Randomized study of individualized induction therapy with or without VCR, and of maintenance of 4 or 12 courses in adult AML: JALSG-AML87. Japan Adult Leukemia Study Group (JALSG).

We asked 2 questions in this study. First was the additional effect of VCR in induction therapy, and the second was the duration of maintenance therapy. Adult AML were treated by an individualized response-oriented induction therapy with behenoyl Ara-C 200 mg/m2 daily + 6MP 70 mg/m2 daily + prednisolone 40 mg/m2 on days 1-4 + DNA 40 mg/m2 on days 1-3 and additionally on days 7, 8, 11, 12 (for M3, DNR 50 mg/m2 daily) (BHAC-DMP) until bone marrow became severely hypoplastic with less than 5% of blasts. Patients were randomized to BHAC-DMP or BHAC-DMP + VCR 0.35 mg/m2 on days 1-4. After obtaining CR, 3 courses of intensive consolidation therapy were given together with I.T. MTX+Ara-C+PSL. Maintenance intensification therapy was randomized to either 4 or 12 courses given every 2 months. Patients of age greater than or equal to 60 received about 2/3 reduced doses. From June 1987 to Sept. 1989, 265 consecutive adult AML were registered from 19 institutions and 258 were evaluable. Age ranged from 15 to 79 (med., 48). Out of 258, 200 (77.5%) achieved CR (80% in 209 of age less than 60 and 65% in 49 of age greater than or equal to 60). Unexpectedly, addition of VCR reduced the high CR rate of BHAC-DMP significantly (84% to 70%, p = 0.007). At the median follow-up of 37 mo., overall survival is 37%, and event-free survival (EVS) 27%. Survival, continuing CR and disease-free survival (DFS) rates of 200 CR cases are 45%, 40% and 35%, respectively. Patients received 12 courses of maintenance therapy showed better DFS (P = 0.0555). The VCR group had significantly worse EFS. By multivariate analysis, significant prognostic factors for the achievement of CR were age less than 60, PS 0-2 and no addition of VCR. Significant factors for longer DFS were induction of CR by one course, FAB M3 or M5 and age less than 50. The present multi-institutional study confirmed the high CR rates of the response-oriented individualized therapy reported from several centers in Japan, but failed to support an additional effect of VCR reported from one center.