Long Term Follow-up of High-Dose CD20-Targeted Radioimmunotherapy-Based Autologous Transplantation for Patients with Mantle Cell Lymphoma

Background: Autologous stem cell transplantation (ASCT) is commonly employed in the treatment of mantle cell lymphoma (MCL), though long-term remission following this intervention is relatively uncommon, particularly for patients with relapsed/refractory disease. We hypothesized that high-dose radioimmunotherapy-(RIT) based conditioning prior to ASCT could improve outcomes based on the radiosensitivity of MCL and reduced cross-resistance with chemotherapy.

Methods: We evaluated the outcomes of patients with MCL treated with myeloablative conditioning using iodine-131 (¹³¹I) conjugated to the anti-CD20 antibody tositumomab with or without additional chemotherapy followed by ASCT on 4 sequential prospective clinical trials. These trials included single agent RIT, RIT + etoposide + cyclophosphamide, and RIT + fludarabine. We retrospectively identified a comparator group of patients at our institution undergoing ASCT for MCL without RIT, using conditioning regimens including carmustine, etoposide, cytarabine, and melphalan (BEAM); busulfan, melphalan, and thiotepa (BuMelT); and 12 Gy of total body irradiation (TBI) plus cyclophosphamide with or without etoposide. We described and compared the outcomes from each group including differences in event rates evaluated by the Cox proportional hazards regression model. To account for imbalanced covariates between subgroups, adjusted estimates of survival were generated (Storer, et al., Lifetime Data Anal, 2008, p. 484). Follow-up was updated as of October 2013.

Results: From November 1995 to May 2011, 162 sequential MCL patients received high-dose therapy and ASCT at our center, including 61 (38%) who received RIT-based conditioning. RIT patients received a median of 19.8 GBq (534 mCi; range: 7.6-40.7 GBq [205-1100 mCi]) of ¹³¹I to deliver a median of 25 Gy (range: 20-27) absorbed dose to critical organs. Median follow-up from ASCT for all surviving patients was 5.1 yrs (range: 0.3-17.1). Patients treated with RIT-based ASCT were less likely to be in first remission (48% vs 72%; p = 0.002), be in complete remission (CR) at ASCT (26% vs 61%; p < 0.001), or have chemosensitive disease (84% vs 96%; p = 0.006) compared to controls. The 2-year, 5-year, and median PFS were 66%, 49%, and 64 months (respectively) for the control group compared to 70%, 45%, and 54 months (respectively) for the RIT group (p = 0.77; see Table). The 2-year, 5-year, and median OS were 76%, 64%, and 126 months (respectively) for the control group compared to 82%, 62%, and 80 months (respectively) for the RIT group (p = 0.75; see Table). Multivariate analysis (adjusting for chemosensitivity, remission status, ASCT in first remission, and age) indicated that RIT-based conditioning was associated with a reduced risk of mortality and treatment failure (i.e., death or relapse; see Table and Figure). Furthermore, the relative benefit of RIT increased with worsening pre-ASCT disease status (from CR to partial remission to stable/progressive disease), with HRs for mortality 1.1, 0.53, and 0.04, and HRs for treatment failure 1.2, 0.38, and 0.07 (respectively). Incidences were comparable between the RIT and control groups in terms of non-relapse mortality (NRM) (7% vs 8%, respectively) and secondary malignancies (9.8% vs 12%), including acute myeloid leukemia or myelodysplasia (4.9% vs 3%).

Conclusions: RIT-based conditioning is associated with improved outcomes following ASCT for MCL after controlling for imbalances in important risk factors, with the greatest benefit among patients with more advanced disease. Rates of NRM and secondary malignancies were comparable with both approaches. These data support the further study of high-dose RIT in a risk-adapted approach to ASCT for MCL.

Figure: Estimated PFS after adjusting for chemosensitivity, remission status, ASCT in first remission, and age following high-dose RIT-based or conventional (Control) ASCT for MCL.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal