Abstract Book for the 27th Congress of the European Hematology Association

Background: Pts with AML in RCTs often do not reflect the population seen in clinical practice due to strict eligibility criteria. Aims: To evaluate patient outcomes of IC vs venetoclax (VEN)-containing regimens based on eligibility criteria from a recent RCT (the VIALE-A trial) in a broad cohort of real-world pts with AML from the Connect® Myeloid Disease Registry (NCT01688011). Methods: Pts were stratified into 3 groups based on the non-IC phase 3 VIALE-A trial eligibility criteria: 1) 'eligible' pts who would have met all VIALE-A inclusion criteria; 2) 'unfit' pts who would have been ineligible for VIALE-A due to ≥ 1 of the following: abnormal liver/kidney function, high ECOG, recent prior malignancy, comorbidities score ≥ 2 by ACE-27, hepatic grade ≥ 1, AIDS grade ≥ 1; 3) 'fit' pts who would have been ineligible for a VEN-based regimen in VIALE-A because they would have qualified for IC (defined as: ≤ 74 y of age, low ECOG, no apparent cardiovascular/renal comorbidities, and did not meet any criteria in #2). Baseline characteristics were summarized by eligibility group. Overall survival by group was estimated using the Kaplan-Meier method. Induction regimens were categorized as IC (any regimen containing 7+3, MEC, CLAG, FLAG) or VEN-based. Hazard ratios (HR) for induction regimens among each group were estimated using Cox models adjusted for age, ELN risk, ECOG, frailty score, and comorbidity index. Results: Of 734 enrolled pts with AML (Dec 2021), most were male (61%) and white (84%); median age 71 y (range, 55-97). Only 26% of pts (n = 192) were eligible for a non-IC RCT, 45% (n = 327) were ineligible due to unfitness, and 29% (n = 215) were ineligible due to overall fitness. The main reason for non-IC RCT ineligibility was high overall comorbidity grade (n = 265 [36%]). At baseline, fit pts intended to undergo transplant more often compared with unfit pts. Median duration of overall survival for eligible, unfit, and fit pts was 14, 10, and 22 months, respectively (HR [95% CI], eligible vs unfit, 0.02 [-0.19 to 0.22], P = 0.8735; eligible vs fit, 0.57 [0.33-0.81], P < 0.0001; unfit vs fit, 0.55 [0.33-0.77], P < 0.0001). Among unfit pts, those receiving IC had significantly longer overall survival compared with pts receiving a VEN-based regimen (median overall survival, 14 vs 6 months, respectively; HR, 0.51 [95% CI, 0.27-0.98]; P = 0.042; Figure). Unfit patients who received IC went on to transplant more frequently than those who received VEN-based therapies (16% [n = 17] vs 1% [n = 1], respectively). Eligible pts who received IC (n = 31) tended to have shorter median overall survival (13 months) vs pts who received VEN-based therapies (n = 27; 23 months; HR, 1.45, 95% CI, 0.66-3.17; not sig.). Among fit pts, median overall survival was 19 months for those receiving IC (n = 69) but could not be estimated for pts receiving VEN-based therapies due to small sample size (n = 10). Image: Summary/Conclusion: The majority of pts with ND AML in the Connect® Myeloid Disease Registry would have been ineligible for a non-IC RCT due to being too fit or unfit. Among pts ineligible for an RCT due to unfitness, there was an association with increased overall survival in pts receiving IC vs those receiving VEN-based therapies, and pts in the IC group were more likely to receive a transplant. This analysis suggests that RCTs may be excluding pts who appear unfit but can potentially tolerate IC and experience improved survival outcomes.