Multicenter evaluation of efficacy and toxicity of venetoclax‐based combinations in patients with accelerated and blast phase myeloproliferative neoplasms

Abstract

To the Editor: BCR-Abl-negative myeloproliferative neoplasms (MPNs), including essential thrombocythemia (ET), polycythemia vera (PV), and primary myelofibrosis (PMF) carry a risk for disease progression to an accelerated phase (AP-MPN) and/or blast phase (BP-MPN). AP-MPN is defined as the presence of 10%–19% peripheral or bone marrow blasts while BP-MPN is defined by ≥20% blasts in peripheral blood or bone marrow. AP/BP-MPN outcomes are poor, with median overall survival (OS) estimated at 3–4 months from the time of diagnosis. Traditional treatment strategies, including intensive chemotherapy and hypomethylating agents (HMAs), elicit inferior response rates in patients with AP/BP-MPN compared de novo acute myeloid leukemia (AML).1 The B-cell lymphoma 2 inhibitor, venetoclax (VEN) is approved in combination with either a HMA or low-dose cytarabine (LDAC) for initial therapy of AML in unfit or elderly patients. A recent Phase III study comparing VEN + azacitidine (AZA) versus AZA monotherapy for treatment-naïve (TN) AML patients ineligible for intensive therapy demonstrated that VEN + AZA significantly improved median OS (14.7 vs. 9.6 months). Notably, patients with a history of prior MPN were largely excluded, thus raising the question of applicability of these data to AP/BP-MPN patients.2 Retrospective series have assessed the clinical utility of VEN-based therapy in MPN patients. Gangat et al. conducted a retrospective, multicenter evaluation of 32 patients with treatment naïve (n = 23) and previously treated (n = 9) BP-MPN who received HMA + VEN. The rate of composite CR/CRi was 44% and median OS was 8 months. The authors suggested VEN + HMA produces a superior rate of response with minimal incremental toxicity when compared to historical controls treated with HMA.3 By contrast, a single-center retrospective review of 31 BP-MPN patients evaluated VEN in combination with HMA (58%) or a variety of other agents. Both relapsed or refractory (R/R) (n = 17) and TN (n = 14) patients were included. No responses were observed in the R/R setting and the median OS was three months. Significant treatment-related toxicity was observed and the 8-week mortality rate was 32%.4 Given the rarity of AP/BP-MPN and limited and conflicting data to-date with VEN-based therapies for AP/BP-MPN, we performed an IRB-approved, multicenter, retrospective analysis of treatment outcomes of 27 patients across three academic cancer centers in the United States (Memorial Sloan Kettering Cancer Center n = 12, Moffitt Cancer Center n = 10, Roswell Park Comprehensive Cancer Center n = 5). VEN-naïve patients with a diagnosis of either R/R or TN AP-MPN or BP-MPN treated with at least seven days of VEN in combination with LDAC, decitabine (DEC), or AZA between January 1, 2016 and January 7, 2020 were included. Best response was determined by retrospective review of bone marrow biopsy reports and peripheral blood studies by the local investigators. Responses were confirmed by two independent reviewers. Patients were evaluable for a response within 42 days of initial VEN dose. Overall response rate (ORR) was considered attainment of a complete molecular response, complete cytogenetic response (CCR), acute leukemia response-complete (ALR-C), or acute leukemia response-partial (ALR-P) by MPN-BP Consensus Criteria.5 EFS was defined as the time between VEN initiation and date of disease progression or death. OS was defined as the time between VEN initiation and date of death. OS and EFS were evaluated by Kaplan–Meier method and the difference between groups was determined by log rank test. Associations between ORR and patient and disease characteristics were tested by Fisher's exact test and odds ratios (ORs) were estimated by logistic regression. The effects of patient and disease characteristics on OS and EFS were estimated by univariate Cox proportional hazard model with p < .05 being considered significant. All statistical analyses were performed using R. A total of 27 subjects were evaluated, including (n = 21, 77%) BP-MPN and (n = 6, 22%) AP-MPN. The median age was 72 years (range, 56–82). Baseline characteristics are described in