Predictors of clinical outcome in myeloproliferative neoplasm, unclassifiable: A Bone Marrow Pathology Group study
Genevieve M Crane,Julia T Geyer,Beenu Thakral,Sa A Wang,Geoffrey D Wool,Ke David Li,Adam R Davis,Leonardo Boiocchi,David Bosler,Carlos E Bueso-Ramos,Daniel A Arber,Tracy I George,Adam Bagg,Robert P Hasserjian,Attilio Orazi,Eric D Hsi,Heesun J Rogers
DOI: https://doi.org/10.1093/ajcp/aqae033
2024-04-10
American Journal of Clinical Pathology
Abstract:Abstract Objectives Myeloproliferative neoplasm, unclassifiable (MPN-U, revised to MPN, not otherwise specified in the fifth edition of the World Health Organization classification) is a heterogeneous category of primary marrow disorders with clinical, morphologic, and/or molecular features that preclude classification as a more specific MPN subtype due to stage at diagnosis, overlapping features between MPN subtypes, or the presence of coexisting disorders. Compared with other MPN subtypes, the contribution of the mutational landscape in MPN-U in conjunction with other clinical and morphologic biomarkers to prognosis has been less well investigated. Methods We performed a multicenter, retrospective study of MPN-U (94 cases) to better define the clinicopathologic features, genetic landscape, and clinical outcomes, including subgroups of early-stage, advanced-stage, and coexisting disorders. The Dynamic International Prognostic Scoring System (DIPSS) plus scoring system was applied to assess its relevance to MPN-U prognosis. Results Multivariate analysis demonstrated bone marrow blast count and DIPSS plus score as statistically significant in predicting overall survival. Univariate analysis identified additional potential poor prognostic markers, including abnormal karyotype and absence of JAK2 mutation. Secondary mutations were frequent in the subset analyzed by next-generation sequencing (26/37 cases, 70.3%) with a borderline association between high molecular risk mutations and overall survival. Conclusions This study, as one of the largest of MPN-U studies incorporating both clinicopathologic and molecular data, moves toward identification of biomarkers that better predict prognosis in this heterogeneous category.
pathology