Volume 127, Issue 22 p. 4213-4220
Original Article
Free Access

Ten-day decitabine with venetoclax versus intensive chemotherapy in relapsed or refractory acute myeloid leukemia: A propensity score-matched analysis

Abhishek Maiti MBBS

Abhishek Maiti MBBS

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Courtney D. DiNardo MD, MSCE

Courtney D. DiNardo MD, MSCE

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Wei Qiao PhD

Wei Qiao PhD

Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Tapan M. Kadia MD

Tapan M. Kadia MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Elias J. Jabbour MD

Elias J. Jabbour MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Caitlin R. Rausch PharmD

Caitlin R. Rausch PharmD

Division of Pharmacy, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Naval G. Daver MD

Naval G. Daver MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Nicholas J. Short MD

Nicholas J. Short MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Gautam Borthakur MD

Gautam Borthakur MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Naveen Pemmaraju MD

Naveen Pemmaraju MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Musa Yilmaz MD

Musa Yilmaz MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Yesid Alvarado MD

Yesid Alvarado MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Kathryn S. Montalbano RN

Kathryn S. Montalbano RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Allison Wade RN

Allison Wade RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Rita E. Maduike RN

Rita E. Maduike RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Julio A. Guerrero RN

Julio A. Guerrero RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Kenneth Vaughan RN

Kenneth Vaughan RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Carol A. Bivins RN

Carol A. Bivins RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Sherry Pierce RN

Sherry Pierce RN

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Jing Ning PhD

Jing Ning PhD

Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Farhad Ravandi MD

Farhad Ravandi MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Hagop M. Kantarjian MD

Hagop M. Kantarjian MD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Search for more papers by this author
Marina Y. Konopleva MD, PhD

Corresponding Author

Marina Y. Konopleva MD, PhD

Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas

Corresponding Author: Marina Y. Konopleva, MD, PhD, Department of Leukemia, The University of Texas MD Anderson Cancer Center, 1400 Holcombe Blvd, Unit 428, Houston, TX 77030 ([email protected]).

Search for more papers by this author
First published: 03 August 2021
Citations: 15

We thank the patients, their caregivers, and members of the study teams involved in these trials.

Abstract

Background

Relapsed/refractory (R/R) acute myeloid leukemia (AML) has poor outcomes. Although lower-intensity venetoclax-containing regimens are standard for older/unfit patients with newly diagnosed AML, it is unknown how such regimens compare with intensive chemotherapy (IC) for R/R AML.

Methods

Outcomes of R/R AML treated with 10-day decitabine and venetoclax (DEC10-VEN) were compared with IC-based regimens including idarubicin with cytarabine, with or without cladribine, clofarabine, or fludarabine, with or without additional agents. Propensity scores derived from patient baseline characteristics were used to match DEC10-VEN and IC patients to minimize bias.

Results

Sixty-five patients in the DEC10-VEN cohort were matched to 130 IC recipients. The median ages for the DEC10-VEN and IC groups were 64 and 58 years, respectively, and baseline characteristics were balanced between the 2 cohorts. DEC10-VEN conferred significantly higher responses compared with IC including higher overall response rate (60% vs 36%; odds ratio [OR], 3.28; P < .001), complete remission with incomplete hematologic recovery (CRi, 19% vs 6%; OR, 3.56; P = .012), minimal residual disease negativity by flow cytometry (28% vs 13%; OR, 2.48; P = .017), and lower rates of refractory disease. DEC10-VEN led to significantly longer median event-free survival compared with IC (5.7 vs 1.5 months; hazard ratio [HR], 0.46; 95% CI, 0.30-0.70; P < .001), as well as median overall survival (OS; 6.8 vs 4.7 months; HR, 0.56; 95% CI, 0.37-0.86; P = .008). DEC10-VEN was independently associated with improved OS compared with IC in multivariate analysis. Exploratory analysis for OS in 27 subgroups showed that DEC10-VEN was comparable with IC as salvage therapy for R/R AML.

Conclusion

DEC10-VEN represents an appropriate salvage therapy and may offer better responses and survival compared with IC in adults with R/R AML.

