Clinical outcomes of <i>IDH2</i>‐mutated advanced‐phase Ph‐negative myeloproliferative neoplasms treated with enasidenib

Abstract

Philadelphia chromosome-negative (Ph-negative) myeloproliferative neoplasms (MPNs) are clonal haematopoietic stem cell disorders with a variable incidence of transformation to the accelerated or blast phase (MPN-AP/BP). Historical survival of MPN-AP/BP patients has been poor with a median survival of 3–5 months regardless of treatment approach.1 Ivosidenib and enasidenib, inhibitors of the IDH1 and IDH2 mutant enzymes respectively, have provided a new treatment paradigm for IDH-mutated acute myeloid leukaemia (AML).2-5 Approximately 20% of MPN-AP/BP patients have mutations in IDH1/2,6, 7 but there are limited data on IDH inhibitor use in this patient population. Therefore, we conducted a retrospective analysis of IDH-mutated MPN-AP/BP patients treated with IDH inhibitors and report the outcomes. Patients with IDH1- or IDH2-mutated MPN-AP/BP treated with IDH inhibitors from 1/1/2009 till 5/14/2019 were identified using the University of Chicago Next-Generation Sequencing (NGS) database. Treatment response was assessed using the 2017 European LeukemiaNet (ELN) criteria8 and the 2012 Post-MPN AML Consortium (2012 MPN-BP) criteria.9 Adverse events were assessed using National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0. Detailed methods regarding gene mutation analysis, statistical analysis, and treatment response definitions are provided in Appendix S1. Of the 1 527 patients screened utilizing our NGS database, 96 harboured an IDH mutation, including eight IDH-mutated MPN-AP/BP patients that received an IDH inhibitor (Figure S1). Patient characteristics are summarized in Table I. All eight patients had IDH2 mutations; seven had a co-occurring JAK2 mutation and one had a co-occurring MPL mutation. Six received enasidenib (50–100 mg once daily) as initial treatment for MPN-AP/BP (three received enasidenib on clinical trial, including one patient who received azacitidine in conjunction with enasidenib; three received the agent off clinical trial) while two received enasidenib in the relapsed/refractory setting (one on clinical trial, one off clinical trial). In the sole patient who received combination therapy, azacitidine was ultimately discontinued due to persistent cytopenias and enasidenib monotherapy was continued. Haematology Median (range) Molecular characteristics at diagnosis of MPN-AP/BP Median (range) The overall response rate (ORR) was 37·5% with 2017 ELN criteria and 75% with 2012 MPN-BP criteria (Table SI). Regarding best response to enasidenib, two patients achieved complete acute leukaemia response (ALR-C), four achieved partial acute leukaemia response (ALR-P), one achieved stable disease (SD), and one had progressive disease (PD) (Fig 1A). Median overall survival (OS) for all eight patients was not reached with a median follow-up time of 272 days (26–1218+) (Fig 1B). Among the three patients reclassified as responders using 2012 MPN-BP criteria, OS for each patient was 67 days, 431+ days, and 1218+ days, respectively. The incidence of grade ≥3 adverse events included two patients with grade 5 differentiation syndrome (Table SII); one patient had interstitial lung disease and the other had familial transthyretin cardiac amyloidosis with heart failure as comorbid conditions. IDH2 mutant variant allele frequency (VAF) and canonical MPN mutation VAF at diagnosis of MPN-AP/BP are summarized in Table I. In addition to an IDH2 mutation and canonical MPN mutation, patients had a median of two co-occurring pathogenic mutations (range 1–6). TP53 was the most commonly found co-mutation (four patients) (Table I, Table SIII). Follow-up VAF data were available for four out of six patients that responded to enasidenib (Table SIV). In these responders, IDH2 became undetectable in two patients (assay sensitivity of 1%); the other two patients had persistently elevated VAFs of 47% and 60% respectively. At the time of best response, the canonical MPN mutation persisted and remained detectable in all four patients. Serial bone marrow examinations of patients that responded to enasidenib revealed that the chronic-phase MPN persisted throughout treatment (Figures S2 and S3). Our analysis of IDH2-mutated Ph-negative MPN-AP/BP patients that received enasidenib provides insight into