Plasma FLT3-L levels predict bone marrow recovery from myelosuppressive therapy

BACKGROUND. During the recovery period after anticancer myelosuppressive th erapy, hematopoietic progenitor cells become mitotically active in order to replenish the bone marrow compartment and remain hyperproliferative even a fter normalization of peripheral white blood cells and platelets. At this s tage, the progenitors are more radiosensitive and chemosensitive. Dosing pa tients with additional totoxic therapy during this phase will likely result in more severe toxicity. For example, the authors have noted that the red bone marrow (RM) dose resulting from radioantibody therapy does not correla te with observed bone marrow toxicity. Several patients given similar RM do ses had Grade 3 or 4 toxicity, whereas others had Grade 0-2 toxicity even t hough their white blood cell (WBC) and (PLT) counts were normal at the time of dosing. The goal of these studies was to establish a noninvasive predic tive marker of bone marrow activity that could determine stem cell and prog enitor cell recovery from previous myelosuppressive therapy. METHODS. A retrospective study was conducted to quantitate plasma levels of 5 cytokines regulating hematopoiesis, namely, 2 stimulatory fms-like tyros ine kinase (FI,TS-L) and stem cell factor (SCF) and 3 inhibitory growth fac tors tumor necrosis factor-alpha (TNF alpha), tumor growth factor-beta, and macrophage inflammatory protein (MIP-1 alpha), by immunoassay in 43 patien ts enrolled in clinical trials at Garden State Cancer Center in Belleville, New Jersey. All patients had had previous chemotherapy with a duration of 1-24 months. The serum cytokine values were with the magnitude of leukopeni a or thrombocytopenia following a single dose radioantibody as the cytotoxi c therapy. RESULTS. Plasma FLT3-L levels predicted excess platelet toxicity in 13 of 1 6 patients (mean = 225 +/- 106 pg/mL) and resulted in a false-positive in o nly 3 of 27 other patients (mean = 80 +/- 41 pg/mL). Plasma FLT3-L > 135 pg /mL resulted in 81% sensitivity and 89% and 86% specificity and accuracy, r espectively, for pre dieting excess toxicity caused by additional cytotoxic therapy. The positive likelihood ratio was 7.5 (95% confidence interval, 2 .5-22.5) and the negative likelihood ratio was 0.19 (95% confidence interva l, 0.05-0.67). CONCLUSIONS. Elevated plasma FLT3-L in patients who previously received che motherapy is a predictive measure of the stage of recovery of the bone comp artment. FLT3-L seems to identify the likelihood that the patient will expe rience Grade greater than or equal to 3 thrombocytopenia if additional cyto toxic therapy is administered. Knowledge of bone marrow activity should per mit therapy that is aggressive by establishing the earliest possible time f or dosing with any agent for which myelosuppression is the dose-limiting to xicity. (C) 2000 American Cancer Society.