Introduction

Acute myeloid leukemia (AML) with intermediate or adverse risk features have poor outcomes and more than 30% to 60% patients experience either refractory disease or relapse.1 Apart from AML with actionable mutations (eg, FLT3, IDH1/2), there are no standard treatments for relapsed or refractory (R/R) AML and outcomes are poor with long-term survival of only 10% to 20%.2, 3 With disease progression or each subsequent therapy, including stem cell transplantation (SCT), it often becomes challenging to continue intensive chemotherapy (IC) in such patients. Consequently, R/R AML remains a therapeutic challenge with an urgent need for novel lower-intensity therapies.

In newly diagnosed patients, venetoclax-containing lower-intensity regimens are the current standard of care for older or unfit patients with AML.4, 5 Venetoclax with hypomethylating agents (HMAs) is potentially better than IC as frontline therapy for older patients including those who are eligible for IC.6 Such venetoclax-based lower-intensity regimens are potentially efficacious in younger patients as well, and prospective trials in this population are currently ongoing. We reported the results of the first prospective trial of venetoclax with decitabine in patients with R/R AML who are older than 18 years.7 However, it is unknown how such venetoclax and HMA regimens compare with IC for R/R AML.

Herein, we compared the outcomes of adult patients with R/R AML treated with 10-day decitabine and venetoclax (DEC10-VEN) on a phase 2 trial (Clinical Trial Registration: NCT03404193) with outcomes of patients treated with IC.

Materials and Methods

Patients and Treatment

The R/R cohort in the prospective phase 2 trial of DEC10-VEN enrolled adults older than 18 years with Eastern Co-operative Oncology Group performance status (ECOG PS) of 3 or lower. Patients with European LeukemiaNet (ELN) favorable risk cytogenetics or prior exposure to venetoclax were excluded. Patients received decitabine 20 mg/m2 daily for 10 days every 4 to 6 weeks with venetoclax 400 mg daily or equivalent (with concomitant azole antifungals) for induction. After achievement of a response, patients received decitabine for 5 days with daily venetoclax for consolidation. Patients received cytoreduction to white blood cells <10 × 109/L before starting. All patients received monitoring and prophylaxis for tumor lysis syndrome and antimicrobial prophylaxis during neutropenia.

Patients in the comparator IC cohort were older than 18 years with ECOG PS of 3 or lower. Patients received salvage therapy with idarubicin with cytarabine (IA), with or without cladribine (CLIA), clofarabine (CIA), or fludarabine (FIA or FLAG-IDA)–based regimens without venetoclax on prospective phase 1b/2 clinical trials or as standard of care (Supporting Table 1A). The IA-based regimen was comprised of idarubicin 10 to 12 mg/m2 intravenously (iv) on days 1 to 3 and cytarabine 1 to 2 g/m2 iv on days 1 to 4. CLIA-based regimens comprised of cladribine 5 mg/m2 iv on days 1 to 5, cytarabine 1-2 g/m2 iv on days 1 to 5, and idarubicin 10 mg/m2 iv on days 1 to 3. CIA-based regimens included clofarabine 15 mg/m2 iv daily on days 1 to 5, idarubicin 10 mg/m2 iv daily on days 1 to 3, and cytarabine 1 g/m2 iv daily on days 1 to 5. FLAG-Ida or FIA-based regimens included fludarabine 30 mg/m2 iv on days 1 to 5, idarubicin 10 mg/m2 iv daily on days 1 to 3, and cytarabine 1 to 2 g/m2 iv daily on days 1 to 5, with or without filgrastim 5 µg/kg subcutaneously on days 1 to 7. Details of additional agents in the overall population and the matched population are given in Supporting Table 1B,C.

Response Assessment and Outcomes

Responses were graded per the ELN 2017 guidelines.8 Overall response rate (ORR) includes complete remission (CR), CR with incomplete hematologic recovery (CRi), and morphologic leukemia-free state (MLFS). Response was established by the time of completion of up to 2 cycles of IC and up to 4 cycles of DEC10-VEN. Minimal residual disease (MRD) was assessed in bone marrow specimens using multiparametric flow cytometry validated to a sensitivity level of 0.1%.9 Adverse events were determined per the National Cancer Institute Common Terminology Criteria for Adverse Events v4.0. Event-free survival (EFS) was defined for all patients and was measured from start of therapy to the date of primary refractory disease, relapse from CR/CRi, or death from any cause; patients not known to have any of these events were censored at last follow-up. Overall survival (OS) was measured from start of therapy until death or censorship at last follow-up. Relapse-free survival was measured from date of achievement of CR/CRi until relapse, death, or censorship at last follow-up.

