Peripheral-blood stem-cell transplantation versus autologous bone marrow transplantation in Hodgkin's and Non-Hodgkin's lymphomas: A new matched-pair analysis of the European group for blood and marrow transplantation registry data

Purpose: To address the question of short-term and long-term advantages of peripheral-blood stem-cell transplantation (PBSCT) over autologous bone marrow transplantation (ABMT), we have reviewed the data of 3,214 patients with lymphoma, 2,859 undergoing ABMT, and 355 undergoing PBSCT. Patients and Methods: Analysis of prognostic factors for progression-free survival (PFS) was conducted separately for non-Hodgkin's lymphoma (NHL) (N = 1,915) and Hodgkin's disease (HD) (N = 1,299). In multivariate analysis, the relevant factors were status at transplant for NHL and sex, size of largest mass at transplant, status at transplant, and conditioning regimen for HD. The pair analysis was carried out by matching NHL and HD patients separately by their prognostic factors. Additionally, NHL patients were matched for histology, whereas both HD and NHL patients were matched for date of transplant, With this method, 454 patients were matched in the NHL group and 256 were matched in the HD group. Results: The overall survival (OS) and PFS unexpectedly were better for ABMT versus PBSCT patients in the HD group (OS, 65.3% at 4 years for ABMT v 52.7% for PBSCT; P = .0198). There was no difference in OS or PFS in the NHL group (OS, 56.6% at 4 years for ABMT v 52.7% for PBSCT; P = .4148). The overall relapse or progression rate at 4 years for NHL was 42% after ABMT and 49.2% after PBSCT (P = .1220); for HD, it wets 40% and 58.6%, respectively (P = .0164). Transplant-related mortality was lower, but not significantly, with PBSCT: 7.0% for ABMT versus 3.5% for PBSCT in NHL (P = .1356) and 7% for ABMT versus 4.7% for PBSCT in HD (P = .6056). Hematologic recovery occurred faster significantly with PBSCT irrespective of disease. Conclusion: This study confirms the advantage of PBSCT in terms of hematopoietic reconstitution, but it fails to show any superiority in the long term. Poorer results for both progression free and overall survival observed in HD patients who are receiving PBSCT are unexplained and should be confirmed with randomized studies. (C) 1997 by American Society of Clinical Oncology.