Citations

Abstract The Janus kinase 2 mutation, JAK2617V>F, is myeloid neoplasm-specific; its presence excludes secondary polycythemia, thrombocytosis, or bone marrow fibrosis from other causes. Furthermore, JAK2617V>F or a JAK2 exon 12 mutation is present in virtually all patients with polycythemia vera (PV), whereas JAK2617V>F also occurs in approximately half of patients with essential thrombocythemia (ET) or primary myelofibrosis (PMF). Therefore, JAK2 mutation screening holds the promise of a decisive diagnostic test in PV while being complementary to histology for the diagnosis of ET and PMF; the combination of molecular testing and histologic review should also facilitate diagnosis of ET associated with borderline thrombocytosis. Accordingly, revision of the current World Health Organization (WHO) diagnostic criteria for PV, ET, and PMF is warranted; JAK2 mutation analysis should be listed as a major criterion for PV diagnosis, and the platelet count threshold for ET diagnosis can be lowered from 600 to 450 × 109/L. The current document was prepared by an international expert panel of pathologists and clinical investigators in myeloproliferative disorders; it was subsequently presented to members of the Clinical Advisory Committee for the revision of the WHO Classification of Myeloid Neoplasms, who endorsed the document and recommended its adoption by the WHO.

In a phase 2 study, 23 patients with myelofibrosis with myeloid metaplasia were treated with imatinib mesylate at a constant dose of 400 mg/d. Treatment was held in 16 patients (70%), after 1 to 12 weeks, because of side effects (neutropenia, 6 patients; musculoskeletal pain, 5 patients; thrombocytosis, 4 patients; edema, 3 patients; diarrhea and hyperbilirubinemia, 1 patient). Including patients in whom retreatment at a reduced dose was possible, 11 patients (48%) were able to continue treatment beyond 3 months. None of the patients experienced a response in anemia, and only 2 had partial responses in splenomegaly. A greater than 50% increase in platelet count was documented in 11 (48%) patients, but not in those with baseline platelet counts of less than 100 × 109/L. In vitro, imatinib mesylate caused variable degrees of growth suppression of myeloid and erythroid progenitors that unfortunately did not translate into clinical benefit.

The lack of diagnostic certainty in some patients makes it difficult to distinguish between primary and secondary forms of thrombocytosis. To augment current diagnostic studies for thrombocytosis, we retrospectively evaluated clinical records and bone marrow trephine specimens of 183 patients with thrombocytosis—164 with essential thrombocythemia (ET), 19 with reactive thrombocytosis (RT)—for bone marrow angiogenesis, bone marrow megakaryocyte c-Mpl staining, and morphologic evidence of megakaryocyte proliferation. Angiogenesis was increased in patients with ET compared with healthy controls (P < .0001) and patients with RT (P = .006). In addition, an increase in angiogenesis was associated with certain disease features such as splenomegaly (P = .004) and reticulin fibrosis (P = .005). Decreased megakaryocyte c-Mpl staining was observed in a heterogeneous pattern in ET compared with healthy controls (P < .0001) and RT (P < .0001). Histologic stratifying criteria incorporating increased angiogenesis, decreased megakaryocyte c-Mpl expression, and marked megakaryocyte proliferation in the bone marrow was highly sensitive (97%) and specific (95%) for distinguishing ET from RT (P < .0001). However, with the current duration of follow-up available on the patients, none of the histologic features evaluated have yet demonstrated prognostic value for subsequent clinical course, vascular events, or survival.

Unlike other leukemia types in which the bone marrow findings are diagnostic, the bone marrow pathology of T-cell granular lymphocytic leukemia (GLL) is subtle and ill-defined. In this study, bone marrow biopsy specimens from 36 patients with T-cell GLL and from 25 control patients with cytopenias and relative or absolute increases in blood large granular lymphocytes were studied by immunohistochemistry using antibodies to the cytolytic lymphocyte antigens CD8, CD56, CD57, TIA-1, and granzyme B. The goals were to clarify the bone marrow pathology of T-cell GLL and to refine the diagnostic criteria for T-cell GLL. Most bone marrow specimens from the T-cell GLL patients contained interstitially distributed clusters of at least 8 CD8+(83%) or TIA-1+ (75%) lymphocytes or clusters of at least 6 granzyme B+ (50%) lymphocytes. Interstitial clusters of CD8+, TIA-1+, or granzyme B+ cells were present in 36%, 12%, and 0%, respectively, of the control bone marrows (all values significantly different, P < .001). An additional T-cell GLL disease-specific finding was the presence of linear arrays of intravascular CD8+, TIA-1+, or granzyme B+ lymphocytes, found in 67% of cases of T-cell GLL and in none of the 25 control samples (P < .001). Staining for CD56 and CD57 was noncontributory. These findings clarify the bone marrow histopathology of T-cell GLL and provide an additional tool by which the discrete, abnormal lymphocyte population required for a diagnosis of T-cell GLL can be identified.