SGLT2 inhibitors may keep ADT working longer in prostate cancer

In prostate cancer patients receiving androgen deprivation therapy (ADT), concurrent use of SGLT2 inhibitors appears to delay hormone therapy failure.

In a target trial emulation study, Hong Kong-based researchers found that time to ADT failure was significantly longer for SGLT2 inhibitor users than nonusers (hazard ratio [HR], 0.63, 95 percent confidence interval [CI], 0.41–0.95; p=0.03), with a 10-year cumulative incidence difference of –11.1 percent (95 percent CI, –18.6 to –0.9). [JAMA Oncol 2026;doi:10.1001/jamaoncol.2025.5869]

The same was true for time to failure of next-generation hormonal agents (NHAs), which was significantly delayed for SGLT2 inhibitor users vs nonusers (HR, 0.44, 95 percent CI, 0.20–0.97; p=0.04). The difference in cumulative incidence at 10 years was –8.4 percent (95 percent CI, –24.3 to –0.1).

The delay in ADT failure did not translate to significant improvements in survival outcomes, although the point estimates suggested a favourable trend among SGLT2 inhibitor users (disease-specific survival [DSS]: HR, 0.60, 95 percent CI, 0.20–1.85; p=0.37; overall survival [OS]: HR, 0.70, 95 percent CI, 0.42–1.16; p=0.17). Meanwhile, the limited sample size of patients taking NHAs (n=215) and number of survival events (DSS, 41 events; OS, 123 events) precluded the researchers from detecting differences in DSS and OS between SGLT2 inhibitor users and nonusers.

“In subgroup analyses, the association between SGLT2 inhibition and delayed ADT failure held up across different patient strata. The beneficial trends were also observed when comparing patients receiving SGLT2 inhibitors to patients without diabetes or to patients taking other glucose-lowering drugs separately,” said lead researcher Dr Ruofan Shi from The University of Hong Kong, Hong Kong, China, and colleagues.

It is possible that the beneficial effect may be linked to the “SGLT2 inhibitors themselves rather than simply reflecting differences in underlying diabetes status,” Shi and colleagues explained. “Meanwhile, we did not detect a substantial difference between dapagliflozin and empagliflozin in clinical outcomes, suggesting the consistent anti–prostate cancer effects within the class of SGLT2 inhibitors.”

Considering that SGLT2 inhibitors are often prescribed together with metformin, Shi and colleagues evaluated the effect of metformin monotherapy. No improvement in disease progression was observed, but there was a significant increase in overall survival (HR, 0.59, 95 percent CI, 0.42–0.83; p=0.002). This indicates that “SGLT2 inhibitors may exert an independent protective effect during hormonal therapy for prostate cancer,” they said.

The study included 14,223 prostate cancer patients (median age at enrolment 74 years), with a median follow-up of 66 months. ADT failure occurred in 6,252 patients (44 percent), with a median time to failure of 55 months. Among 3,358 patients who received subsequent NHA, 1,932 (57.5 percent) experienced treatment failure, with a median time to failure of 50 months. Prostate cancer–specific mortality was 16.8 percent, and the overall mortality was 44.4 percent.

Shi and colleagues called for prospective trials to validate the findings and assess their potential clinical applicability.