Detection of Residual Acute Lymphoblastic Leukemia Cells in Cultures of Bone Marrow Obtained during Remission

We used a semisolid culture assay to quantitate leukemia cells in the bone marrow of patients with childhood acute lymphoblastic leukemia (ALL). In bone marrow cultures from 40 patients with newly diagnosed disease, the colonies that developed in vitro consisted of lymphoblasts with the same surface markers and abnormal karyotype as the original diagnostic marrow specimens. We also studied marrow cultures from 13 patients in chemotherapy-induced remission; 6 of these, including 1 obtained from a patient during successful engraftment after marrow transplantation, also yielded lymphoblast colonies in culture, with the same immunologic phenotype or abnormal karyotype as the original leukemic marrow. Four of these patients, including the one who underwent marrow transplantation, relapsed within 2 to 30 months of the abnormal cultures; the other two are still In remission, one of them 30 months after diagnosis. Bone marrow cultures from eight normal controls and from the other seven patients in remission did not yield lymphoblast colonies; all seven of the latter are still in remission. This assay appears to allow detection of small numbers of residual leukemic cells. We conclude that the technique will be valuable in monitoring the efficacy of chemotherapy and allogeneic bone marrow transplantation in acute lymphoblastic leukemia, as well as in evaluating the quality of purged marrow for autologous marrow transplantation. (N Engl J Med 1986; 315: 538–42.), Approximately 50 percent of patients with childhood acute lymphoblastic leukemia (ALL) are cured with chemotherapy.¹ The others have recurrent disease that is ultimately fatal, unless bone marrow transplantation is successful.^{2 3 4} Currently, most centers use a number of hematologic, clinical, and cytogenetic prognostic factors determined at diagnosis to predict which patients will have early relapse. These prognostic factors are useful but not entirely satisfactory, because of the number of cases in which predicted and clinical outcome differ.¹ To understand why ALL recurs in some patients but not others, we developed a colony-culture assay for the quantitation of blast cells in the….