Statistical Analysis

Propensity score methods in retrospective studies attempt to approximate populations in randomized controlled trials. This is achieved by selecting for comparable patient characteristics by balancing specified covariates between comparator groups using propensity scores to reduce the potential for confounding or bias in estimating treatment effect in observational studies.10 A propensity score is the probability that a given patient would receive a given treatment based on their clinical scenario, treating clinician, and care setting.11 Propensity scores were calculated by logistic regression using baseline characteristics well recognized to be prognostic in the salvage setting, including age, ECOG PS, number of prior lines of therapies, primary refractory versus relapsed disease, CR1 duration, prior allogeneic stem-cell transplantation, ELN cytogenetic risk, and FLT3 and TP53 mutation status.12, 13 Propensity score matching (PSM) with the nearest-neighbor method was used to select the IC comparison group using 1:2 matching with a caliper control of 0.3.14, 15

Pairwise comparison of patient characteristics and outcomes was performed using the t test or the Wilcoxon signed-rank test for continuous variables, and the McNemar test for categorical variables or the Fisher exact test for unpaired comparisons. The Kaplan-Meier method with stratified log-rank test was used to compare time-to-event variables. The Cox proportional hazards model with stratification to account for matched pairs was used to estimate the hazard ratio (HR) between comparator groups. Univariate and multivariable Cox models were used in the unmatched population to evaluate the associations between patient characteristics and OS. Factors significant in the univariate models at significance level 0.1 were included in the multivariate model. Backward model selection was used to remove variables from the model until all remaining variables were statistically significant with P < 0.05 while treatment effect variable was forced in the final multivariable model. SPSS 24 (IBM Corp), R Essentials for SPSS version 24, the PSMATCHING3.03 package, and R version 3.2 (R Foundation) were used for analysis.

Results

Patients

Sixty-five out of 65 patients treated with DEC10-VEN were best matched to 130 out of 291 patients treated with IC (Supporting Fig. 1). The standardized mean differences of variables selected for PSM before and after matching was low (range, -0.04 to 0.23) suggesting significant reduction of bias between the two intervention groups (Supporting Table 2). Patients in DEC10-VEN cohort were treated between January 2018 and September 2020 whereas patients in IC cohort received treatment between June 2005 and July 2019. The median age of the DEC10-VEN cohort was 64 years (range, 18-85); for the IC cohort, it was 58 years (range, 19-80). Baseline characteristics were well balanced between the two groups (Table 1). The median number of prior lines of therapy was similar in the DEC10-VEN and IC cohorts at 2 (range, 1-8) versus 2 (range, 1-10), respectively (P = .721). Details of the regimens administered in the IC cohort are given in Supporting Table 1C. In the DEC10-VEN cohort, patients received a median of 2 cycles of therapy (range, 1-16); in the IC cohort, patients had received a median of 1 cycle of therapy (range, 1-5). FLT3 inhibitors were used in 16 of 16 patients treated with DEC10-VEN compared with 25 of 34 patients treated with IC (P = .043). No patients in either group received isocitrate dehydrogenase 1/2 (IDH1/2) inhibitors.

