A cause-specific hazard rate analysis of prognostic factors among 199 adults with acute lymphoblastic leukemia: the Memorial Hospital experience since 1969.

Results of a multivariable analysis of prognostic factors are reported for 199 previously untreated adults with acute lymphoblastic leukemia (ALL). These patients have long-term follow-up, and the probability of cure is estimated at approximately 35%. The cause-specific hazard rate analysis found lower rates of achieving complete remission (CR) in patients with WBC greater than 10,000/microL, AUL (undifferentiated) morphology, and older age. Since these patients required additional time to respond, fewer of them actually achieved CR. Characteristics directly associated with a higher rate of death during induction therapy due to severe bone marrow suppression were low serum albumin concentration (less than or equal to 3.5 g/dL), age greater than 50 years, acute undifferentiated leukemia (AUL) morphology, low Karnofsky performance status, and weight loss greater than 5%. Factors associated with a higher rate of relapse were WBC greater than 20,000/microL, non-T cell ALL, age greater than 60 years, Ph' + ALL, and time to achieve CR greater than 5 weeks. These criteria were used to identify patients at high risk of relapse. In addition, the predictive value of high WBC was found to disappear by 18 months of continuous CR. Finally, the rate of death following first relapse was higher in patients with a short first remission duration, high percentage weight loss at initial diagnosis, and older age. In summary, factors associated with a higher rate of death during attempted induction (ie, low albumin, high percent weight loss, and poor performance status) had no association with the patient's ability to remain relapse-free. Conversely, factors correlating with more extensive or resistant disease (ie, high WBC, null or B cell ALL, or Ph' + ALL) showed no association with the ability to tolerate therapy. Thus, a less toxic but more effective induction regimen is needed for patients with a poor clinical status, whereas a more intensive form of therapy appears warranted for patients presenting with more extensive or resistant disease.