Hyperleukocytosis in adult acute nonlymphocytic leukemia: impact on remission rate and duration, and survival.

The clinical courses of 353 patients with acute nonlymphocytic leukemia (ANLL) treated between 1971 and 1982 at the Baltimore Cancer Research Program (BCRP) of the National Cancer Institute were reviewed and examined for the impact of presenting WBC count on the initial course and overall outcome of these patients. Group A (WBC greater than 100,000/microL) had significantly more deaths during the first week of therapy than did group C (WBC less than 50,000/microL) (P = .0003). CNS hemorrhage was responsible for a significantly greater number of deaths in group A compared with group C (P less than .004). The group B (WBC 50,000 to 100,000/microL) death rate was intermediate. These findings are consistent with other reports of complications of leukostasis. Rapid intervention with antileukemic therapy and cranial irradiation may have decreased the risk of CNS hemorrhage in group A. If early deaths are removed from analysis, the complete remission rate among patient groups is not significantly different (group A, 59%; group B, 68%; group C, 65%). However, further analyses of patients achieving remission demonstrate significant differences among patient groups based on presenting WBC count. The median complete remission duration of patients in group A (4.2 months) is shorter than that of patients in group B (8.0 months) or C (8.0 months), P = .07. In addition, remission duration has improved with modern aggressive antileukemic therapy in groups B (median before 1977, 7.0 months; after 1977, 22.0 + months) and C (before 1977, 6.0 months; after 1977, 16.0 + months). No such improvement has occurred in group A, in which the median duration of remission was 4.2 months before and after 1977. The same findings are demonstrated in an analysis of survival, with improvement occurring only in groups B (median before 1977, 16.5 months; after 1977, 26.0 + months) and C (before 1977, 13.5 months; after 1977, 24.0 + months). Long-term follow-up (minimum of 4 years) of these patients has allowed an analysis of the effect of presenting WBC count on the overall outcome of adult patients with ANLL.