The presence of SH2B3 mutations identifies a distinct subtype of myelodysplastic neoplasms
Abstract
Abstract Background & aim: While SH2B3 gene mutations have been identified in myeloproliferative neoplasms, where they negatively regulate the TPO/MPL/JAK2 pathway, their pathogenic mechanism and clinical implications in MDS remain unknown. This study aimed to analyze the clinical features, prognostic impact, and functional role of SH2B3 mutations in patients with myelodysplastic neoplasms (MDS). Methods: The study cohort included MDS patients according to WHO 2016 criteria from 19 international centers, harboring SH2B3 mutations (MDS-SH2B3mut) with a variant allele frequency (VAF) < 40%, or VAF ≥ 40% if germline origin was excluded. The control group included MDS patients without SH2B3 mutations (MDS-SH2B3wt), obtained from the Molecular International Prognostic Scoring System (IPSS-M) dataset (Bernard et al., NEJM Evid. 2022). Clinical and molecular data collected at diagnosis (including NGS of myeloid gene mutations), and outcome variables were compared between MDS-SH2B3mut and MDS-SH2B3wt patients. Prognostic risk was assessed using IPSS-R and IPSS-M. Associations with acute myeloid leukemia (AML) progression, overall survival (OS) and leukemia-free survival (LFS) were evaluated. Variables included in the multivariate analysis were SH2B3 mutational status (MDS-SH2B3mut vs MDS-SH2B3wt ) and risk classification: lower-risk, defined as IPSS-R ≤3.5 points or IPSS-M very-low, low, or moderate-low versus higher-risk, defined as IPSS-R >3.5 points or IPSS-M moderate-high, high, or very-high. Statistical analyses were performed using R. Results: 92 MDS-SH2B3mut and 1.842 MDS-SH2B3wt patients were included. Mutated patients were significantly older than MDS-SH2B3wt (median 74 (IQR 68–80) vs. 70 (IQR 64–79) years; p<0.01). A lower proportion of MDS-SH2B3mut cases were classified as MDS with excess blasts-2 (9.8% vs 18.0%; p=0.06) compared to MDS-SH2B3wt. Moreover, high and very-high IPSS-R cytogenetic categories were less frequent in MDS-SH2B3mut patients than MDS-SH2B3wt (4.5% vs. 12.2%; p<0.01). Consequently, fewer MDS-SH2B3mut cases were classified into high and very-high risk groups by IPSS-R (9.8% vs. 24.0%, p<0.02). Median VAF of SH2B3 mutations was 11.8% (IQR 5.6–24.2). Among the 99 SH2B3 variants identified, the most frequent were frameshift (40%) and missense (39%). Notably, 43% of the frameshift variants clustered in exon 2, which encodes the Phe_ZIP domain that is critical for SH2B3-mediated inhibition of cell proliferation. In MDS-SH2B3mut patients, median number of co-ocurring mutations was 5 (IQR 3–6). Compared with MDS-SH2B3wt, MDS-SH2B3mut patients showed significantly higher mutation rates in TET2 (54% vs. 29%), SRSF2 (23% vs. 15%), ZRSR2 (17% vs. 5%), CBL (10% vs. 4%), JAK2 (10% vs. 2%), and PHF6 (9% vs. 3%). Moreover, 38% of MDS-SH2B3mut patients presented TET2–SRSF2 co-mutations (HR 12; p<0.01), showing similarities with the molecular landscape of CMML. Interestingly, preliminary data from ex vivo myeloid-erythroid differentiation assays suggest that SH2B3 knockdown in primary CD34⁺ hematopoietic stem/progenitor cells promote a shift toward monocytic differentiation, aligning with the molecular features observed in MDS-SH2B3mut cases. MDS-SH2B3mut patients exhibited better outcomes compared to MDS-SH2B3wt. LFS was longer in MDS-SH2B3mut [median 59.7 months (95%CI 43.1–NR) vs. 37.8 months (95%CI 29.9–50.1); p=0.01]. OS was also longer in MDS-SH2B3mut patients [median 89.6 months (CI95% 44.4–NR) vs. 47.3 months (95%CI 37.7–55.1); p=0.05]. In the multivariate analysis for LFS including SH2B3 mutational status and IPSS-M, both SH2B3 mutations (HR 0.6; 95%CI 0.4–0.9; p=0.01) and lower-risk IPSS-M categories (HR 0.30; 95%CI 0.22–0.39; p<0.01) emerged as independent favorable prognostic factors. Similarly, in the model including IPSS-R, SH2B3 mutations (HR 0.5; 95%CI 0.3–0.9; p=0.02) and lower-risk IPSS-R categories (HR 0.4; 95%CI 0.3–0.5; p<0.01) were independently associated with improved LFS. Conclusions: Our study, the first to comprehensively characterize MDS patients harboring SH2B3 mutations, identifies a distinct subtype of MDS with CMML-like molecular features and more favorable outcomes.
