Long term follow-up of patients with acute myelogenous leukemia who received the daunorubicin, vincristine, and cytosine arabinoside regimen

Abstract

BACKGROUND. In 1985, the authors published a study of acute myelogenous leukemia (AML) patients treated with a chemotherapeutic regimen that was then considered intensive. Ten years later, the authors reanalyzed the same cohort to determine whether the very promising actuarial results observed at 5 years held after longer follow-up. METHODS. Between 1977 and 1982, 61 patients with AML were treated with a protocol consisting of daunorubicin, vincristine, and cytosine arabinoside induction followed by consolidation and maintenance for a total of 2 years. The complete remission (CR) rate was 66%, 84% in males versus 47% in females (P < 0.005). At the time of the first analysis in 1984, the overall survival (OS) was 17%, the projected 5- year continuous CR rate (CCR) 32%, and the disease free survival (DFS) rate 29%, with the best results observed for males and for patients ages 40-60 years (P < 0.05). RESULTS. When the data were reanalyzed 11 years later in 1995, the results were 14% OS, 23% CCR, and 16% DFS at 5 years. However, these figures dropped to 8%, 18%, and 11% at 10 years and to 8%, 12% and 7% at 15 years, respectively. Among the 40 CR patients, 31 relapsed (up to 13 years after CR), and all died within 1.6 years after relapse. Nine patients were in CCR: 4 died of unrelated causes (suicide, alcoholic cirrhosis, acute peritonitis, or bladder carcinoma), 1 was lost to follow-up after 11 years, 2 were alive and well at 17 years at last follow-up, and 2 were transplanted in first CR and were doing well at 13 and 14 years at last follow-up. The survival advantage for males over females persisted (P = 0.0197), but the advantage for patients age 40-60 years did not hold. CONCLUSIONS. These long term data indicate that actuarial analysis at 5 years may overestimate the cure rate of AML patients because a number of late relapses do occur. However, the picture is blurred by the incidence of death not related to leukemia or its treatment; and when these patients were censored at the time of death, 17% of CR patients were still projected to be alive and free of leukemia after 17 years.