Sodium-glucose cotransporter-2 inhibitors may modulate complement pathway activation in paroxysmal nocturnal hemoglobinuria

Abstract PNH is a bone marrow failure syndrome characterized by intravascular hemolysis and thrombotic proclivity. Complement inhibitors (CI) have changed the management of PNH patients by extending patients' life expectancy by improving anemia and decreasing the rate of thrombotic complications. Sodium-glucose cotransporter 2 inhibitors (SGLT2i), originally used for glycemic control, are reported to exhibit anti-inflammatory properties and decrease complement activation by modulating glutamine metabolism, activating AMPK, decreasing CRRY expression, and inhibiting PI3K/AKT/mTOR signaling. We noticed an improvement in hematologic parameters occasionally in PNH patients who received SGT2i and queried whether our initial clinical observation could be further substantiated by a systematic analysis of the global registry database. We conducted a retrospective, propensity score-matched (PSM) cohort study using the TriNetX Analytics Network database. Our analysis included patients ≥18 years with recorded diagnosis of PNH with (n=145) and without (n=4,267) SGLT2i between 2010 and 2022. PSM (1:1) was applied using a greedy nearest-neighbor method with a caliper of 0.1 pooled standard deviations, adjusting for demographics, PNH therapy (including complement inhibitors), comorbidities, laboratory parameters, and cardiovascular medications. Notwithstanding the limitations of registry data, primary outcomes included clinical parameters such as lactate dehydrogenase (LDH), haptoglobin, and hemoglobin (Hb). Secondary outcomes were the rate of venous thromboembolism (VTE), all-cause hospitalizations, and all-cause ICU admissions. The difference between lab values was compared using a t-test, and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated over a 1-year follow-up. Analysis of hemolytic parameters in PNH patients with and without SGLT2i showed a significant difference in mean LDH (315 vs 503), t=-1.43, p=0.03, haptoglobin (182 vs 106), t=3.84, p<0.001, and higher Hb values (12.44 vs 11.01), t=5.40, p<0.001. Accounting for clinical heterogeneity positive results of SGLT2i persisted post PSM, after PSM, 135 matched pairs (mean age was 59.9±15.5 years for patients receiving SGLT2i and mean age 60.3±15.7 years for patients not receiving SGLT2i, M/F ratio 1.17) were analyzed by measuring LDH levels (299.04 ± 197.35 vs 704.84 ± 990.08); t=-2.88, p=0.005, haptoglobin (187.60±139.80 vs 99.28 ± 91.12); t=3.09, p=0.003 and Hb (12.39 ± 2.78 vs 10.41 ± 2.53); t=5.36, p<0.001. Furthermore, SGLT2i therapy was associated with lower odds of VTE (OR=0.429, 95% CI 0.206-0.895, p =0.021),all-cause hospitalizations (OR=0.458, 95% CI 0.279-0.753, p ≤0.001), and all-cause ICU admissions (OR=0.419, 95% CI 0.215-0.820, p ≤0.01). Our study presents compelling evidence suggesting that SGLT2i may positively influence the clinical course of PNH, with the potential to delay or defer the initiation of CI therapy and enhance response rates to CIs. Despite limitations such as limited detailed clinical granularity and an observational design that precludes causality, repurposing SGLT2i for PNH management could improve outcomes, especially in patients with coexisting type 2 diabetes or heart failure. Given these potential benefits, SGLT2i should be strongly considered as part of the therapeutic approach in this high-risk patient population, warranting further investigation to confirm the safety and efficacy of SGLT2i in prospective trials.