Mast cell cytomorphology and treatment outcome in mast cell leukemia

Systemic mastocytosis (SM) in adults almost always involves the bone marrow (BM) and less frequently, the skin and other organs, including the liver, spleen, and bones.1 The extent of neoplastic mast cell (MC) proliferation (i.e., MC burden), MC-associated organopathy, and co-occurrence of an associated myeloid neoplasm (AMN) determine disease severity and allow subclassification into indolent SM (ISM), smoldering SM (SSM), aggressive SM (ASM), SM-AMN, and mast cell leukemia (MCL).2 According to the revised fourth and the fifth editions of the World Health Organization (WHO) classification system (WHO-2022),3, 4 diagnosis of MCL requires the presence of ≥20% MCs in BM aspirate smear, without reference to MC cytomorphology. By contrast, the first edition of the International Consensus Classification (ICC-2022)2, 5 requires the presence of atypical/immature MCs, in addition to the ≥20% threshold, for diagnosis of MCL.2 Unlike the case with WHO-2022, ICC-2022 no longer distinguishes between leukemic (circulating MC ≥ 10%) and aleukemic variants of MCL, and has also eliminated the use of the term “chronic MCL,” which refers to the mature morphology seen in some cases of MCL.6 In a recent communication,7 we examined survival outcomes in 16 patients with MCL, with the main objective of determining the impact of “immature” MC cytomorphology2 on overall survival; after independent pathology review, eight patients were classified as having “immature” and eight “mature” MC cytomorphology; immature MC cytomorphology clustered with CD2 (86% vs. 29%) and CD25 expression (100% vs. 75%) while mature MC cytomorphology clustered with KIT D816V (86% vs. 50%); after a median follow-up of 3.9 months, only four patients were alive at 0.5 (immature lost to follow-up), 14 (immature on avapritinib), 23 (mature under no treatment), and 180 months (mature post-transplant); in multivariable analysis, immature MC morphology (hazard ratio [HR] 7.4) and absence of KIT D816V (HR 6.8) were independently associated with inferior survival and their combined consideration allowed significantly different survival projections; KIT D816V mutation-negative or immature morphology versus mutation-positive and mature morphology (median survival 0.9 vs. 33 months; HR 13.1). The main objective of the current study was to provide a detailed account of treatment outcomes in the abovementioned study and examine the impact of MC cytomorphology on treatment outcomes. Mayo Clinic hematopathology and clinical databases were reviewed to identify MCL cases, after institutional review board approval. Diagnosis of MCL followed the revised fourth edition WHO classification criteria,3 in order to capture all cases with ≥20% MCs in BM aspirate smear, regardless of MC cytomorphology. Clinical and laboratory data were collected at the time of MCL diagnosis. BM slides were reviewed by three Mayo Clinic hematopathologists (KKR, KK, and DC), with each slide set being independently reviewed by at least two pathologists, using prespecified criteria. In addition, digitized images of the selected slides were reviewed by an expert external hematopathologist (AO). MCs were cytomorphologically classified as being “mature” or “immature” (promastocytes, metachromatic blast-like forms, multinucleated or highly pleomorphic MCs), as previously described.5, 8 Patient follow-up was updated in August 2023; data were censored at last follow-up. Conventional statistical methods were applied using JMP Pro 16.0.0 software (SAS Institute, Cary, NC). Sixteen cases of MCL were identified: 37% de novo and 63% with prior history of SM; 8 (50%) were classified as “MCL-immature,” using ICC-2022 criteria,2 (median age 66 years; 50% females) and 8 (50%) “MCL-mature” (median age 61 years; 50% females). Antecedent or AMN was more frequent in patients with mature versus immature cytomorphology (50% vs. 0%; p < .01). On the other hand, immature MC cytomorphology clustered with absolute eosinophil count >0.5 × 109/L (50% vs. 13%; p = .09; median 1.2 vs. 0.06 × 109/L), circulating MCs (63% vs. 25%; p = .13), subnormal serum albumin (67% vs. 25%; p = .11; median 3.2 vs. 3.9 g/dL), while mature cytomorphology clustered with KIT D816V (86% vs. 50%; p = .15); the two morphology groups were otherwise similar in their expression of palpable splenomegaly (50% vs. 83%; p = .2), hepatomegaly (50% vs. 75%; p = .3), abnormal karyotype (13% vs. 29%; p = .4), prior history of SM (63% vs. 63%; p = 1.0), biopsy-proven CM (25% vs. 25%; p = 1.0), weight loss (38% vs. 38%; p = 1.0), C-findings (75% vs. 75%; p = 1.0), anemia with hemoglobin <10 g/dL (63% vs. 63%; p = 1.0), platelet count <100 × 109/L (38% vs. 50%; p = .61; median 139 vs. 103 × 109/L), leukocyte count (median 10.5 vs. 7.5 × 109/L; p = .8), absolute monocyte counts (median 0.5 vs. 0.4 × 109/L; p = .6), serum tryptase > 500 ng/mL (57% vs. 50%; p = .8; median 1055 vs. 389 ng/mL), and increased serum alkaline phosphatase (80% vs. 63%; p = .5; median 171 vs. 142 U/L), respectively. On multivariable analysis, immature MC morphology (p = .03; HR 7.4, 95% confidence interval [CI] 1.1–51.3) and absence of KIT D816V (p = .04; HR 6.8, 95% CI 1.0–49) were independently predictive of worse survival. As depicted in Figure 1A, six (75%) of eight patients with MCL-immature were either deceased (N = 5) or lost to follow-up (N = 1) in the first 30 days after diagnosis; treatments given during this short period included solumedrol, imatinib, and in one case, one cycle of cladribine, all without benefit. Only two (25%) patients with MCL-immature were known to survive beyond 1 year of initial diagnosis; patient 2 remains alive at 14 months from diagnosis and had received, sequentially, imatinib × 15 days, midostaurin × 30 days, cladribine × 2 cycles, and avapritinib × 9+ months (Figures 1A and 