[Corticosteroid-dependent reduction of leukocyte count in blood as a prognostic factor in acute lymphoblastic leukemia in childhood (therapy study ALL-BFM 83)].

In therapy study ALL-BFM 83 a total of 630 patients with acute lymphoblastic leukemia (ALL) have prospectively been evaluated for initial response on therapy with corticosteroids. It was the aim to qualify the in vivo cytoreduction as a new predictor for therapy failure. All patients were exposed for 7 days to prednisone before combination chemotherapy at day 8 has been started. At day 0 one additional dose of Methotrexate was given intrathecally. Therapy for all patients with non-B-ALL has been stratified according to initial tumor burden (risk factor) providing four therapy branches: standard risk low (SR-L), standard risk high (SR-H), medium risk (MR), high risk (HR). After a median duration of study of 21 months, event-free survival (EFS) is for all 630 patients 73%, 81% for SR-L, 76% for SR-H, 69% for MR, and 35% for HR patients (date of evaluation Jan. 1, 1987). In this prospective study, a small subgroup of patients (n = 48; 7.6% of total group) is characterized by greater than 1000 leukemic blasts/mm3 peripheral blood at day 8 after exposure to prednisone. In this subgroup the EFS is only 43% in contrast to 76% in the complementary group of 582 patients with less than 1000 leukemic blasts/mm3 peripheral blood at day 8. Patients of that risk group are derived from therapy branches SR-H, M and HR, the latter contributing relatively most patients. In this negatively selected group all patients with an initial high white blood count, CNS disease at diagnosis, immune subtypes as prae-T-ALL (n = 6), T-ALL (n = 18), null-ALL (n = 5), and males clearly dominate. Of 48 patients with greater than 1000 blasts/mm3 at day 8 4 subsequently failed to enter remission and 8 were qualified as lateresponders. 18 patients relapsed, most of them earlier compared to those of the complementary group. The initial in vivo response on corticosteroid therapy is considered a supplementary prognostic predictor for early failure. It will be utilized in trial ALL/NHL-BFM 86 to qualify patients at the highest risk for relapse. This group of patients is supplemented in addition by non- and lateresponders and children with acute undifferentiated leukemia (AUL). The in vivo corticosteroid test is simple, generates early and reliable results and can be obtained almost always. Thus it may be recommended for use in a multicenter trial.