Allogeneic Hematopoietic Cell Transplantation (HCT) for Advanced Mycosis Fungoides and Sezary Syndrome (MF/SS): Impact of Increasing the Use of Unrelated Donors (UD) - the EBMT Lymphoma Working Party Experience

Introduction: The EBMT Lymphoma Working Party has previously reported on the long-term outcome of allogeneic HCT for patients with advanced MF/SS [J Clin Oncol 2014; 32: 3347-8]. Among a number of disease and transplant factors influencing patient outcome, the use of UD showed to be the strongest independent factor influencing overall survival (OS). The main shortcoming of the original reports was a limited number of 60 cases in the series, of which only 15 received allogeneic HCT from UD. As UD have been increasingly used during recent years, we sought to extend our previous analysis (1997-2007) to include patients with MF/SS allografted between 2008-2011. 

Patient and Methods: Endpoints were OS, progression-free survival (PFS), non-relapse mortality (NRM) and incidence of disease relapse/progression (DRP). Eligible were patients >= 18 years who were registered with the EBMT and had received an allogeneic HCT for MF/SS between 1997-2011. Centers with eligible patients were contacted to provide additional treatment and follow-up information including a written diagnostic report. Data were collected from the EBMT Registry (closed in July 2014), and endpoints were defined and analyses performed according to EBMT statistical guidelines (www.ebmt.org).

Results: Eligible for final analysis were a total of 113 patients, including our original 60 cases (1997-2007) and 53 new cases (2008-2011): 71 men and 42 women, median age at HCT 48 years (21-72), 77 MF (68%) and 36 SS (32%), with 7 EORTC/ISCL stage IIB, 17 stage III, 46 stage IV-A and 26 stage IV-B (17 missing). Demographics of both time periods were comparable, except for a marked increase in the use of UD (15/60, 25% vs 29/53, 55%; p=0.001) and a reduction in the use of TBI for conditioning (30/60, 50% vs 15/53, 28%; p=0.031) in recent years. At HCT, 52 patients (46%) were refractory or in relapse/progression and 61 (54%) in complete or partial remission. Eighty-six patients (76%) received reduced-intensity (RIC) and 27 (24%) myeloablative (MAC) conditioning regimens, including TBI in 45 cases (40%).

With a median follow up in survivors of 72 months (IQR: 39-97), allogeneic HCT for MF/SS offers an estimated OS of 56% at 1 year, 44% at 3 years and 38% at 5 years, and PFS of 34% at 1 year, 28% at 3 years and 25% at 5 years. NRM was 26% at 1 year and 28% at 3 years and thereafter. DRP was the main cause of treatment failure, with a probability of 40% at 1 year, 44% at 3 years and 47% at 5 years, and a mortality rate after DRP of 70% (35/50). It is worth noting that 15 patients (30%) remain alive at last follow up despite DRP, suggesting that some of these patients can be successfully rescued with donor lymphocyte infusions and other therapeutic interventions. The cumulative incidence of acute GVHD was 47% at day 100, and chronic GVHD 35% at 1 year, 45% at 3 years and 48% at 5 years. Interestingly, the univariate analysis showed a statistical trend towards a poorer OS in the cohort of new cases registered from 2008 (p=0.106). However, transplant period had no significant impact when included as a covariate in multivariate analysis, and appears to associate with the higher percentage of UD transplants in the new cohort. The use UD remained the main negative independent factor for OS (HR: 0.490; 95CI: 0.283-0.848; p=0.011) and PFS (HR: 0.468; 95CI: 0.259-0.843; p=0.011) in the multivariate models unstratified and stratified by inclusion period. The use of TBI in conditioning appears to have an independent effect to reduce the risk of DRP in the multivariate analysis (HR: 0.427; 95CI: 0.199-0.917; p=0.029), but does not translate into OS or PFS.

Conclusions: This extended series confirms the existence of a clinically relevant graft-versus-lymphoma effect in MF/SS, permitting long-term disease control in a substantial proportion of high-risk patients. Follow-up studies need to address the still significant adverse effect of UD and the role of TBI conditioning in order to improve HCT results in these otherwise fatal disorders.