TABLE 1. Baseline Characteristics of Matched Patients With Relapsed or Refractory Acute Myeloid Leukemia Who Received 10-Day Decitabine With Venetoclax (DEC10-VEN) versus Intensive Chemotherapy (IC)
Patient characteristics DEC10-VEN (N = 65) IC (N = 130) P
Age, y 64 [18-85] 58 [19-80] .113
<60 y 26 (40) 74 (57) .038
Male sex 39 (60) 72 (55) .539
AML type
Primary refractory 28 (43) 54 (42)
AML
Relapsed AML 37 (57) 76 (58) .837
Secondary AML from AHD 11 (17) 21 (16) .891
ECOG performance status
0-1 46 (71) 95 (73) .734
≥2 19 (29) 35 (27)
Bone marrow blasts, % 34 [1-96] 35 [0-98] .536
ELN 2017 cytogenetic group
Favorable 0 (0) 0 (0)
Intermediate 41 (63) 80 (62) .958
Adverse 24 (37) 50 (38)
Mutations
NPM1 14 (22) 15 (12) .605
FLT3-ITD/TKD 16 (25) 34 (26)
IDH1/2 11 (17) 28 (22)
TP53 18 (28) 32 (25)
RUNX1 12 (18) 22 (17)a
ASXL1 10 (15) 14 (11)a
ELN 2017 risk group
Favorable 10 (15) 11 (8) .302
Intermediate 12 (18) 30 (23)
Adverse 43 (66) 89 (68)
No. of prior therapies 2 [1-8] 2 [1-10] .721
Prior therapies
HMA only 16 (25) 24 (18) .214
IC only 31 (48) 53 (41)
IC and HMA 16 (25) 53 (41)
SCT 18 (28) 36 (28)
  • Abbreviations: AHD, antecedent hematological disorder; AML, acute myeloid leukemia; ECOG, Eastern Co-operative Oncology Group; ELN, European LeukemiaNet; HMA, hypomethylating agent; IC, intensive chemotherapy; SCT, allogeneic stem cell transplantation.
  • Results reported as No. (%), or median [range].
  • a RUNX1 and ASXL1 status was unknown for 9 patients.

Efficacy: Response Rates

ORRs, CRi, MLFS, and MRD negativity by flow cytometry were significantly higher with DEC10-VEN compared with IC. ORR was achieved in 60% patients with DEC10-VEN versus 36% patients treated with IC (odds ratio [OR], 3.28; 95% CI, 1.73-6.23; stratified test P < .001; Table 2). CRi rates were 19% versus 6%, respectively (OR, 3.56; 95% CI, 1.32-9.61; stratified test P = .012); rates of MLFS were 15% versus 5%, respectively (OR, 3.33; 95% CI, 1.21-9.17; stratified test P = .020), and the MRD-negativity rate was 28% versus 13%, respectively (OR, 2.48; 95% CI, 1.17-5.24; stratified test P = .017). The rate of refractory disease was significantly lower with DEC10-VEN (35% vs 55%; OR, 0.46; 95% CI, 0.25-0.84; stratified test P = .011). The median number of cycles to best response in the DEC10-VEN cohort was 1 (range, 1-4) and in the IC cohort was 1 (range, 1-3).

TABLE 2. Outcomes of Propensity Score Matched Patients With Relapsed or Refractory Acute Myeloid Leukemia Who Received 10-Day Decitabine With Venetoclax (DEC10-VEN) versus Intensive Chemotherapy (IC)
Outcomes DEC10-VEN (N = 65) IC (N = 130) OR/HR (95% CI) Stratified P
Overall response rate (ORR) 37 (60) 36 (28) 3.28 (1.73-6.23) <.001
CR/CRi 27 (42) 30 (23) 2.52 (1.26-5.03) .009
CR 15 (23) 22 (17) 1.44 (0.70-2.94) .320
CRi 12 (19) 8 (6) 3.56 (1.32-9.61) .012
MLFS 10 (15) 6 (5) 3.33 (1.21-9.17) .020
MRD negative by FCM 18 (28) 17 (13) 2.48 (1.17-5.24) .017
Refractory 22 (35) 71 (55) 0.46 (0.25-0.84) .011
Inevaluablea 1 (2) 1 (1) 2.00 (0.13-31.98) .624
Relapse 19 (51) 17 (47) 0.75 (0.24-2.38) .630
30-Day mortality 5 (8) 16 (12) 0.58 (0.2-1.70) .322
60-Day mortality 8 (12) 35 (27) 0.40 (0.18-0.91) .029
  • Abbreviations: CR, complete remission; CRi, CR with incomplete hematologic recovery; HR, hazard ratio; MLFS, morphologic leukemia-free state; MRD, minimal residual disease evaluated using multiparametric flow cytometry validated to a sensitivity level of 0.1%; NA, not applicable; OR, odds ratio; ORR, CR + CRi + MLFS.
  • Results reported as No. (%), n/N (%).
  • a One patient was taken off DEC10-VEN after receiving only 2 days of therapy and another patient was taken off IC after 1 day off therapy based on adverse event.