the potential promise of this strategy for inducing durable responses that extend beyond the historically poor survival associated with MPN progression.1 The ORR was 75% using MPN-BP criteria and the median OS had not been reached with a median follow-up time of 272 days. Clinical trials evaluating IDH inhibitors both as frontline treatment for AML and in the relapsed/refractory setting have not specifically reported on outcomes in patients with antecedent MPNs; thus our experience may offer insight into the potential benefits of this treatment strategy.2-5 Traditional response criteria for AML may underestimate the benefit of IDH inhibition in MPN-AP/BP patients and currently there is a lack of standardization of criteria for response assessment in the treatment of MPN-BP.1, 9 ORR was 37·5% when using 2017 ELN criteria and 75% when using 2012 MPN-BP criteria. The median OS for three patients reclassified as responders was not reached, after a median follow-up of 431 days. Thus, traditional response criteria may not fully capture the benefit of IDH inhibitor therapy in patients with MPN-AP/BP. This may be due in part to the fact that traditional AML response criteria do not account for persistence of histopathological findings associated with chronic-phase MPNs. We also found both molecular and morphologic evidence that the chronic-phase component of disease persisted in all patients even at the time of best response (Table SIII, Figures S2 and S3). While we observed encouraging responses, we also noted an increased frequency of differentiation syndrome (DS). Grade ≥3 DS occurred in two patients (25%) in the setting of significant comorbid conditions. This highlights the potential risks associated with this targeted therapeutic approach. Our retrospective analysis suggests that the use of enasidenib in IDH2-mutated advanced-phase MPNs can offer long-lasting responses in some patients and provides a rationale for this treatment approach.1 This study is limited by the small sample size, single-centre location, and retrospective design which may have led to overestimation of both the efficacy and toxicity signals. The persistence of the chronic-phase MPN component suggests the need for prospective studies of rational IDH inhibitor-based combination therapies that target both the acute and the chronic phase of MPN-AP/BP, especially given pre-clinical work demonstrating cooperative efficacy with combined JAK2/IDH inhibition.10 AAP, KC, ACK and SK: none. HL: honoraria – Agios; research funding – BMS, Karyopharm; PI of clinical trials – Novartis, Arog. SG: honoraria – Up to Date. MT: honoraria – Up to Date; consultancy/advisory board/monitoring committee: AbbVie, Astra Zeneca, Celgene, Janssen, Roche/Genentech; research funding – AbbVie (Inst), Merck (Inst), Pharmacyclics (Inst), TG Therapeutics (Inst), Gilead Sciences (Inst). ASA: research funding – Miltenyi Biotech. RAL: consulting or advisory role – Agios, Celgene. WS: honoraria – Up to Date, Research to Practice; consulting or advisory role – Agios, Astellas, Daiichi, Kite, Pfizer. J.S: consulting or advisory role – Astra Zeneca, Merck, BMS; research funding – BMS (Inst), AbbVie (Inst). OO: consulting or advisory role – AbbVie, Celgene; research funding – Celgene (Inst), Incyte (Inst), Astra Zeneca (Inst), Astex Pharmaceuticals (Inst), NS Pharma (Inst), AbbVie (Inst), Gilead Sciences (Inst), Janssen Oncology (Inst), Oncotherapy (Inst), Agios (Inst), CTI (Inst). AAP and KC contributed equally to this manuscript and share first authorship. AAP, KC and OO designed research, performed research, analyzed data, wrote the manuscript. ACK, SG and JS provided pathology data, analyzed data and reviewed the manuscript. HL, MJT, SK, ASA, RAL and WS reviewed the manuscript and provided essential clinical input. Appendix S1. Supplementary materials and methods Fig S1. Patients with IDH1/2 mutations identified by next-generation sequencing. Fig S2. Pathologic characteristics of illustrative case #1. Fig S3. Pathologic characteristics of illustrative case #2. Table SI. Treatment response to enasidenib. Table SII. Frequency of adverse events (≥grade 3). Table SIII. Co-occurring mutations at diagnosis of MPN-AP/BP. Table SIV. Variant allele frequency (VAF) of IDH2 and canonical MPN mutations in patients with multiple NGS samples. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts