Prognostic importance of blast cell DNA content in childhood acute lymphoblastic leukemia

Using flow cytometric techniques, we determined the pretreatment distribution of DNA content in propidium iodide-stained leukemic blasts from 205 children with "standard-risk" acute lymphoblastic leukemia (ALL). Risk assignment was based on an initial WBC count less than 100 X 10(9)/L, no thymic mass, no meningeal leukemia, and lymphoblasts lacking sheep erythrocyte receptors or surface immunoglobulin. A single aneuploid leukemic line was detected in 74 cases (36.1%): 70 hyperdiploid and four hypodiploid. For hyperdiploid cases, the DNA index (DI, or ratio of the DNA content of leukemic v normal G0/G1 cells) ranged from 1.06 to 2.0 (median, 1.20). A secondary leukemic line with hyperdiploid cellular DNA content was identified in 21 cases with diploid primary lines. Children whose primary leukemic line showed a DI greater than or equal to 1.16 (n = 57) had significantly better responses to treatment than did those with either a diploid DI (n = 130; P = .002) or values in the range of 1.01 to 1.15 (n = 14; P = .001). The relative risk of failure for hyperdiploid cases with DI greater than or equal to 1.16, corresponding to greater than or equal to 53 chromosomes, was one-third that of the other two groups. Treatment responses of patients with both diploid and hyperdiploid lines were identical to those associated with single diploid lines, but significantly worse than those associated with single hyperdiploid lines with DI greater than or equal to 1.16 (P = .016). The most favorable prognostic variables selected by a Cox proportional hazards model were: DI greater than or equal to 1.16 (P = .001), white race (P = .022), WBC less than or equal to 25 X 10(9)/L (P = .032), age between 2 and 9 years (P = .075), and hemoglobin less than 7.0 g/dL (P = .094). DNA index greater than or equal to 1.16 retained its significant prognostic impact even after adjustment for other variables (P = .001). With the combination of DI greater than or equal to 1.16 and WBC less than or equal to 25 X 10(9)/L, one can identify a group of children with ALL who have a low probability of relapse when treated with current therapy. If they remain disease-free after longer follow-up, it may be advisable to treat them with less intensive, hence less toxic, chemotherapy than patients with higher WBC counts or lower DI values.