Efficacy: Survival Outcomes

The median follow-up for the DEC10-VEN cohort was 17.5 months and for the IC cohort was 49.3 months. The median EFS with DEC10-VEN was 5.7 months versus 1.5 months with IC (HR, 0.46; 95% CI, 0.30-0.70; stratified test P < .001; Fig. 1A). The median OS with DEC10-VEN was 6.8 months compared with 4.7 months with IC (HR, 0.56; 95% CI, 0.37-0.86; stratified test P = .008; Fig. 1B). Relapse-free survival was similar with DEC10-VEN and IC (8.4 vs 8.4 months; HR, 1.103; 95% CI, 0.59-2.07; P = .743; Supporting Fig. 2). Fifteen patients (23%) in the DEC10-VEN cohort and 21 patients (16%) in the matched IC cohort received SCT after achievement of a response (P = .247). The median OS after SCT was numerically higher in DEC10-VEN versus IC cohorts (18.7 vs 12.9 months; HR, 1.11; 95% CI, 0.44-2.84; P = .813; Supporting Fig. 3).

Details are in the caption following the image
(A) Event-free survival (EFS), and (B) overall survival (OS) in patients with relapsed or refractory acute myeloid leukemia (AML) who received salvage therapy with 10-day decitabine with venetoclax (DEC10-VEN) versus intensive chemotherapy (IC). (C) Exploratory subgroup analysis for OS in unmatched populations. The dotted line represents a hazard ratio (HR) of 1.00. Only patients receiving FLT3 inhibitors were chosen for the subgroup analysis comparing outcomes in FLT3mut AML. All 16 patients treated with DEC10-VEN received FLT3 inhibitor including gilteritinib (n = 7), sorafenib (n = 6), and midostaurin (n = 3); all 48 selected patients treated with IC received FLT3 inhibitor including sorafenib (n = 33), crenolanib (n = 10), gilteritinib (n = 2), midostaurin (n = 2), and ponatinib (n = 1). AHD, antecedent hematological disorder; CR1, first remission duration; ECOG PS, Eastern Co-operative Oncology Group performance status; HMA, hypomethylating agent; MRD, minimal residual disease; sAML, secondary AML; SCT, stem-cell transplantation.

Exploratory subgroup analyses for OS in unmatched cohorts showed that DEC10-VEN was comparable to IC across most subgroups of age, diagnosis, cytogenetic risk, line of salvage therapy, type of previous therapy, and major mutational subgroups (Fig. 1C). Although 95% CIs were wide, the HRs marginally favored DEC10-VEN in most subgroups; the benefit was significant in patients with CR1 duration of more than 6 months and in patients with antecedent hematological disorder. In patients with CR1 duration of more than 6 months, the median OS with DEC10-VEN was 21.8 months (n = 13) compared with 6.5 months with IC (n = 76; HR, 0.52; 95% CI, 0.28-0.96; P = .023). In patients with antecedent hematological disorder, the median OS with DEC10-VEN was 6.7 months (n = 11) compared with 3.5 months with IC (n = 56; HR, 0.42; 95% CI, 0.23-0.76; P = .017). Patients with mutated fms-like tyrosine kinase 3 (FLT3mut) AML treated with DEC10-VEN and FLT3 inhibitor (n = 16) versus IC with FLT3 inhibitor (n = 48) had comparable OS (6.4 months vs 7.0 months; HR, 0.83; 95% CI, 0.44-1.58; P = .587).

