Comparative long-term cardiovascular and hematologic outcomes of gemtuzumab ozogamicin containing regimens versus idarubicin containing regimens in acute promyelocytic leukemia: A propensity-matched analysis of real-world multicenter data

Abstract Background All-trans retinoic acid (ATRA)-based regimens have revolutionized the treatment of acute promyelocytic leukemia (APL). While ATRA combined with idarubicin remains a frontline standard, ATRA plus gemtuzumab ozogamicin (GO), a CD33-directed antibody-drug conjugate, has emerged as a chemotherapy-sparing alternative. Comparative long-term data on cardiovascular and hematologic toxicities of these regimens remain limited. Methods We conducted a retrospective cohort study using the TriNetX global research platform to identify patients diagnosed with APL who received either ATRA + GO (n=180) or ATRA + idarubicin (n=4,096). Propensity score matching (1:1) was performed based on age, race, BMI, hypertension, hypertensive heart disease with heart failure, diabetes mellitus, malnutrition, asthma, and emphysema, resulting in matched cohorts of 166 patients each. The primary outcome was the 5-year incidence of all-cause mortality, and secondary outcomes included the incidence of acute arterial thrombosis, DIC, acute MI, cardiac arrest, heart failure, atrial fibrillation/flutter, AKI, readmission rate, acute ischemic stroke, and the incidence of secondary cancers. Risk ratios, 95% confidence intervals, and p values were calculated. Results After propensity score matching each cohort comprised of 166 patients. All-cause mortality was significantly lower in the ATRA + GO group compared to the idarubicin group (15.7% vs. 33.7%; RR 0.46, 95% CI 0.30-0.70, p<0.001). The idarubicin group higher incidence of cardiac arrest (RR 1.8, 95%, p=0.001). The ATRA+GO cohort experienced a higher incidence of DIC ( 19.8% VS. 6.0%, RR:3.3, 95%CI: 1.68-6.47, p<0.001). There was no statistically significant difference in the outcomes of acute MI, heart failure, atrial fibrillation/flutter, acute ischemic stroke, AKI, readmission rate, hypothyroidism, secondary tumors incidence, isolated arterial thrombi, and deep venous thrombosis. Conclusions In this real-world, propensity-matched analysis of APL patients, ATRA + GO was associated with significantly lower 5-year mortality and incidence of cardiac arrest but increased risks of DIC compared to ATRA + idarubicin. These findings highlight the importance of individualized risk-benefit assessment when selecting induction regimens for APL. Table 1. Summary of Efficacy Outcomes from Meta-analysis Outcome Risk Ratio (RR) [95% CI] P-value Heterogeneity (I²) 2-year Progression-Free Survival (PFS) 1.50 [0.50–4.50] 0.254 85.3% PFS (Sensitivity analysis) 2.05 [1.18–3.55] 0.038 0% 3-year Overall Survival (OS) 1.00 [0.96–1.05] 0.593 0% Overall Response Rate (ORR) 1.09 [0.81–1.49] 0.697 65.2% Complete Response (CR) 0.78 [0.25–2.47] 0.549 82.2% Partial Response (PR) 1.20 [0.74–1.94] 0.325 24.9% Table 2. Summary of Adverse Events from Meta-analysis Adverse Event Risk Ratio (RR) [95% CI] P-value Heterogeneity (I²) Neutropenia (Any Grade) 1.56 [0.22–11.09] 0.434 75.8% Neutropenia (Grade ≥3) 1.39 [0.12–16.03] 0.620 86.9% Thrombocytopenia (Any Grade) 1.09 [0.19–6.10] 0.855 13.6% Thrombocytopenia (Grade ≥3) 1.71 [0.34–8.59] 0.289 0% Anemia (Any Grade) 1.20 [0.17–8.41] 0.727 60.1% Anemia (Grade ≥3) 2.96 [0.12–71.57] 0.504 N/A Skin Rash 2.79 [1.08–7.17] 0.033 65.2% Diarrhea 1.98 [1.65–2.38] 0.004 0% Fatigue 1.09 [0.40–2.99] 0.747 79.6% Nausea/Vomiting 0.91 [0.14–5.80] 0.844 83.3% Elevated Aminotransferases 1.00 [0.03–31.60] 0.997 11.3%