Is There Still a Place for Total Body Irradiation (TBI) In the Conditioning Regimen of Autologous Stem Cell Transplantation In Mantle Cell Lymphoma ?: a Retrospective Study From the Lymphoma Working Party of the EBMT

Prognosis of Mantle cell lymphoma (MCL) has recently improved by the use of combination of monoclonal anti-CD20 immunotherapy, high-dose cytarabine containing regimen and autologous stem cell transplantation (ASCT). During the last decade the use of TBI, as part of the conditioning regimen of ASCT, has drastically decreased. This attitude is questionable since first, MCL cell lines exhibit a high sensitivity to radiotherapy in vitro and second, a retrospective study including 18 patients suggested that TBI may improve the EFS after ASCT in MCL (Milpied et al, BMT 1998). Aim. We performed a large retrospective study from the EBMT registry to assess the role of TBI in the conditioning regimen of ASCT in MCL. Patients and methods. All patients who underwent an ASCT between 2000 and 2007 in complete or first partial remission (CR/PR1) registered within the EBMT data-base with complete MedB forms available were eligible for this study. Results. Altogether 418 patients met the eligibility criteria. Median age was 51 years (range 29 to 65 years). At diagnosis 93% of the patients presented with a stage III/IV disease. Most of the patients (85%) had received one line of chemotherapy prior to ASCT. According to reported treatments, 122 patients (31%) had received Rituximab (R) and high-dose cytarabine (HD-Cyt) before transplantation and 111 patients (29%) had not received any of these drugs. At transplant, 283 patients (68%) were in CR and 135 (32%) in PR1. Conditioning regimen for ASCT contained TBI in 152 patients (36.5%) and consisted on a BEAM-based chemotherapy in 92% of the 266 patients transplanted with a non-TBI regimen. Stem cell source was peripheral blood for 99.3% of patients. With a median follow-up of 29 months, median overall and disease free survival (OS and DFS) of all patients were 99 and 57 months, respectively. Disease status at transplant appeared as a significant predictive factor for DFS and relapse incidence (RI) but at the time of analysis had no impact on OS. In comparison to patients transplanted in CR, those transplanted in PR1 had an impaired DFS (median DFS 52 months versus 40 months, respectively) (RR= 1,53, 95% CI 1.10–2.14, p=0.01) with an increased incidence of relapse (RR= 1.49, 95% CI 1.04–2.13, p= 0.03). Since we found a significant interaction between the use of TBI and disease status on the incidence of relapse, all further analysis were stratified on disease status at transplant. The use of R and HD-Cyt before ASCT or TBI in the conditioning of ASCT did not have any impact on OS, DFS and RI of patients transplanted in CR. In contrast, for patients transplanted in PR1, pre-transplant treatment with R and HD-Cyt was associated with a prolonged DFS in comparison to those who did not receive these treatments (median DFS 52 months versus 27 months, respectively, p=0.0891). Moreover, in patients in PR1 at transplant, the use of TBI was associated with a significant reduction of the incidence of relapse in both univariate and multivariate analysis (p=0.034 and RR= 0.409, 95% CI 0.213–0.786, p=0.007, respectively). There was a trend to a better DFS in patients transplanted in PR1 with TBI (median DFS 60 months with TBI versus 33 months without TBI, p=0.123). At the time of analysis, overall survival and non relapse mortality (NRM) were similar between patients who received TBI and those who did not. Three secondary malignancies were reported in this series of patients with two occurring in the TBI group Conclusion. In this retrospective series of autografted MCL, the risk of relapse after ASCT is mainly dependant on the disease status at transplant. TBI or other radiotherapy-based conditioning can be probably avoided in patients transplanted in CR but should still be considered in patients who can not achieve a better response than PR at transplant.

No relevant conflicts of interest to declare.