2); MCL-related symptoms were substantially improved by midostaurin and subsequently sustained by cladribine and avapritinib (100 mg/day), but BM MC burden remained unaltered (80%–90%) until a higher dose of avapritinib (200 mg/day) was implemented, resulting in MC burden decrease to 20% (Figure 2). Patient 1 with MCL-immature experienced symptomatic relief from midostaurin for 6 months before receiving ASCT; unfortunately, post-transplant BM biopsy at 6 months showed persistent MCL and midostaurin was restarted and continued until the patient's demise at 16 months from MCL diagnosis. Unlike the case with MCL-immature, six (75%) of eight patients with MCL-mature survived beyond the first year of diagnosis and four (50%) beyond 2 years from diagnosis (Figure 1B). The longest living patient with MCL-mature (patient 9) had received ASCT within 1 year of diagnosis after a brief treatment period with imatinib, IFN-a, and decitabine for associated MDS/MPN. Another long-lived patient with MCL-mature (patient 10) was treated with multiple cycles of cladribine before and after MCL diagnosis with documentation of a favorable effect on BM MC burden. Two patients with MCL-mature survived beyond 2 years, with one patient still alive with no active treatment (patient 12; Figure 1B). Two patients with MCL-mature died early within 1 month of diagnosis and received either lenalidomide or AML-like induction chemotherapy. Two patients (patients 1 and 9) received ASCT for treatment of MCL while one (patient 8) had received ASCT for antecedent SM-AMN, before progression into MCL; of these three patients, one (patient 9) with MCL-mature (KIT D816V+) remains alive at 15 years from diagnosis of MCL and 14 years from time of ASCT (Figure 1B). The second patient (patient 1) with ASCT died 16 months from diagnosis of MCL-immature/KIT D816V− and 10 months from time of ASCT, with BM examination performed at 6 months post-transplant showing persistent neoplastic MCs. The observations in the current study highlight the prognostic as well as therapeutic relevance of distinguishing MCL-immature from MCL-mature disease, always in association with establishing KIT D816V mutational status. The presence of immature MC cytomorphology or absence of KIT D816V should be regarded as a marker of aggressive disease biology that should be treated as an oncologic emergency, requiring immediate institution of MC-directed therapy. A similar treatment strategy is also advised and expected to perform better, in the setting of MCL-mature. The choice between the aforementioned MC-directed drugs depends on a number of factors: (i) drug access and availability for rapid institution; (ii) presence of thrombocytopenia, where the use of avapritinib is relatively contraindicated; (iii) presence of drug–drug interactions; (iv) presence of comorbid conditions that might dictate route of administration (intravenous vs. oral); (v) dosing uncertainty in the presence of renal or hepatic dysfunction; (vi) objective of treatment (palliative vs. deep clearance of BM MCs in preparation for ASCT); and (vii) out-of-pocket cost to patients. Our experience suggests that avapritinib, midostaurin, and cladribine are all effective in alleviating MC-associated complications in MCL, while avapritinib might deliver better in terms of BM MC clearance, which might come in handy for bridging patients to ASCT. The superior performance of avapritinib in reducing BM MC burden appeared to be dose-dependent, but also truncated because it failed to induce complete eradication of neoplastic MCs. In previously published phase-1 and phase-2 studies of avapritinib in MCL, complete or partial remissions were reported in 23% and 25%–31% of patients, respectively9, 10; similar response rates (50%) were also reported for midostaurin11; details on MC cytomorphology were not provided in either the avapritinib or midostaurin clinical trials. Regardless, it is unlikely that drug therapy alone is adequate for long-term survival in either MCL-immature or MCL-mature and that ASCT should be pursued as consolidation therapy; in the current series, two patients received ASCT, with long-term survival (15 years+) documented in one patient with MCL-mature/KIT D816V+ and significantly shorter survival (14 months) in the second patient with MCL-immature/KIT D816V− disease. Whether or not MC cytomorphology affects the outcome of MCL in patients undergoing ASCT or receiving KIT-targeted kinase inhibitors requires additional studies in larger cohorts of informative cases. This is of critical importance because previous reports on the subject matter have not accounted for MC cytomorphology and might have, therefore, over- or underestimated survival rates stratified by MCL-mature versus MCL-immature.12 The same can be said regarding KIT mutational status. Our observations underscore the need to abide by the 2022 ICC requirement for the presence of atypical immature MCs (i.e., promastocytes, metachromatic blast-like forms, and multinucleated or highly pleomorphic MC), for the diagnosis of MCL, while what is referred to as “MCL-mature” might need to be reclassified as ASM or SM-AMN, with a notation of excess BM MCs.2, 13 Animesh Pardanani, Ayalew Tefferi, Aref Al-Kali, Mrinal Patnaik, William J. Hogan, Kebede Begna, Michelle A. Elliott, Jeanne M. Palmer, Nandita Khera, and Naseema Gangat participated in patient care. Attilio Orazi, Katalin Kelemen, Dong Chen, and Kaaren K. Reichard provided hematopathology expertise. Dong Chen and Kaaren K. Reichard provided central review of all cases and prepared all images. Animesh Pardanani, Ayalew Tefferi, Kaaren K. Reichard, Dong Chen, and Attilio Orazi participated in concept and design of the study. Ayalew Tefferi and Animesh Pardanani wrote the paper. All authors reviewed and approved the manuscript. The authors declare no conflicts of interest. The data are available on request.