Table 1. Blast phase patients (n = 21): European LeukemiaNet (ELN) risk stratification for BP-MPN patients consisted of intermediate (n = 6, 29%) and adverse/poor risk (n = 15, 71%). Mutations in JAK2, ASXL1, TET2, and RUNX1 were the most common, observed in (n = 14, 66%), (n = 8, 38%), (n = 8, 38%), and (n = 7, 33%) of patients with available data, respectively. Antecedent MPN was evenly distributed between ET, PV, and MF. Eight patients (38%) received VEN-based therapy as first line treatment, five patients (24%) received prior HMA, and two patients were post alloSCT (9.5%). VEN was given with concurrent AZA 75 mg/m2 × 7 days (n = 9, 43%), DEC 20 mg/m2 × 5 days (n = 7, 33%), LDAC 20 mg/m2 × 10 days (n = 3, 14%), and AZA 75 mg/m2 × 5 days (n = 2, 10%). ORR was 52.3%: ALR-C (n = 5, 24%), ALR-P (n = 4, 19%), and CCR (n = 2, 10%). Further details of responses are described in Figure 1A. Median time to best response was 29.5 days (range, 7–114 days) and median response duration was 87 days (range, 21–447). Disease relapse occurred in 6 of the 11 responders by the end of study evaluation period. Median OS was 6 months (95% confidence interval [CI], 4.1–NR), with significantly inferior survival for patients not receiving AZA (hazard ratio [HR], 17.9; 95% CI, 2.34–136.98; p = .001) and patients with a TP53 mutation (HR, 3.42; 95% CI, 1–11.8; p = .03). NRAS mutations trended toward inferior OS but did not meet statistical significance (p = .05). Patients that received prior alloSCT had a markedly inferior median OS of 0.85 months (p = .019) compared to the rest of the cohort (Figure 1B,C). Median EFS was 0.9 months (95% CI, 0–NR). EFS was inferior for patients on LDAC (HR, 13.47; 95% CI, 2.27–79.84; p = .004) or DEC (HR, 4.72; 95% CI, 1.21–18.4; p = .03) when compared to AZA and those with poor cytogenetic risk (OR, 2.7; 95% CI, 1–7.43; p = .04). Patients who received DEC- and LDAC-based therapy were also less likely to respond compared to AZA-based therapy (p = .01). NRAS and ASXL1 mutations were associated with numerically lower response rates which trended toward statistical significance (p = .05). Notably, line of therapy, antecedent MPN, prior HMA, or drug cessation for toxicity did not affect EFS, OS, or response rate. Three of the four patients who proceeded to alloSCT post VEN (n = 3 ALR-C, n = 1 ALR-P) were alive at the end of the study evaluation period. Seven patients (33%) warranted VEN interruption for nonhematologic toxicity and 11 patients (52.4%) required VEN interruption for severe myelosuppression. Six patients (28%) had documented infections, including four fatal infection-related events [hospital acquired pneumonia (n = 2) and Gram-negative bacteremia (n = 2)]. Four patients (19%) developed grade 3 bleeding events including: Gastrointestinal bleed (n = 1) subdural hemorrhage (n = 1), hematuria (n = 1), and epistaxis (n = 1); no deaths were attributed to bleeding events. There were no documented cases of clinical or laboratory tumor lysis syndrome (TLS). Three patients died within 60 days of VEN initiation. Accelerated phase patients (n = 6): Of the six AP-MPN patients, antecedent MPN included ET (n = 3, 50%) and PMF (n = 3, 50%). VEN was given with concurrent DEC 20 mg/m2 × 5 days (n = 3, 50%) or AZA 75 mg/m2 × 7 days (n = 3, 50%); three patients received prior HMA. ORR was 50%, with all responders obtaining ALR-C (n = 2 post PMF, n = 1 post ET); further details of responses are included in Figure 1D. The median time to best response was 28 days (7–58 days) and median duration of response was 55 days. Median EFS was 0.93 months and median OS was 3.58 months. Half of the cohort (n = 3, 50%) was hospitalized for neutropenic fever, but no infections, TLS, or bleeding events were documented. No patients died within 60 days of treatment. Two patients were alive at the end of the study period with one patient proceeding to alloSCT. We determined that VEN-based combination treatment has modest clinical activity in BP and AP-MPN patients but is associated with notable toxicities. Recent data from Gangat et al. demonstrated an overall CR/CRi rate of 44% using ELN criteria with CR/CRi observed in 48% and 33% of TN and R/R patients, respectively. By contrast, Masarova et al. reported CR/CRi rate of 43% in TN but 0% in R/R patients.3, 4 Our data demonstrated