Morphologic Assessment of Bone Marrow Samples in AML Patients Treated with Inhibitors to IDH1 and IDH2—a Single Center Experience

Abstract

INTRODUCTION:Mutations in isocitrate dehydrogenase 1 (IDH1) and isocitrate dehydrogenase 2 (IDH2) are reported in about 8% and 13% of patients with acute myeloid leukemia (AML), respectively, and are less common in patients with myelodysplastic syndromes. These recurrent mutations in AML lead to neomorphic enzymatic activity resulting in accumulation of 2-hydroxyglutarate, DNA and histone hypermethylation, blockage of cellular differentiation and, ultimately, leukemogenesis. Inhibitors to both IDH1 and IDH2 are being developed and have shown clinically significant responses. Perhaps unsurprisingly, these IDH inhibitors appear to promote the differentiation of myeloblasts.Serial bone marrow biopsies are considered the standard of care for the evaluation of the effectiveness of AML-directed therapies. However, due to the unique mechanism of action of IDH1 and IDH2 inhibitors, the morphologic assessment of response to these drugs may be challenging to pathologists and hematologists who may not be familiar with their use. Here we describe our single-center experience with IDH1 and IDH2 inhibitors with the goal of improving our understanding of the bone marrow topography in treated patients.METHODS:This retrospective study reports on all patients at our institution with refractory or relapsed AML who have been treated with either an IDH1 inhibitor (FT-2102) on a clinical protocol (NCT02719574) or an IDH2 inhibitor (AG-221, Enasidenib) on a compassionate use basis. A bone marrow biopsy and aspiration was performed on all patients after each 4-week cycle of therapy. All baseline and on-treatment bone marrow specimens were reviewed by a dedicated hematopathologist. Due to the small sample size, which is reflective of the low prevalence of this particular patient population, reported results are primarily descriptive in nature.RESULTS:A total of three patients were treated with an IDH1 inhibitor (cases 1,2,3), one of whom was treated in combination with azacitidine (case 3). Two patients were treated with an IDH2 inhibitor (cases 4,5). One patient (case 5) is still undergoing early treatment and therefore a full assessment of response and bone marrow morphology is not included at the time of this writing. Ages ranged from 53 to 71 years. All patients had relapsed and/or refractory AML with a median of three prior lines of therapy. Complete response (CR) to IDH1 or IDH2 inhibitor therapy was seen in 3 out of the 4 evaluable patients (cases 2,3,4). Two of the CRs were associated with incomplete platelet count recovery (CRp).In 3 of 4 patients, the end of cycle 1 bone marrow was markedly hypercellular (>90%) with sheets or large aggregates of myeloblasts suggestive of refractory disease. In case 1, the IDH inhibitor therapy was discontinued. In the other two cases, therapy was continued and CRp was eventually achieved after 2 cycles in case 2 and after 3 cycles in case 4. In cases 2 and 3 the morphology in the bone marrow at the time of remission was considered markedly abnormal by the pathologist. In case 2, the marrow was markedly hypercellular (>95%) with a profound myeloid expansion and increased reticulin fibrosis reminiscent of a myeloproliferative disorder. In case 3, the marrow was again markedly hypercellular (>95%) with marked myeloid and megakaryocytic hyperplasia and significant dysplastic features in all three lineages.CONCLUSION:Bone marrow assessments in patients on IDH inhibitor therapy can be challenging. Clinicians should be aware that responses may be delayed and early discontinuation of therapy may be premature particularly in patients who otherwise appear to be deriving clinical benefit. Pathologists should be made aware of this new class of drugs with a novel mechanism of action which can produce unusual bone marrow morphologies, presumably due to a differentiating effect on leukemic myeloblasts. This may mimic other myeloid malignancies such as myelodysplasia and myeloproliferative neoplasms. Thus, communication between the treating hematologist and interpreting pathologist is important for accurate response evaluations and optimal patient outcomes. Results will be updated at the ASH Annual Meeting including results from case 5 and photomicrographs of representative marrow specimens. DisclosuresNo relevant conflicts of interest to declare.