Impact of autologous bone marrow infusion on hematopoietic recovery after high-dose cyclophosphamide, etoposide, and cisplatin

Because of potential tumor contamination and inadequacy of current purging technique of bone marrow in patients with solid tumors, we investigated an alternative approach to high-dose therapy without autologous bone marrow (ABM) infusion. Three levels of nonmyeloablative doses of cyclophosphamide 4.5 to 5.25 g/m2, etoposide 750 to 1,200 mg/m2, and cisplatin 120 to 165 mg/m2 (CVP) were administered to patients with metastatic solid tumors. Patients were randomized to ABM (n = 46) or no-ABM (NABM) (n = 46) infusion after CVP to study the impact of ABM on hematopoietic recovery, morbidity, and mortality. All patients had ABM harvested, underwent conventional chemotherapy, and then received CVP. Seventy-three patients received two courses of similar doses. The following were the median days to absolute neutrophil count (ANC) of 0.1 x 10(9)/L: for the ABM arm, 19, 21, and 19 and for the NABM arm, 23, 20, and 21 at levels 1, 2, and 3, respectively, during course 1 (P = .01, .80, and .01, respectively). During course 2, ANCs to 0.1 x 10(9)/L and 0.5 x 10(9)/L were attained significantly faster at levels 1 and 3 in the ABM arm. ANC to 1.0 x 10(9)/L was comparable in both arms. Incidence of infection and duration of fever were similar in both arms. Although mortality and the incidence of delayed hematopoietic recovery were more frequent in the NABM arm, this was not statistically significant. Platelet recovery was consistently prolonged in course 2 in both arms, with demonstrable benefit of ABM in course 2 when dose levels were collectively considered. We conclude that (1) ABM enhanced recovery of ANC to 0.1 x 10(9)/L; (2) ABM did not decrease the incidence of infections and the duration of fever; and (3) CVP can be safely given without ABM to carefully selected patients.