Article Details
Authors (41)
Maria Julia Montoro
1Vall d'Hebron University Hospital, Vall d'Hebron Institute of Oncology, Hematologic Genetic Diagnosis and Counseling Unit, Barcelona, Spain
Pamela Acha
1Vall d'Hebron University Hospital, Vall d'Hebron Institute of Oncology, Hematologic Genetic Diagnosis and Counseling Unit, Barcelona, Spain
Franceso Versino
2Kings College Hospital, London, United Kingdom
Paula Garcia Olloqui
3Clínica Universidad de Navarra - CIMA - CCUN, Hematologic Oncology Program, Navarra, Spain
Manja Meggendorfer
46Munich Leukemia Laboratory, Munich, Germany
Najla Al Ali
9Moffitt Cancer Center, Department of Malignant Hematology, Tampa, United States
Raja Prince-Eladnani
18Allegheny Health Network, Pittsburgh, United States
Gabriela Fernandez-Ruz
3Clínica Universidad de Navarra - CIMA - CCUN, Hematologic Oncology Program, Navarra, Spain
Sherry Pierce
1MD Anderson Cancer Center, Leukemia, Houston, United States
Carlos Bravo-Perez
1Translational Hematology & Oncology Research, Cleveland Clinic, Cleveland, OH
Marie Adelaine Migeon
9Hôpital St Louis/université de Paris, Département (DMU) d'hématologie et immunologie, APHP Nord Service d'hématologie séniors, Paris, France
Julia Mestre
10Institut Català Oncologia-Hospital Germans Trias i Pujol, Josep Carreras Leukemia Research Institute, Universitat Autònoma de Barcelona, Spain, Department of Hematology, Badalona, Spain
Klaus Metzeler
4Department of Hematology, Cell Therapy, Hemostaseology and Infectious Diseases, University of Leipzig, Leipzig, Germany
Michael Wulfert
12University Medical Center Düsseldorf, Heinrich-Heine-University, Department of Hematology, Oncology and Clinical Immunology, Düsseldorf, Germany
Andres Jerez
1Vall d'Hebron University Hospital, Vall d'Hebron Institute of Oncology, Hematologic Genetic Diagnosis and Counseling Unit, Barcelona, Spain
Kar Lok Kong
2King's College Hospital, London, United Kingdom
Marina Díaz-Beyá
7Hospital Clínic Barcelona, Hematopathology Section, Barcelona, Spain
Beate Betz
6University Medical Center Düsseldorf, Heinrich-Heine-University, Department of Hematology, Oncology and Clinical Immunology, Düsseldorf, Germany
Emmanuelle Clappier
2Institut de Recherche Saint-Louis, Université Paris Cité, Paris, France
Mar Tormo
Hospital Clinico Universitary. INCLIVA Research Institute, Valencia 46010, Spain
Helena Pomares
1Institut Català d'Oncologia - Hospital Duran i Reynals, Hematology, Hospitalet de Llobregat, Spain
Mónica Del Rey
8Department of Hematology. Salamanca-IBSAL University Hospital., Salamanca, Spain
Silvia Saumell
1Vall d'Hebron Hospital, Department of Hematology, BARCELONA, Spain
Laura Barberó
18Hospital Ramón y Cajal, Department of Hematology, Madrid, Spain
Maria Eugenia Rivero
2Arnau de Vilanova University Hospital, Department of Hematology, Lleida, Spain
Maria Diez-Campelo
11Hospital Clínico Universitario de Salamanca, Salamanca, Spain
David Sallman
Moffitt Cancer Cancer and Research Institute, Tampa, Florida, United States
Weina Chen
Department of Pathology, University of Texas Southwestern Medical Center
Zhuoer Xie
Moffitt Cancer Center, Tampa, Florida, United States
Francesc Bosch Albareda
5Vall d'Hebron University Hospital, Vall d'Hebron Institute of Oncology, Barcelona, Spain
Ulrich Germing
Uwe Platzbecker
Francesc Sole
8Institut de Recerca Contra la Leucèmia Josep Carreras, Barcelona, Spain
Pierre Fenaux
Jaroslaw Maciejewski
1Department of Translational Hematology and Oncology Research, Cleveland, United States
Guillermo Garcia-Manero
Rami Komrokji
Moffitt Cancer Cancer and Research Institute, Tampa, Florida, United States
Torsten Haferlach
7Munich Leukemia Laboratory, Munich, Germany
Austin Kulasekararaj
22King's College Hospital NHS Foundation Trust, London and King's College London, London, United Kingdom
Teresa Ezponda
7Department of Oncology-Hematology, CIMA Universidad de Navarra-IDISNA-CCUN. Centro de Investigación Biomédica en Red de Cáncer, CIBERONC. Clínica Universidad de Navarra, Pamplona, Spain
David Valcárcel