Low relapse rate in children with acute lymphoblastic leukemia after risk-directed therapy.

PURPOSE: Even though acute lymphoblastic leukemia (ALL) responds well to chemotherapy, relapse remains the major problem. This study documents relapse and survival rates in 85 consecutive children (33 at good risk, 52 at high risk) with ALL diagnosed in 1991 to 1996. PATIENTS AND METHODS: Until 1993, the New York II protocol for the high-risk group and a combination of UKALL XI (induction) and R blocks of ALL-REZ BFM-87 (intensification) regimens for patients at good risk were used. To reduce toxicity, the protocols were subsequently modified. Consolidation treatment was the same for both groups, consisting of a lower cytarabine dose and methotrexate removal, whereas intensification was changed only for the high-risk group using the BB block of the NHL-BFM-90 protocol. The bone marrow clearance of leukemia was assessed on day 22, and minimal residual disease was detected using polymerase chain reaction analysis of Ig heavy-chain gene rearrangements. RESULTS: Seventy patients had common precursor B lineage ALL, six had pre-B-ALL, eight had T-ALL, and one had B-ALL. Two patients never achieved remission and died. Six patients died of consolidation-related complications. Four more patients died, two during induction and two during maintenance therapy. Two other children had relapse (2.3%), both of whom were treated with the earlier protocols and then underwent bone marrow transplantation. Four more children with morphologically complete remission showed minimal residual disease (which reached the levels of 1 leukemic cell among 10(2)-10(4) normal cells) with the use of clone-specific probes at several points of the study intervals, but never had relapse. The 5-year overall and event-free survival rates were 86% and 83%, respectively. The 5-year overall survival rates for good-risk and high-risk groups were 94% and 81%; the corresponding event-free rates were 91% and 78%. The 5-year event-free survival rate in the patients at high risk was significantly higher after the protocol change (90% vs. 65%, P = 0.04). CONCLUSIONS: The modification proved to be effective in diminishing the therapeutic toxicity and improving the efficacy, mainly for the high-risk group.