Prognostic Diversity of Cytogenetic Lesions in Myelofibrosis with Myeloid Metaplasia: A Prospective Study of 88 Patients.

Abstract Background : Thirty to 50% of patients with myelofibrosis with myeloid metaplasia display detectable bone marrow cytogenetic abnormalities at diagnosis with some(Demory et al. Blood1988;72:855, Dupriez et al. Blood1996;88:1013, Reilly et al. BJH1997;98:96) but not other studies (Cervantes et al. BJH1998;102:684, Okamura et al. Int J Hematol2001;73:194) suggesting a detrimental effect on survival. Methods : The current prospective study was conducted as a follow-up of the observations from an antecedent retrospective study that identified sole abnormalities of either 13q- or 20q- as ‘favorable’ to survival (Tefferi et al. BJH2001;113:763). All clinical and laboratory measurements, including bone marrow examination and karyotype analysis, were performed concurrently in consecutive cases of histologically confirmed MMM regardless of either duration of disease or presence of active therapy. Survival, leukemic transformation rate, and clinical progression were determined from the time of cytogenetic analysis. Results : Karyotypic analysis was successfully performed in a total of 88 patients (median age 61 years; 31 females) and the results were normal in 49 (56%) and abnormal in the remaining 39 patients with 9 (10%) displaying ‘favorable’, and 30 (34%) ‘other’ clonal abnormalities. At the time of study entry, 57 patients were not recieving any form of therapy, 18 were newly diagnosed, and the Dupriez prognostic score distributions were ‘0’ in 33 patients, ‘1’ in 35 and ‘2’ in 20 (Dupriez et al. Blood1996;88:1013). All patients were followed for a minimum of 3 years and a median of 42 months. The 3 cytogenetic categories were similar in age and gender distribution. Survival in the favorable cytogenetic group was similar to that of the normal group and significantly better than the group with other cytogenetic abnormalities (p=0.02). Figure Figure Similarly, no patient (0%) in the ‘favorable’ group transformed into acute leukemia while 6 of 30 (20%) in the ‘other’ and 5 of 49 (10%) in the normal groups did (p=0.21). In a Cox regression analysis, cytogenetic lesions other than the ‘favorable’ kind, hemoglobin below 10 g/dL, leukocytosis, hypercatabolic symptoms, increased blood blast %, blood CD 34 count above 100 x 106/L were individually associated with shortened survival. In a multivariate analysis, only hemoglobin level and leukocyte count retained their significance. Conclusion : Single chromosomal abnormalities of either 13q- or 20q- should be considered as ‘favorable’ cytogenetic lesions in MMM while other abnormalities are associated with an adverse prognosis. The current observation explains the discrepancy in the literature regarding the prognostic relevance of cytogenetic findings in MMM and warrants additional studies to determine the value of incorporating the specific information in prognostic scoring systems.