Clinical-pathological features of triple-negative essential thrombochytemia

Background: Lack of demonstrable mutations affecting driver genes JAK2, CALR or MPL within the spectrum of BCR-ABL1-negative myeloproliferative neoplasms (MPN) is currently referred as triple-negative (TN) genotype, which is most commonly found in the setting of essential thrombocythemia (ET). ET represents an MPN with good prognosis, whose differential diagnosis is crucial, particularly versus reactive thrombocytosis and, most notably, pre-fibrotic primary myelofibrosis (PMF), a condition which significantly impacts patient’s management for its increased risk of progression. Aims: In the present study, we aim to evaluate the morphologic features of the bone marrow biopsies of ET cases, defined upon strict adherence to the 2017 WHO criteria and featuring a TN genotype. Methods: We collected a continuous, single centre series of TN-ET patients (n = 47), being diagnosed at the Unit of Pathology and followed up at the Unit of Hematology of our Institution, according to the 2017 WHO Classification criteria between February 1999 and July 2017. The series was matched to a group of mutated ET patients (n = 40). Clinical and laboratory data included blood cells count, LDH level and assessment of splenomegaly. Driver gene mutational status was assessed by routine analysis in the clinical practice. Morphologic features were reviewed by two hematopathologists, on haematoxylin/eosin, Giemsa and Gomori’s silver impregnation stained slides. Blast count was assessed both on morphology and via immunohistochemistry with anti-CD34 antibody (QBEnd-10 clone, Dako-Agilent). Results: The two study cohorts turned out to be rather homogeneous on the side of clinical-laboratory features, with the exception of a higher prevalence of the female sex and hematocrit values more frequently under normal limits in TN-ET. The histological review, on the other hand, highlighted some differences between the two groups: TN-ET showed a tendency to mild hypercellularity, with a dyshomogeneous distribution (p = 0.02), a slight increase in the erythropoiesis (p = 0.0008) and in the proerythroblasts (p < 0.0001), and left-shifted granulopoiesis. Megakaryocytes were morphologically typical of ET in both groups, with presence of giant, hyperlobulated forms, but TN-ET more frequently featured also polymorphic (p = 0.009) or dysmorphic megakaryocytes (p = 0.001). Furthermore, besides no significant differences in fibrosis and blast counts, an increase in vascular density was significantly more common in TN-ET. Summary/Conclusion: Our study highlighted differential histological traits between “mutated” and TN-ET. Notably, as the clinical findings were well matched among “mutated” and TN-ET, the latter carried some histologic features on the boundary with other MPNs, in particular with PMF, whereas clear-cut polycythemia vera-like features were not present. On the opposite, cytological features, particularly those of megakaryocytes, supported in all cases the clinical picture of MPN over the alternative setting of a reactive thrombocytosis. The question therefore arises as to which is the most appropriate histological diagnosis in this group of cases, whether of ET or MPN, unclassifiable, to be followed over time and classified according to the clinical evolution and any progressive bone marrow changes. Finally, it should be stressed that, in the present study, a TN definition was applied in the context of molecular characterization performed on routine basis: however, clonal nature may be demonstrated by extended panels, which may also be able to detect variant mutations affecting driver genes.