In Vitro Drug Sensitivity as a Predicitive Tool of Early Clinical Response in Childhood Acute Lymphoblastic Leukemia.

Background: Residual disease or rapidity of response to induction therapy is among the most powerful predictors of outcome in pediatric Acute Lymphoblastic Leukemia (ALL). Various methods to determine response during induction have been in use in clinical investigation. We hypothesize that drug sensitivity at the cellular level predicts rapidity of response to induction therapy in ALL. We recently developed a high resolution flow cytometry based cytotoxicity assay for in vitro cellular drug response profiling for pediatric ALL. This method has a turnaround time of 48 hours and the ability to measure drug effect specific to leukemic cells regardless of number of admixed normal cells. We report preliminary data that correlate results of this drug sensitivity assay with rapidity of response to induction therapy among patients with ALL. Methods: We performed in vitro tests, applying a multiparameter flow cytometric drug cytotoxicity assay, on bone marrow (BM) samples of 23 patients with newly diagnosed standard (n = 10), high (n = 11), and very high (n = 2) risk ALL, as defined by the Children’s Oncology Group (COG) and NCI risk classification. Fourteen patients were rapid early responders (RER) and 9 were slow early responders (SER) by COG criteria at day 15 and 29. Cryopreserved cells from BM samples were thawed and determined to have adequate viability by trypan blue dye exclusion. Drugs were tested at three different concentrations, each in triplicate. Concentrations tested were based on an empirically derived cut-off concentration (EDCC) adopted from published in vitro studies, chosen to produce a large scatter of survival index values among samples. Leukemic blasts were specifically identified by surface markers, CD 45, CD 19 and CD 10 or CD 3, while cytotoxicity was measured with Annexin V based apoptosis. Leukemic cell survival index (LCSI = Average Replicate /Average Control x 100) was determined at 48 hours after in vitro exposure to individual standard induction agents for pediatric ALL: vincristine, asparaginase, dexamethasone, prednisone and daunomycin. LCSI differences between RER and SER were compared using Wilcoxon rank sum test for each drug and concentration. The mixed effects model was used to evaluate the overall difference of LCSI between RER and SER over the three concentrations (referred to as “averaged concentrations”). Results: For dexamethasone, a significantly lower LCSI was seen in the RER compared with the SER cohort at individual and averaged concentrations: RER mean LCSI = 40.2%, SER mean LCSI = 70.1% (p = 0.01, mixed effects model). A trend toward a lower mean LCSI in the RER compared with the SER group was noted for asparaginase and vincristine at individual and averaged concentrations (p < 0.1). Mean LCSI was not different between the RER and SER groups for daunomycin and prednisone at individual or averaged concentrations. Conclusions: This in vitro drug sensitivity assay provides a response profile for dexamethasone that correlates with in vivo response to induction therapy. Research is ongoing with more patient samples in order to achieve a greater statistical power to detect a smaller difference for all drugs tested. Further research will also correlate clinical response with LCSI using drug combinations in vitro. Results of these studies will determine the potential value of this assay for early risk stratification and modification of therapy in de novo or relapsed ALL.