Prognostic factors and therapy in acute lymphoblastic leukemia of childhood: CCG‐141: A report from childrens cancer study group

From 1975 to 1978, 880 previously untreated patients with acute lymphoblastic leukemia (ALL) were entered on CCG-141, a protocol designed to determine the importance of clinical predictors of remission-induction, duration of complete continuous remission (CCR), and survival, and to determine the benefit of intensive therapy in patients with a poor prognosis. Patients with initial leukocyte count 20 × 109/1 were randomly assigned to this standard regimen or to a more intensive four-drug regimen (PDN, VCR, LASP, and cyclophosphamide) and a maintenance program consisting of alternating cycles of PDN, VCR, 6-MP, and MTX (POMP) and PDN, VCR, cytosine arabinoside, and Adriamycin (POCA). The overall rate of complete remission (CR) was 93%. Factors associated with a significantly lower rate of CR were: M3 bone marrow (BM) on day 14, CNS leukemia at diagnosis, L3, morphology, less than 40% PAS-positive lymphoblasts, low IgG, age >10 years, and L2 morphology. The relapse rate in patients with an initial leukocyte count 20 × 109/1. By multivariate analysis, adverse predictors of duration of CCR were leukocyte count >50 × 109/1, Hb >10 g/dl, low IgM, massive splenomegaly, age 20 × 109/1. By life table analysis, 80% of patients with good prognostic factors remain in CCR at four years. In patients with poor prognostic factors, the CCR rate at four years is 43%. This study demonstrates the utility of clinical prognostic factors in identifying subsets of patients at low and high probability of treatment failure. Intensive induction and maintenance therapy as used in this protocol did not benefit the poor prognosis group.