Addition of venetoclax to FLAI induction chemotherapy (VFLAI) improves survival in patients with intermediate and high-risk AML vs FLAI and 3+7: A GIMEMA Italian cooperative analysis

Background: Venetoclax combined with intensive chemotherapy proved to be safe with promising activity. The GIMEMA AML1718 trial (NCT03455504) was based on the administration of venetoclax-FLAI (fludarabine, idarubicin, cytarabine) to intermediate/high-risk ELN2017 AML. Methods: To investigate the potential advantage of the addition of venetoclax, patient-level data from GIMEMA AML1718 (V-FLAI group, n=57) were compared to “3+7”-like induction (AML1310, Venditti et al. 2019, 3+7 group, n=445), and to FLAI real-life experience (Guolo et al. 2016, FLAI group, n=155) using a propensity score matching. In the model, we included the following variables: age at diagnosis, gender, ELN2017 risk classification and allogenic transplant rate. For determining adequate balance, the diagnostic method employed was the standardized mean difference, visualized through a Love plot; a standardized bias score less than 0.2 was used as a criterion. The final weighted sample sizes are 183 (3+7), 54 (V-FLAI) and 155 (FLAI). Weights were calculated with a multinomial logistic regression model. The Odds Ratio (OR) were estimated using logit models. Survival curves were compared by standard and pairwise Log-rank test. Results: FLAI, V-FLAI and 3+7 cohorts differed in terms of age (median: 52 vs 54 vs 49 years, respectively, p<0.001), in terms of risk category (p< .001) - since the low risk was not represented both in V-FLAI and FLAI trials -, in terms of transplant rate (45% vs 49% vs 27%, p<0.001) and female sex (43% vs 35% vs 48%, p=0.15). Being more recent, AML1718 had a shorter median follow-up (10.5 vs 75.8 vs 104.8 months). The matching weights produced the best balance, with SMDs <0.18 for all variables. After weighting, the CR rate observed in the V-FLAI group was higher than FLAI and 3+7, as well as MRD-negativity rate. In terms of Odds Ratio (OR) the probability of achieving CR in V-FLAI and FLAI patients was significantly higher compared to 3+7 patients (p=0.01 and p=0.03, respectively). Conversely, V-FLAI and FLAI did not differ in their probability to achieve CR (p=0.34). Concerning MRD, the OR associated to V-FLAI group was significantly higher (namely, a higher probability of MRD-negativity) of that associated to both FLAI and 3+7 treatments (p=0.005 vs FLAI; p=0.001 vs 3+7). Looking at survival outcomes at 12 months, upon weighting, DFS estimate of the V-FLAI group was higher than “3+7” and FLAI (p=0.048). Conclusions: VFLAI resulted in a deeper disease response as compared to other regimens. These results highlight the potential benefit of venetoclax added to intensive induction chemotherapy and suggest the role of fludarabine-based regimens as a backbone for high-risk disease.