Carvedilol for prevention of heart failure in anthracycline-exposed survivors of childhood cancer: Results from COG ALTE1621

10013 Background: Carvedilol improves cardiac function in patients with heart failure (HF), but remains untested as cardioprotective therapy in long-term childhood cancer survivors at risk for HF due to high-dose anthracycline exposure. We hypothesized that low-dose carvedilol would be safe and efficacious in reducing anthracycline-related heart failure risk in long-term survivors without evidence of HF. Methods: This was a randomized, double-blind, placebo-controlled phase 2B trial (NCT02717507) conducted at 28 sites. Eligibility included ≥250 mg/m 2 cumulative anthracycline exposure by age 21 years (y), ≥2y from completion of cancer treatment, and preserved systolic function (ejection fraction [EF] ≥55% and/or fractional shortening [FS] ≥28%) at enrollment. Patients were randomized 1:1 to take low-dose carvedilol (12.5 mg/day) or placebo for 2y; dose up-titration started at 3.125 mg/day. The primary objective was to determine the impact of carvedilol on echocardiographic indices of HF risk: standardized left ventricular (LV) wall thickness/dimension ratio (LVWT/Dz; primary endpoint), LV end-systolic wall stress (ESWS), LV chamber diameters, systolic and diastolic function. These indices were obtained every 6 months and measured centrally in a blinded fashion. Common Terminology Criteria for Adverse Events (CTCAE v4.0) was used to assess toxicity. Cardiac event was defined as relative EF/FS decline < 20% from baseline or both < 55% [EF] and < 28% [FS]. Treatment effects were analyzed using the Generalized Estimating Equation for normally distributed data, assuming linear time-trend. Group differences were evaluated by treatment*time interaction. Results: Of 196 participants enrolled, 182 were eligible and randomized. Characteristics (median, range): age at enrollment, 24.7y (10.4-53.9); time from cancer diagnosis, 13.4y (2.9-43.6); anthracycline dose, 374 mg/m 2 (250-1760); 19% had received chest-directed radiotherapy. There were no group differences in baseline participant demographics, cardiotoxic treatment exposure, or cardiac endpoints. LVWT/Dz was similar (+0.053, p = 0.15) between arms at 2y. However, treatment effects varied by time since diagnosis (p = 0.008) suggesting greater efficacy among longer-term survivors. Compared to placebo, carvedilol arm participants had significantly better LV end-diastolic diameter (-0.16 cm, p = 0.005) and LVESWS (-7.53, p = 0.027) at 2y. There were no CTCAE grade ≥3 events; all except 2 participants tolerated full dose escalation. Six out of eight cardiac events occurred on the placebo arm; longer-term follow-up of study participants is ongoing. Conclusions: In this randomized trial of long-term survivors of childhood cancer treated with high-dose anthracyclines, low-dose carvedilol was safe and effective in reducing established biomarkers of HF risk and could serve as a risk reduction strategy. Clinical trial information: NCT02717507 .