Multivariate analysis for OS in unmatched population showed that DEC10-VEN was independently associated with significant improvement in OS (HR, 0.66; 95% CI, 0.48-0.92; P = .012; Supporting Table 3), and IDH1/2 mutation was associated with a lower risk of death (HR, 0.70; 95% CI, 0.51-0.97; P = .029). There was a higher risk of death in patients with an antecedent hematological disorder (HR, 1.84; 95% CI, 1.35-2.51; P < .001), ELN adverse-risk cytogenetics (HR, 1.57; 95% CI, 1.21-2.03; P < .001), TP53 mutation (HR, 1.71; 95% CI, 1.25-2.34; P < .001), and failure of 2 or more lines of therapy. Overall survival was comparable among patients who achieved CR or CR/CRi with DEC10-VEN or IC (Supporting Figs. 4 and 5).

Safety

The 60-day mortality was 12% (n = 8) with DEC10-VEN compared with 27% (n = 35) with IC (OR, 0.40; 95% CI, 0.18-0.91, stratified test P = .029). Prospectively collected adverse event data were available for all patients in the DEC10-VEN cohort compared with 28 patients (22%) in the IC cohort. Nonhematologic grade 3/4 treatment-emergent adverse events, at least possibly related to the study regimen, occurred in 21 (75%) of 28 patients receiving IC and 45 (69%) of 65 patients receiving DEC10-VEN (P = .628; Table 3). Grade 3/4 infectious adverse events occurred in 25 out of 65 (38%) patients in the DEC10-VEN cohort compared with 5 of 28 (18%) patients in the IC cohort (P = .057). Febrile neutropenia was noted in 20 patients (30%) in the DEC10-VEN cohort compared with 14 patients (50%) in the IC cohort (P = .101).

TABLE 3. Adverse Events in Propensity Score-Matched Patients With Relapsed or Refractory Acute Myeloid Leukemia Who Received 10-Day Decitabine With Venetoclax (DEC10-VEN) Versus Intensive Chemotherapy (IC)
Event DEC10-VEN (n = 65) IC (n = 28)
Any Grade Grade 3/4 Any Grade Grade 3/4
Infection with ANC <1.0 × 109/L 25 (38) 25 (38) 6 (21) 5 (17)
Febrile neutropenia 20 (30) 20 (30) 14 (50) 14 (50)
Infection with ANC >1.0 × 109/L 14 (21) 11 (16) 3 (10) 3 (10)
Mucositis 4 (6) 4 (14) 1 (3)
Hyperbilirubinemia 3 (4) 3 (4) 8 (28) 2 (7)
Nausea 3 (4) 1 (3)
Tumor lysis syndrome 3 (4) 1 (1) 1 (3) 1 (3)
Constipation 2 (3)
Renal failure 2 (3) 2 (3)
Pain 1 (1) 1 (1) 4 (14) 2 (7)
Hypokalemia 1 (1)
Vomiting 1 (1)
ALT/AST elevation 10 (35) 1 (3)
Hemorrhage 4 (14) 3 (10)
Rash 4 (14)
Creatinine elevation 3 (10)
Altered mental status 1 (3) 1 (3)
Cardiac arrhythmia 1 (3)
Cholecystitis 1 (3) 1 (3)
Dyspnea 1 (3) 1 (3)
Hypoxia 1 (3) 1 (3)
Pruritus 1 (3)
  • Abbreviations: ALT, alanine aminotransferase; ANC, absolute neutrophil count; AST, aspartate aminotransferase.

Discussion

Heterogeneous biology in R/R AML without actionable mutations makes it a challenging disease for drug development; hence, only modest progress has been made in the last several decades. Although IC has been the mainstay so far, declining functional status in R/R AML and treatment-related toxicities are considerable barriers to treatment and limit the testing of novel agents because of the risk of additive toxicity. Venetoclax-based lower-intensity regimens have been a paradigm-shifting development in the field because of modest end-organ toxicity apart from myelosuppression and resultant infections. This has enabled treatment delivery in frail patients who otherwise would not have tolerated intensive therapies and have accelerated the development of rational combination therapies in AML.

