Data on all patients with acute lymphoblastic leukemia (ALL) aged 60 or older, referred to our institution over a 18-year period, were studied to determine the incidence and range of clinical and biological subtypes, and the outcome of different therapeutic approaches. Sixty-nine ALL cases (median age: 68 years) were diagnosed between 1980 and 1998 (18% of all adult ALL seen during this period). Ten of them (14%) had a past history of previous malignancy. Karyotypic analysis was performed successfully in 42 cases. Ten patients were diagnosed as Philadelphia chromosome positive (Ph(+)) ALL. Immunophenotyping was performed in 63 cases. Fifty-six patients had B-cell lineage ALL. T lymphoid markers were observed only in 5 cases. Co-expression of myeloid markers was observed in 19% of tested cases. Five patients died before any chemotherapy could be given. All other patients received "curative" treatment according to different protocols used during the period of study. Overall complete remission (CR) rate of these patients was 62% (95% confidence interval (CI): 50-74%). Thirty-nine patients achieved CR after one course of chemotherapy and 4 patients after salvage therapy. Median disease-free survival (DFS) of the entire cohort was 8.3 months (95% CI: 5-12.8 months) and median overall survival was 7 months (95% CI: 6-10 months). In multivariate analysis, the presence of hemorrhage (P = 0.02) was a poor prognostic for CR achievement. Higher WHO performance status (P = 0.003) and the presence of hemorrhage (P = 0.01) at diagnosis were poor prognostics for overall survival. When patients were stratified into three groups according to the time of admission, survival appeared significantly longer for patients admitted between July 1992 and December 1998 (median overall survival at 10 months) than for patients admitted before July 1992 (P = 0.04). "Age-adapted" therapy appeared superior to "young adult-like" therapy in terms of CR rate (96% versus 60%; P = 0.007). However, "age-adapted" therapy did not show any advantage in terms of DFS or overall survival, making the difference in CR rates questionable. We conclude that the pejorative overall outcome in elderly ALL points to the need for new therapeutic trials taking into account the specific characteristics of ALL in this age group.
Copyright 2001 Wiley-Liss, Inc.