Stratification by Prognostic Factors in the Design and Analysis of Clinical Trials for Acute Lymphoblastic Leukemia

Abstract

Children with acute lymphoblastic leukemia (ALL) were treated until a decade ago as a homogeneous group of patients with the same diagnosis. Experience has illustrated the extraordinary clinical and biologic heterogeneity of patients with this diagnosis. Now numerous characteristics of the disease which can be evaluated at the time of diagnosis are known to be associated with either favorable or poor outcome to current treatments and, therefore, are important indicators of prognosis. With the identification of an increasing number of such prognostic factors, it has become possible to identify subsets of patients with ALL which can be expected to have a very favorable outcome to current treatments and other subsets that are at high risk of early relapse and death. It has become essential for the design and analysis of clinical trials to use important prognostic characteristics to stratify patients with ALL into relatively homogeneous subgroups and to design clinical trials which pose therapeutic questions appropriate for each stratum. Apart from clinical trials, the knowledge and judicious use of the most important prognostic factors are necessary for the selection of the most appropriate treatment for each patient group.

The work was supported in part by grants from the Division of Cancer Treatment, National Cancer Institute, NIH, DHHS, and the T. J. Martell Foundation.