Our retrospective comparison shows that DEC10-VEN is an appropriate option for venetoclax-naïve patients with R/R AML. In a population with a median of 2 prior lines of therapy and a majority with adverse-risk AML, DEC10-VEN showed significantly higher ORR, rate of CRi, MRD negativity, lower rates of 60-day mortality, refractory disease, and significantly longer EFS and OS. The OS benefit noted was further confirmed on multivariate analysis. Although OS was only modestly improved compared with IC, it was encouraging to note that despite potential confounders, the HRs across most subgroups favored DEC10-VEN over IC. In particular, the trend noted in patients with antecedent hematological disorder—CR1 duration longer than 6 months and RUNX1 mutations—was encouraging. It is possible that the improvement in OS may have been caused by patients surviving with MLFS. Although significantly more patients achieved MLFS compared with IC, it did not translate into a significantly higher number of patients transitioning to SCT in the DEC10-VEN cohort. However, attaining MLFS does provide the opportunity to take such patients to transplantation because suboptimal responders can still experience OS benefit with SCT even in the absence of full-count recovery.16 Although our safety analysis focused on nonhematologic toxicity, some CRi and MLFS responses noted with DEC10-VEN can be potentially related to venetoclax-induced myelosuppression.

Based on these results, we have initiated two phase 2 trials of oral regimens of 10-day and 5-day decitabine/cedazuridine (ASTX727) with venetoclax for R/R AML (Clinical Trial Registration: NCT04746235, NCT04975919); however, prospective randomized studies are needed to confirm these findings. Ongoing evaluation of venetoclax with FLAG-IDA (Clinical Trial Registration: NCT03214562), CPX-351 (Clinical Trial Registration: NCT03629171), and CLIA (Clinical Trial Registration: NCT02115295) have shown promising early results and will help determine the optimal backbone for venetoclax in R/R AML.17, 18 However, for patients with multiply R/R AML with poor PS who are venetoclax-naïve and ineligible for intensive therapy, DEC10-VEN maybe a potential salvage option.

This was a retrospective study with all inherent limitations. However, propensity score matching for important effect modifiers and treatment in prospective clinical trials helped to minimize such bias, as noted from the low standardized mean differences after matching and similar baseline characteristics in the cohorts. Although propensity score matching can balance important effect modifiers, there could have still been other factors that influenced the differences observed. We had limited patients in the DEC10-VEN cohort for exploratory subgroup analysis, which resulted in large CIs decreasing the precision of our findings. All these trials were single-center studies and outcomes may have been different in other settings, particularly with other regimens like mitoxantrone, etoposide with cytarabine, cladribine, cytarabine with G-CSF, etc. Provision of two more cycles of treatment for response assessment with DEC10-VEN compared with IC introduced the potential for immortal time bias in our analysis. However, other retrospective results of venetoclax with HMA in R/R AML have shown identical or better outcomes supporting the reproducibility of our findings.19

In conclusion, this retrospective analysis suggests that DEC10-VEN may be considered as a potential salvage option offering outcomes comparable to non-venetoclax-based IC in patients with R/R AML who have not received venetoclax previously. The high ORR, rates of CRi, MLFS, MRD negativity, better OS, EFS, and its low-intensity nature makes this regimen a potential option as a bridge to SCT and as a reasonable choice for a low-intensity backbone for adding novel therapies in the salvage setting.

Funding Support

This study was supported in part by the MD Anderson Cancer Center (Support Grant CA016672) from the National Cancer Institute, and the Research Project (Grant Program R01CA235622) from the National Institutes of Health. Abhishek Maiti was supported in part by the American Society of Clinical Oncology Young Investigator Award from the Conquer Cancer Foundation.

