Ten-Day Decitabine with Venetoclax (DEC10-VEN) in Acute Myeloid Leukemia: Updated Results of a Phase II Trial

Authors:

Abstract
Background: Elderly patients (pt) with acute myeloid leukemia (AML) or pts with relapsed/refractory (R/R) AML have dismal outcomes. Venetoclax (VEN) synergizes with hypomethylating agents (HMA) and is now an approved combination for newly diagnosed (ND) AML in older/intensive chemotherapy (IC)-ineligible patients. We hypothesized that VEN with 10-day decitabine (DEC) may offer superior efficacy in AML. Methods: Between January and November 2018 we enrolled 101 pts with ND AML (n=40), untreated secondary AML (sAML, n=9), treated sAML (n=19) and R/R AML (n=33). Eligibility included ECOG PS ≤3, WBC ≤10 x109/L, and adequate organ function. DEC was given 20 mg/m2 IV daily on day 1-10 until CR/CRi, followed by 5-day cycles. VEN was given on day 1-28 in cycle 1 but was interrupted on C1D21 until count recovery if the day 21 bone marrow (BM) had ≤5% blasts. VEN could be reduced further to 14, 10 or 7 days in subsequent cycles for pts with ongoing cytopenias/myelosuppression. VEN dose was 400 mg PO daily (reductions allowed for concomitant CYP3A4 inhibitors) All pts received tumor lysis syndrome (TLS) prophylaxis. Cytoreduction prior to start, and concomitant BCR-ABL1 (n=1) and FLT3 inhibitors (n=15) were allowed as indicated. Primary objective was overall response rate and secondary objectives included safety and overall survival. The data cut-off date was 03.08.19. Results: This high-risk cohort included 52% of pts ≥70 yrs (interquartile range [IQR], 61-74), 54% pts were men, 28% pts had ECOG PS ≥2, and 67% pts had adverse-risk AML (Table 1). 30 and 60-day mortality were both 2.5% for ND pts, and 5% and 9%, respectively, for all pts. Notable treatment-emergent adverse events were infections with grade 3/4 neutropenia (61%), febrile neutropenia (35%), and TLS (4%, Table 2). Four pts developed grade 3 or 4 reversible TLS; their median WBC count was 7.1x109/L (1.6, 2.1, 12, 28) and median peripheral blasts was 64% (IQR 39-83) and led to a subsequent requirement for WBC ≤10 x109/L prior to start. The CR/CRi rate in ND AML was 95%, in untreated sAML was 67%, in treated sAML was 37%, and in R/R AML was 27% (Table 3). Out of 67 pts with C1D21±3 BM, ≤5% blasts was achieved in 57%, and 74% of ND pts. In previously untreated AML (defined as ND AML + untreated sAML), CR/CRi rate in ELN favorable, intermediate and adverse risk pts were 100% (n=11), 100% (n=6), and 84% (n=32). CR/CRi rate in t-AML was 100% (n=7). In previously treated AML (treated sAML + R/R AML), CR/CRi rate in HMA refractory pts was 39% (n=18), in pts with prior IC was 21% (n=33), in pts with prior HMA and IC was 17% (n=12), in pts with prior stem cell transplantation (SCT) was 22% (n=18). At median follow-up of 8.1 months (mo), the 6-mo OS in ND AML was 90%, in untreated sAML was 56%, in treated sAML was 62%, and R/R AML was 53% (Fig. 1). Among all pts, achievement of CR conferred better duration of response (DOR) and OS compared to CRi/MLFS (median DOR not reached [NR] vs 4.0 mo, censored at SCT, hazard ratio [HR] 0.21, 95% confidence interval [CI] 0.09-0.49, p=0.002, and median OS NR vs 7.8 mo, censored at SCT, HR 0.31, 95% CI 0.13-0.74). 70% of all responding pts were minimal residual disease negative (MRD-) by flow cytometry and MRD- status was associated with longer OS (NR vs 7.8 mo, censored at SCT, HR 0.35, 95% CI 0.14-0.88, p=0.01, Fig. 2a). Median DOR in pts achieving CR/CRi in ND AML was 8.5 mo, in the untreated sAML was 6.3 mo, in treated sAML was 4.8 mo, and in R/R AML was 6.6 mo. Among ND AML pts achieving CR/CRi, median time to count recovery after 1st, 2nd and 3rd cycles for ANC >0.5x109/L was 42 days (95% CI 37-46), 40 days (95% CI 35-44), and 38 days (95% CI 32-NR), respectively; and for platelet ≥50 x109/L was 28 days (95% CI 25-32), 25 days (95% CI 0-34), and 26 days (95% CI 19-33), respectively. 78 pts (77%) discontinued treatment, and 59 pts (58%) are alive. Common reasons for discontinuation were non-response in 24 pts (24%), SCT in 20 pts (20%), and relapse in 18 pts (18%). Pts receiving SCT (n= 20) had excellent outcomes with 100-day post-SCT mortality of 0% (Fig 2b and c). Baseline BCL2 expression in CD34+ blasts showed a trend between higher BCL2 expression and MRD- rate (Fig. 2d). Conclusions: DEC10-VEN is an effective therapy for ND elderly pts, as well as in R/R AML as an effective bridge to SCT. Trial continues to accrue (NCT03404193).