MULTIVARIATE-ANALYSIS OF RISK-FACTORS FOR SURVIVAL AND RELAPSE IN CHRONIC GRANULOCYTIC-LEUKEMIA FOLLOWING ALLOGENEIC MARROW TRANSPLANTATION- IMPACT OF DISEASE-RELATED VARIABLES (SOKAL SCORE)

Patients with chronic granulocytic leukemia (CGL) can be classified in different groups according to risk factors at diagnosis (Sokal). The aim of the present study was to assess the impact of Sokal's risk fact ors in 100 patients with CGL undergoing allogeneic BMT. Patients were in first chronic phase (CP) (n = 65), or with advanced disease (n = 35 ), grafted from an HLA-identical sibling following conditioning with c yclophosphamide and total body irradiation (TBI). Median follow up for survivors is 1783 days (429-3533 days). Variables recorded at diagnos is to calculate Sokal's prognostic index were: leukocyte, platelet and peripheral blood blast cell counts, age and spleen volume. The median value of the Sokal index was 0.87. Projected survival for all patient s at 9 years was 28% (95% confidence limits (CL) 6-49), 48% (34-62) fo r first CP patients and 15% (0-36) for more than first CP patients (p = 0.04). Survival was 25% and 31% for all patients with a Sokal index of < 0.87/greater-than-or-equal-to (p = 0.07) and 55% vs 39% for first CP patients only (p = 0.03). The relapse rate was similar for patient s with Sokal index < 0.87/greater-than-or-equal-to (41% vs 39%, p = 0. 9) and this was also true for first CP patients (33% vs 26%, p = 0.8). In multivariate analysis, an interval between diagnosis and BMT of > 2 years was the most significant negative predictor for survival for t he whole group of patients (p = 0.01) and more so for first CP patient s (p = 0.0004). A high Sokal score had a borderline impact on survival for all patients (p = 0.04) and a significant impact on first CP pati ents (p = 0.01). The risk of relapse was predicted in multivariate ana lysis by the use of T cell depletion (p = 0.002) and by disease phase (p = 0.0008) but not by the Sokal score (p = 0.2). These results sugge st that negative prognostic factors at diagnosis of CGL, as identified by Sokal score, do have an impact on transplant-related mortality and a modest effect on leukemia relapse. They also highlight the striking impact on survival of the time interval between diagnosis and transpl ant.