Conflict of Interest

Abhishek Maiti received research funding from Celgene Corporation. Courtney D. DiNardo received personal fees from AbbVie, personal fees from Agios, Novartis, ImmuneOnc, Daiichi Sankyo, Celgene, Jazz Pharmaceuticals, and Notable Labs, from outside the submitted work. Elias J. Jabbour received consultancy research funding from Takeda, BMS, Adaptive, Amgen, AbbVie, Pfizer, and Cyclacel Ltd; and research grants with Amgen, AbbVie, Spectrum, BMS, Takeda, Pfizer, and Adaptive. Naval G. Daver reports grants from AbbVie, Genentech, Astellas, Daiichi-Sankyo, Pfizer, BMS, Immunogen, Novimmune, and Forty-Seven; and personal fees from AbbVie, Genentech, Astellas, Daiichi-Sankyo, Pfizer, BMS, Immunogen, Jazz Pharmaceuticals, Trillium, Forty-seven, Gilead, Kite, and Novartis. Nicholas J. Short reports grants from Takeda Oncology and Astellas, as well as personal fees from Takeda Oncology, AstraZeneca, and Amgen. Gautam Borthakur received research funding from AbbVie, Incyte, and Janssen. Naveen Pemmaraju reports personal fees from Pacylex Pharmaceuticals and Incyte; grants and other support from Affymetrix; grants from SagerStrong Foundation; personal fees and other support from Novartis; personal fees from LFB Biotechnologies, Roche Diagnostics, and Blueprint Medicines; personal fees, nonfinancial support, and other support from Stemline Therapeutics and AbbVie; personal fees and nonfinancial support from Celgene, MustangBio, and DAVA Oncology; other support from Samus Therapeutics, Cellectis, Daiichi Sankyo, and Plexxikon, outside the submitted work. Hagop M. Kantarjian received grants and other support from AbbVie, Agios, Amgen, Immunogen, and Pfizer; grants from Ariad, Astex, BMS, Cyclacel, Daiichi-Sankyo, Jazz Pharmaceuticals, and Novartis; and other support from Actinium and Takeda, outside the submitted work. Farhad Ravandi received personal fees and research grants from AbbVie. Hagtop M. Kantarjian received research grants from AbbVie, Amgen, Ascentage, BMS, Daiichi-Sankyo, Immunogen, Jazz Pharmaceuticals, Novartis, Pfizer, and Sanofi; and honoraria from AbbVie, Actinium (Advisory Board), Addaptive Biotechnologies, Amgen, Aptitude Health, BioAscend, Daiichi-Sankyo, Delta Fly, Janssen Global, Novartis, Oxford Biomedical, Pfizer, and Takeda Oncology. Marina Y. Konopleva received grants from the National Institutes of Health, the National Cancer Institute, AbbVie, Genentech, Stemline Therapeutics, Forty-Seven, Eli Lilly, Cellectis, Calithera, Ablynx, and Astra Zeneca; consulting/honorarium from AbbVie, Genentech, F. Hoffman La-Roche, Stemline Therapeutics, Amgen, Forty-Seven, and Kisoji; clinical trial support from Ascentage; and has stock in and receives royalties from Reata Pharmaceutical.

Author Contributions

Abhishek Maiti: Conception and design, data analysis and interpretation, and manuscript writing. Courtney D. DiNardo: Conception and design, administrative support, provision of study materials or patients, data analysis and interpretation, and manuscript writing. Wei Qiao: Conception and design, collection and assembly of data, and data analysis and interpretation. Tapan M. Kadia: Administrative support and provision of study materials or patients. Elias J. Jabbour: Administrative support and provision of study materials or patients. Caitlin R. Rausch: Provision of study materials or patients. Naval G. Daver: Provision of study materials or patients. Nicholas J. Short: Provision of study materials or patients. Gautam Borthakur: Provision of study materials or patients. Naveen Pemmaraju: Provision of study materials or patients. Musa Yilmaz: Provision of study materials or patients. Yesid Alvarado: Provision of study materials or patients. Kathryn S. Montalbano: Collection and assembly of data. Allison Wade: Collection and assembly of data. Rita E. Maduike: Collection and assembly of data. Julio A. Guerrero: Collection and assembly of data. Kenneth Vaughan: Collection and assembly of data. Carol A. Bivins: Collection and assembly of data. Sherry Pierce: Collection and assembly of data. Jing Ning: Conception and design, collection and assembly of data, and data analysis and interpretation. Farhad Ravandi: Conception and design, administrative support, and provision of study materials or patients. Hagop M. Kantarjian: Administrative support and provision of study materials or patients. Marina Y. Konopleva: Conception and design: Administrative support, provision of study materials or patients, data analysis and interpretation, and manuscript writing. All authors added critical revisions for important intellectual content. All authors reviewed and approved the final version of the manuscript.

Data Availability

Please contact the corresponding author with request for original patient-level data.