All-trans retinoic acid significantly increases 5-year survival in patients with acute promyelocytic leukemia: long-term follow-up of the New York study.

All-trans retinoic acid (ATRA) induces a high incidence of complete remission (CR) in patients with acute promyelocytic leukemia (APL); however, the magnitude of this agent's contribution to increased rates of cure of this disease has not yet been established. From 1990 to 1995 we used RA as remission induction therapy in 103 APL patients (73 newly diagnosed and 30 previously treated) who were retinoid-naive and were treated on the basis of initial morphology. Patients whose diagnosis was changed on the basis of the results of molecular testing (n = 13) were withdrawn from RA treatment and given chemotherapy alone. After achieving a CR, previously untreated patients received several cycles of consolidation chemotherapy, usually with idarubicin and cytosine arabinoside. Among individuals whose diagnosis was molecularly confirmed, 54 of 65 new patients (83%) and 25 of 30 previously treated patients (83%) achieved a CR. All induction failures in molecularly diagnosed cases were due either to early death or to premature withdrawal. Median disease-free and overall survival rates recorded for all newly diagnosed patients are currently > 40+ and > 43+ months, respectively. We subsequently examined a subset of 27 newly diagnosed patients treated during the first 2 years of this program whose actual median follow-up period is now > 5 years. Median disease-free and overall survival rates recorded for this group are > 57+ and > 58+ months, respectively; 56% of these patients are alive in first remission. These results significantly exceed those achieved using chemotherapy alone in a historical control group of 80 patients consecutively treated at this center from 1975 to 1990, whose median disease-free and overall survival rates were 11 and 19 months, respectively; only 22% of these patients were alive in first remission at 5 years. Although a high proportion of previously treated patients also achieved a CR after RA treatment, median disease-free and overall survival rates noted for that group were markedly lower (i.e., 7.5 and 10.9 months, respectively). Thus, data from patients whose median follow-up period is now > 5 years have confirmed earlier projections and indicate that the use of RA for remission induction yields an approximately 2.5-fold increase in the proportion of patients who have presumably been cured of this disease.