Treatment Free Remission (Tfr) After Ceasing Venetoclax-Based Therapy In Responding Patients With Acute Myeloid Leukemia

Context: Ph3 studies confirmed the benefit of adding venetoclax (VEN) to hypomethylating agents (HMA) or low-dose cytarabine (LDAC) in older/unfit patients with newly diagnosed AML with a median response duration of 12-18 months(mo). It is currently unknown whether patients should continue therapy until progression and whether elective therapy cessation could detrimentally impact outcomes. Objective: To compare the natural history of patients receiving ≥12mo of VEN-based therapy in first remission. Design, Setting and Patients: Retrospective analysis of patients with AML receiving VEN plus HMA or LDAC for ≥12mo and in sustained first remission from three institutions. Two treatment approaches were compared: elective therapy cessation in remission followed by monitoring (STOP) or continued therapy until relapse (CONT). Main Outcome Measures: Primary: treatment free remission (TFR), progression free survival (PFS). Secondary: overall survival (OS), molecular correlates of PFS. Results: 28 patients were identified: 14 in STOP, 14 in CONT. Median follow-up was 60.9mo. Treatments included VEN-HMA (68%) or VEN-LDAC (32%). Median time on therapy (TOT) was shorter in STOP (19.0mo vs 32.7mo) and fewer cycles were received in STOP (median 17 vs 30). In STOP, treatment cessations were due to patient request (43%) or medical (57%), with a median TFR of 45.8m (95%CI 9.5-not reached). 7/14(50%) have relapsed, with 3 relapsing >31mo of TFR. In CONT, 9/14(64%) have relapsed, with 6 relapsing >24mo of therapy. Cytogenetic evolution was more frequently observed at relapse in CONT (67% vs 43%). A landmark analysis was performed (landmarked at median TOT in STOP to exclude lead-time bias from early relapses in CONT), which revealed no significant difference in PFS (p=0.14) or OS (p=0.24). 10/15 patients with NPM1 and/or IDH2 mutations remain in remission (6/8 STOP, 4/7 CONT), of which 9/10 achieved undetectable MRD (by flow/qPCR). 1 had detectable IDH2 mutation at pre-leukemic levels. Multivariate analysis showed that mutant NPM1 is an independent predictor of OS (HR for death 0.002, p=0.037) and PFS (HR for relapse 0.012, p=0.024). Conclusion: Durable TFR is highest among patients with NPM1/IDH2 mutations and MRD negative after ceasing >12 months of VEN-based treatment. A prospective, randomised, discontinuation study in pts in CRMRDneg after 12 months of VEN-based therapy should be considered. Ph3 studies confirmed the benefit of adding venetoclax (VEN) to hypomethylating agents (HMA) or low-dose cytarabine (LDAC) in older/unfit patients with newly diagnosed AML with a median response duration of 12-18 months(mo). It is currently unknown whether patients should continue therapy until progression and whether elective therapy cessation could detrimentally impact outcomes. To compare the natural history of patients receiving ≥12mo of VEN-based therapy in first remission. Retrospective analysis of patients with AML receiving VEN plus HMA or LDAC for ≥12mo and in sustained first remission from three institutions. Two treatment approaches were compared: elective therapy cessation in remission followed by monitoring (STOP) or continued therapy until relapse (CONT). Primary: treatment free remission (TFR), progression free survival (PFS). Secondary: overall survival (OS), molecular correlates of PFS. 28 patients were identified: 14 in STOP, 14 in CONT. Median follow-up was 60.9mo. Treatments included VEN-HMA (68%) or VEN-LDAC (32%). Median time on therapy (TOT) was shorter in STOP (19.0mo vs 32.7mo) and fewer cycles were received in STOP (median 17 vs 30). In STOP, treatment cessations were due to patient request (43%) or medical (57%), with a median TFR of 45.8m (95%CI 9.5-not reached). 7/14(50%) have relapsed, with 3 relapsing >31mo of TFR. In CONT, 9/14(64%) have relapsed, with 6 relapsing >24mo of therapy. Cytogenetic evolution was more frequently observed at relapse in CONT (67% vs 43%). A landmark analysis was performed (landmarked at median TOT in STOP to exclude lead-time bias from early relapses in CONT), which revealed no significant difference in PFS (p=0.14) or OS (p=0.24). 10/15 patients with NPM1 and/or IDH2 mutations remain in remission (6/8 STOP, 4/7 CONT), of which 9/10 achieved undetectable MRD (by flow/qPCR). 1 had detectable IDH2 mutation at pre-leukemic levels. Multivariate analysis showed that mutant NPM1 is an independent predictor of OS (HR for death 0.002, p=0.037) and PFS (HR for relapse 0.012, p=0.024). Durable TFR is highest among patients with NPM1/IDH2 mutations and MRD negative after ceasing >12 months of VEN-based treatment. A prospective, randomised, discontinuation study in pts in CRMRDneg after 12 months of VEN-based therapy should be considered.