformin, on prognosis for non-muscle-invasive bladder cancer (NMIBC). Our aim was to investigate the association between hypoglycemia agents, especially metformin, and long-term survival outcomes for patients with NMIBC treated with bacillus Calmette-Guérin.

METHODS

All patients with NMIBC treated with intravesical BCG therapy from 2001 to 2020 were identified in a territory-wide database in Hong Kong. Patients were stratified into two groups according to whether or not they were taking a hypoglycemia agent at BCG treatment initiation. We analyzed data for overall survival (OS), cancer-specific survival (CSS), recurrence-free survival (RFS), and progression-free survival (PFS) using the Kaplan-Meier method. Multivariable Cox regression analysis was used to adjust for potential confounding factors and estimate hazard ratio (HRs) and 95% confidence intervals (CIs). Subgroup analyses were conducted to assess the specific influence of metformin on survival outcomes.

KEY FINDINGS AND LIMITATIONS

Of 2602 patients with NMIBC treated with intravesical BCG, 19.5% (n = 507) were taking a hypoglycemia agent at BCG initiation (treatment group) and 80.5% (n = 2095) were not (control group). At median follow-up of 11 yr, Kaplan-Meier analysis revealed a significant difference in OS between the groups (p < 0.01), but not in CSS (p = 0.36), RFS (p = 0.19), or PFS (p = 0.05). Subgroup analysis comparing outcomes for patients taking metformin, patients taking a hypoglycemia agent other than metformin, and control subjects revealed significant differences in OS (p < 0.01) and RFS (p = 0.02), but not in CSS (p = 0.59) or PFS (p = 0.08). Multivariable Cox regression analysis identified metformin-based treatment for hypoglycemia as an independent risk factor for RFS (HR 1.22, 95% CI 1.02-1.46), whereas hypoglycemia agents other than metformin were not significantly associated with RFS (HR 0.71, 95% CI 0.47-1.06).

CONCLUSIONS AND CLINICAL IMPLICATIONS

Metformin-based hypoglycemia treatment was an independent risk factor for RFS in BCG-treated NMIBC. Hypoglycemia treatment with an agent other than metformin was not related to long-term survival outcomes.

PATIENT SUMMARY

We investigated the relationship between treatment for high blood sugar and long-term survival for patients with intermediate-risk or high-risk non-muscle-invasive bladder cancer. The patients had received BCG (bacillus Calmette-Guérin) treatment in Hong Kong for their bladder cancer over the past two decades. Our results show that metformin, but not other drugs used to treat high blood sugar, was associated with poorer survival free from bladder cancer recurrence for these patients.