Acute lymphoblastic leukemia in a developing country: preliminary results of a nonrandomized clinical trial in El Salvador.

PURPOSE: To improve outcome and study biology of childhood acute lymphoblastic leukemia (ALL) in El Salvador. PATIENTS AND METHODS: Between January 1994 and December 1996, 153 children of El Salvador had newly diagnosed ALL treated in a collaborative program between Hospital Benjamin Bloom and St. Jude Children's Research Hospital (SJCRH). Therapy was based on a modified SJCRH protocol, with uniform remission induction (prednisone, vincristine, L-asparaginase) followed-up by consolidation with teniposide/cytarabine and/or high-dose methotrexate. Continuation treatment was risk-stratified: 123 patients assigned to the high-risk group received weekly rotational drug pairs, and 16 assigned to the standard-risk group received daily 6-mercaptopurine, weekly methotrexate, and monthly pulses of vincristine plus dexamethasone. High risk was defined as: DNA index < 1.16, age 12 months or younger, white blood cell count > or = 50 x 10(9)/L, T-cell immunophenotype, anterior mediastinal mass, central nervous system leukemia at diagnosis, or t(4;11), t(1;19), or t(9;22). Duration of the continuation treatment was 2.5 years in both groups. The median age at diagnosis of all patients was 4.8 (range I d-17 yrs), median leukocyte count was 15 (range 1-766) x 10(9)/L, and sex distribution was equal. RESULTS: Immunophenotypes were early beta-progenitor in 79%, T-cell in 3.9%, and inconclusive in 17% of cases. DNA index was <1.16 in 80.5% and was > or = 1.16 in 19.5% of the 123 known cases. For the analyzes, patients who refused therapy (abandoned treatment) were considered to have treatment failure as of their last follow-up dates. Complete remission was achieved in 126 of 151 (82.4%) patients (11 abandoned therapy during induction). The overall 4-year event-free survival (EFS) rate +/- 1 standard error was 48 +/- 6%. The 4-year EFS rates in patients at high-risk and standard-risk were 46 +/- 7% (n = 121) and 69 +/- 15% (n = 16), respectively (P = 0.20). When patients who refused further treatment are censored, the corresponding 4-year estimates of EFS are 51 +/- 8% and 75 +/- 14%, respectively. CONCLUSIONS: These results suggest that the biology of childhood ALL in El Salvador appears to be similar to that seen in the United States. Risk-directed chemotherapy can successfully be used in developing countries, but risk factors must be carefully determined and applied.