36
Impact of Conditioning Regimen on Outcomes for Patients with Lymphoma Undergoing High-Dose Therapy with Autologous Hematopoietic Cell Transplantation (AutoHCT)
Introduction: High-dose therapy with AutoHCT is a standard component of therapy for many patients with Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL). Choice of high-dose therapy regimens has historically been driven by institutional tradition. No randomized prospective clinical trials have shown a difference in outcomes between conditioning regimens.
Methods: Using the Center for International Bone Marrow Transplant Registry (CIBMTR), we studied 4,917 patients with lymphoma (NHL n=3,905; HL n= 1,012) who underwent AutoHCT from 1995-2008 using the most commonly used conditioning regimens: BEAM (n=1730), CBVlow (n=1249), CBVhigh (n=604), BuCy (n=789), and TBI (n=545)-based regimens. The CBV cohort was divided based on the BCNU dose and its impact on outcome (above [high] vs. below [low] 375 mg/m2).
Results: The cohorts differed in: age <54y (BEAM 47%, CBVlow 61%, CBVhigh 66%, BuCy 50% and TBI 60%), HL (BEAM 18%, CBVlow 23%, CBVhigh 37%, BuCy 21% and TBI 4%) and year of transplant ≥2002 (BEAM 70%, CBVlow 18%, CBVhigh 26%, BuCy 49% and TBI 19%). 1-year rates of TRM ranged from 4% to 8% and did not differ across regimens in multivariate analysis. There were a significant interaction between regimens and disease (HL and NHL). For patients with NHL, there was no significant difference in outcomes with BEAM, CBVlow, BuCy, and TBI. In contrast, NHL patients treated with CBVhigh had higher progression (HR 1.28 [1.08, 1.50], p=0.003), and mortality (HR 1.27 [1.08, 1.49], p=0.003). For patients with HL, BuCy and TBI were associated with higher progression at 3 years compared to BEAM (47%, 57% vs. 36%) and shorter overall survival (65%, 47% vs. 79%). Multivariable results for HL are demonstrated in the table.
Conclusion: The impact of specific high-dose regimens with AutoHCT on overall outcomes appears to be different in patients with NHL and HL. Use of CBVhigh in patients with NHL and BuCy or TBI-containing regimens in HL were associated with worse outcomes compared to BEAM. Future investigation should be geared towards guiding the choice of regimen based on specific patient and disease characteristics.
Groups | HL | Progression | Overall Mortality | ||
| N | HR (95% CI) | p-value | HR (95% CI) | p-value |
BEAM | 313 | 1.00 | - | 1.00 | - |
CBVlow | 279 | 1.12 (0.85-1.47) | 0.43 | 1.53 (1.16-2.02) | 0.003 |
CBVhigh | 219 | 0.95 (0.71-1.28) | 0.74 | 1.54 (1.15-2.05) | 0.003 |
BuCy | 162 | 1.52 (1.13-2.05) | <0.001 | 1.77 (1.30-2.43) | <0.001 |
TBI | 23 | 1.94 (1.09-3.45) | 0.024 | 3.39 (2.03-5.64) | <0.001 |
See more of: BMT Tandem "Scientific" Meeting