The Influence of PET/Gallium (PET/Gal) Status and High-Dose Rituximab (HDR) in Patients with Aggressive, Large, B-Cell Lymphoma (LBCL) Receiving Autologous Stem Cell Transplants (ASCT).

Background: PET/Gal status has been reported to be an important predictor of outcome in patients with LBCL who receive an ASCT. Newer conditioning regimens which include high-dose rituximab (HDR) have been shown to improve results (Khouri, JCO, 2005). The impact of HDR on the outcome of patients based on PET/Gal status has not been determined.

Methods: A retrospective review of patients with chemo-sensitive, LBCL who received an ASCT on a research protocol at MD Anderson between 1995 and 2005 was performed. Factors that were considered for outcome included: Age, IPI, # of prior chemotherapies, B2-microglobulin, disease status at transplant, HDR and PET/Gal status. In patients who received HDR, it was given with stem cell mobilization and then again on days +1 and +8 following transplant.

Results: A total of 188 patients were identified. Median age was 49 years with 108 (57%) male patients. 147 patients (78%) had de novo LBCL and 41 (22%) had a LBCL of follicular origin (LBCL-F). 83 (39%) patients received HDR. At transplantation, 95 patients (50%) were in PR, 71 (38%) in CRU and 22 (12%) in CR. 142 (76%) patients were PET/Gal negative, 37 (20%) PET/Gal positive and 9 (4%) were unknown. Median follow-up was 47 months. On multivariate analysis, for patients with de novo LBCL, PET/Gal status and HDR were the only predictors for progression and progression free survival (PFS). Patients who were PET/Gal negative and those that received HDR had a hazard ratio (HR) of 0.3 (p<0.001) and 0.5 (p=0.02) for progression, respectively (see the table below for the cumulative incidence (CI) for progression and PFS at 54 months according to HDR and PET/Gal status for de novo LBCL undergoing ASCT). PET/Gal Status and HDR were also found to be predictive for patients with LBCL- F on univariate analysis, however due to the small numbers in this subset; multivariate analysis could not be performed. PFS at 54 months for patients with LBCL-F who were PET/Gal negative was 40% versus 17% in the PET/Gal positive group, (p=0.006). PFS for those LBCL-F patients who received HDR was 81% as compared to 23% for those who did not receive HDR, (p=0.007).

Conclusions: The two most important predictors of outcome following ASCT are PET/Gal status and whether HDR was given with the transplant regimen. The addition of HDR to the transplant regimen decreases the risk for progression irrespective of PET/Gal status; however the improvement is more significant in patients with a negative PET/Gal scan.