an ORR of 53% in BP-MPN, with ORR of 28% in R/R patients, and 50% in AP-MPN. In our study, response did not appear to translate into a survival benefit, with a median OS appreciably less than 1 year. Similar to the findings of Gangat et al., alloSCT appears to be important for long-term survival of responding BP-MPN patients.3 Our findings suggest that the risk benefit profile of VEN-based therapy in both AP/BP-MPN patients should be carefully considered, noting an appreciable rate of infectious complications and bleeding events. Limitations to this study include the retrospective nature of the study and limited sample size. As well, our use of MPN Consensus criteria may offer variation in interpretation across other clinical studies. VEN-based combinations have activity in patients with AP/BP-MPN, demonstrating the ability to facilitate alloSCT in a small portion of patients. However, despite the ability of VEN-based regimens to achieve responses, the short duration of response and noted toxicities may limit the overall clinical benefit in this population. Globally, outcomes for AP/BP-MPN patients remain dismal, and further investigational studies are direly needed for this patient population. This study was supported by Cancer Center Support Grant/Core Grant to Memorial Sloan Kettering Cancer Center (P30 CA008748); Raajit K. Rampal is supported by National Cancer Institute P01 CA108671 11. Amber C. King has received advisory/consulting fees from: Astellas, Abbvie, and Blueprint Medicines. Raajit K. Rampal has received consulting fees from: Constellation, Novartis, Kartos, Incyte, Celgene/BMS, Promedior, CTI, Blueprint, Stemline, Galecto, Pharmaessentia, Abbvie and research funding from Incyte, Constellation, Stemline, and Zentalis. Aaron D. Goldberg has served as a consultant or on advisory boards for AbbVie, Aptose, Astellas, Celgene, Daiichi Sankyo, and Genentech, has received research funding from AbbVie, Aptose, Cellularity, ADC Therapeutics, Aprea, AROG, Pfizer, and Prelude, and has received honoraria from Dava Oncology. Maximilian Stahl has consulted for the Boston Consulting Group. Michael J. Mauro has disclosed the following relationships: Novartis; consulting/advisory/research support (institution), Takeda; consulting/advisory/research support (institution), Sun Pharma/SPARC; research support (institution), Bristol Myers Squibb; consulting/advisory, Pfizer: consulting/advisory. Martin S. Tallman has disclosed the following relationships: Abbvie: Research Funding; Cellerant: Research Funding; Orsenix: Research Funding; ADC Therapeutics: Research Funding; BioSight: Membership on an entity's Board of Directors or advisory committees and Research Funding; Glycomimetics: Research Funding; Rafael: Research Funding; Amgen: Research Funding; Bioline rx: Membership on an entity's Board of Directors or advisory committees; Daiichi Sankyo: Membership on an entity's Board of Directors or advisory committees; KAHR: Membership on an entity's Board of Directors or advisory committees; Rigel: Membership on an entity's Board of Directors or advisory committees; Delta Fly Pharma: Membership on an entity's Board of Directors or advisory committees; Oncolyze: Membership on an entity's Board of Directors or advisory Committees; Jazz Pharma: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees; UpToDate: Patents & Royalties. Eunice S. Wang has received consulting fees from Abbvie, Amgen, Astellas, BMS, Genentech, Gilead, GlaxoSmithKline, Jazz, Kite, Kura, Mana Therapeutics, Novartis, Stemline, Takeda; speaker roles for Astellas, Dava Oncology, Kura, Pfizer, Stemline; Data monitoring committees for Abbvie and Rafael Pharmaceuticals. Andrew T. Kuykendall has received consulting feels/honoraria from Blueprint, Novartis, Incyte, Celgene/BMS, Abbvie, CTI Biopharma, Pharmaessentia, Protagonist and research funding from Protagonist, Prelude, and Sierra. The remaining authors have no relevant disclosures. Amber C. King, Taylor M. Weis, Andriy Derkach, and Raajit K. Rampal analyzed the data and wrote the manuscript draft. Amber C. King, Taylor M. Weis, and Raajit K. Rampal designed the research study. All authors were involved in review and revision of the manuscript. All authors approved the final version. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Topics & Concepts