Timing of platelet recovery is associated with adequacy of leukapheresis product yield after cyclophosphamide and G-CSF in patients with lymphoma

A subgroup of patients with refractory Hodgkin's (HD) or non-Hodgkin's (NHL ) lymphoma may be cured with high-dose chemotherapy and peripheral blood pr ogenitor cell rescue. To investigate the relationship of adequate leukapher esis yield and time course of platelet recovery after mobilization chemothe rapy, we retrospectively analyzed the leukapheresis yields in seven patient s with Hodgkin's disease and fifteen patients with non-Hodgkin's lymphoma u ndergoing high-dose chemotherapy. Our goal was to develop a rule to determi ne when to initiate leukapheresis and then to prospectively validate this r ule. All patients were mobilized with cyclophosphamide and G- CSF (granuloc yte-colony stimulating factor). A total of 144 leukaphereses were completed and analyzed. Based on the CD34 content in the initial harvest product, fi fteen patients were defined as poor mobilizers (CD34 < 0.15 x 10(6)/kg) and seven were good mobilizers. The platelet count on the first day of harvest ing was significantly associated with the poor mobilizers (P = .03). Age, s ex, marrow involvement, disease (HD vs. NHL), prior radiation, time since l ast chemotherapy, and total number of cycles of prior chemotherapy were not predictive of poor mobilizers. By using a platelet count cut off of 35 x 1 0(9)/L, we retrospectively analyzed 144 individual leukapheresis products, to test whether CD34 yield was predicted by the peripheral blood platelet c ount on the day of leukapheresis. This rule had an excellent sensitivity, 9 1%, and a specificity of 67%. Subsequently, we validated this rule with the next twenty-four patients undergoing leukapheresis of which there were 143 leukaphereses. The prediction rule exhibited a sensitivity of 72% and a sp ecificity of 68% in the validation set. There does appear to be utility in using the platelet count to guide the initiation of leukapheresis after che motherapy and G-CSF mobilization. (C) 1999 Wiley-Liss, Inc.