Effect Modification by Hydroxychloroquine on the Nephrolithiasis Preventive Effect of SGLT2 Inhibitors

Background: Recent studies have highlighted that SGLT2 inhibitors (SGLT2i) can reduce urolithiasis. This effect is theorized to be due to the stimulation of autophagy by SGLT2 inhibitors, distinct from other antidiabetic drugs. However, this hypothesis lacks validation in clinical settings.

Objectives: To explore whether the inhibition of autophagy, through exposure to hydroxychloroquine (HCQ), diminishes the effect of SGLT2i in preventing urolithiasis.

Methods: A retrospective cohort study was conducted, utilizing Taiwan's National Health Insurance Research Database. The study included type 2 diabetic patients aged over 40, initiating SGLT2i or Dipeptidyl peptidase-4 inhibitors (DPP4i) between 2016 and 2020. Patients with simultaneous SGLT2i or DPP4i prescriptions or end-stage renal disease were excluded. Two cohorts were formed: patients without a history of urolithiasis and those with such a history. The index date was the first prescription date of SGLT2i or DPP4i within the study period. HCQ was selected as the autophagy inhibitor for stratification, testing the effects of SGLT2i and DPP4i on urolithiasis occurrence. HCQ exposure was defined as any prescription within one year prior to the index date. A multivariable Cox regression model was utilized for confounder adjustment, using DPP4i users without HCQ exposure as a reference group to calculate relative hazard ratios (HR) of urolithiasis.

Results: The study identified 84,749 SGLT2i initiators and 437,325 DPP4i initiators without a history of urolithiasis. Compared to the reference group, DPP4i users with prior HCQ exposure showed no significant risk alteration (HR 1.12, 95% CI 0.90-1.39). SGLT2i users without HCQ exposure had a significantly lower risk of developing urolithiasis (HR 0.82, 95% CI 0.78-0.86). However, this risk reduction was not significant in SGLT2i users exposed to HCQ (HR 0.89, 95% CI 0.44-1.77). The impact of HCQ on SGLT2i's urolithiasis prevention was statistically significant (p for interaction=0.009). Similar findings were observed in patients with a history of nephrolithiasis.

Conclusions: Our findings indicate that the nephrolithiasis preventive effects of SGLT2i are mitigated upon exposure to HCQ, an autophagy inhibitor. This supports the clinical significance of autophagy stimulation by SGLT2i in reducing the risk of urolithiasis.