Primary myelodysplastic syndromes: analysis of prognostic factors in 235 patients and proposals for an improved scoring system.

In an attempt to identify prognostic factors for survival and leukemic transformation, 235 untreated patients with primary myelodysplastic syndromes (MDS) were analyzed in a single center retrospective study. To the well known FAB classification of MDS a supplementary group of patients with pure sideroblastic anemia (PSA) was added, characterized by the absence of dysplastic features of non-erythroid cells. Accordingly, the morphological subtypes were refractory anemia (RA), n = 55; PSA, n = 40; RA with ring sideroblasts (RARS), n = 33; RA with excess of blasts (RAEB), n = 53; RAEB in transformation (RAEB/T) n = 29; and chronic myelomonocytic leukemia (CMML), n = 25. Having screened 28 clinical, cytological, and laboratory parameters by univariate analysis, multiple regression analysis identified six variables with independent prognostic value: percentage of bone marrow blasts, serum LDH activity, PSA, hemoglobin concentration, age, and platelet count. If patients with PSA were excluded, the FAB classification no longer contributed independent prognostic information. Based on the results of this multivariate analysis, a simple scoring system was devised for predicting the survival of patients with MDS. A score of unity was allocated to each of the following parameters: bone marrow blasts greater than or equal to 5%, LDH greater than 200 U/I, hemoglobin less than or equal to 9 g/dl, and platelets less than or equal to 100 x 10(9)/I. As a function of their total score, patients were divided into three risk groups (group A, score 0; group B, score 1-2; group C, score 3-4), which differed significantly in both survival and rates of leukemic transformation. The cumulative survival 2 years after diagnosis was 91% in group A, 52% in group B, and 9% in group C (p less than 0.00005). The actuarial risk of transformation to acute myeloid leukemia at 2 years was 0, 19, and 54%, respectively (p less than 0.05). The inclusion of LDH enzyme levels qualified this scoring system for an accurate assessment of patients with CMML whose prognosis is viewed too favorably when rated by other scores. Furthermore, this score was able to identify those patients with RA and RARS who, without showing an excess of marrow blasts, have an unfavorable prognosis.