In vivo purging of bone marrow in children with poor-risk neuroblastoma for marrow collection and autologous bone marrow transplantation

PURPOSE To evaluate the following prospectively in poor-risk neuroblastoma (NBL) patients: (1) the feasibility and efficacy of in vivo purging of bone marrow; and (2) the outcome after autologous bone marrow transplantation (ABMT) when immunologically tumor-free, unpurged autografts were used. PATIENTS AND METHODS Twenty-three children with poor-risk NBL were evaluated during induction chemotherapy by repeat bone marrow examinations, including aspirate, biopsy, and an immunofluorescence method using the anti-GD2 monoclonal antibody 3A7. Nineteen patients completed the program with surgery with or without local irradiation followed by ABMT. RESULTS Autologous bone marrow grafts, both immunologically and cytologically clean, were obtained and used in 19 of 23 children. The overall 4-year disease-free survival of the 19 grafted children was 53%, with a toxic death rate of 16% and a posttransplant relapse rate of 37%. According to the in vivo purging efficacy of the 18 children with initial marrow disease, the following three groups were formed: patients with (1) perfect in vivo purging (n = 5); (2) eventually successful in vivo purging (n = 8); and (3) unsuccesful in vivo purging (n = 5). The 4-year DFS was 100%, 67%, and 0%, respectively (P < 0.001). The five patients with unsuccessful in vivo purging failed because of resistant/progressive bulky disease. CONCLUSION In patients with poor-risk NBL, in vivo purging of bone marrow by conventional chemotherapy is feasible, can be monitored, and the purging efficacy during the first 3 months after diagnosis is a strong prognostic factor reflecting tumor responsiveness to therapy. Autografting with immunologically clean, unpurged marrows gives a DFS well comparable to previous studies using ex vivo purging.