Influence of background medical therapy on efficacy and safety of dapagliflozin in patients with heart failure with improved ejection fraction in the DELIVER trial

Aims: The Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure (DELIVER) trial demonstrated the sodium–glucose cotransporter 2 inhibitor dapagliflozin to be beneficial in patients with symptomatic heart failure (HF) with improved ejection fraction (HFimpEF; those with prior left ventricular ejection fraction ≤40% that had improved to >40% by enrolment). Whether this benefit differs by background medical therapy is unclear. The current study aims to determine the efficacy and safety of dapagliflozin among patients with HFimpEF by background medical therapy. Methods and results: Treatment effects on the primary endpoint (worsening HF or cardiovascular death) were assessed by number of background HF medical therapies (angiotensin-converting enzyme inhibitor/angiotensin receptor blocker/angiotensin receptor–neprilysin inhibitor, evidence-based beta-blocker, and mineralocorticoid receptor antagonist). Among the 6263 patients randomized in DELIVER, 1151 (18%) had HFimpEF. Of those, 21% of patients were on 0–1 therapies, 44% were on two therapies, and 35% were on three therapies. During 2.3 years of median follow-up, the incidence rate of the primary outcome was 9.7, 8.8, and 8.4 per 100 person-years for patients on 0–1, 2 and 3 HF medications at baseline, respectively. Treatment effects with dapagliflozin on the primary outcome may be greater in patients with HFimpEF on 0–1 therapies at baseline (p_interaction = 0.09), driven mostly by a significant interaction for HF hospitalization (p_interaction = 0.023) with no evidence of effect modification for cardiovascular death (p_interaction = 0.65). Treatment effects of dapagliflozin on the primary outcome were, however, consistent when assessed across the modified Heart Failure Collaboratory Medical Therapy Score integrating both therapeutic use and dosing (p_interaction = 0.39). The use of dapagliflozin was not associated with changes in use or doses of background HF therapies, and among patients on three HF medications at baseline, the addition of dapagliflozin did not lead to higher adverse events. Conclusions: In patients with HFimpEF, the safety and efficacy of dapagliflozin were largely similar by background use and dosing of HF medical therapies. The benefit of dapagliflozin in reducing HF events tended to be greater in those patients on 0-1 medications at baseline. Among patients already on three HF medical therapies, the addition of dapagliflozin was safe without requiring de-escalation of other therapies.