Allogeneic Stem Cell Transplantation in Hodgkin Lymphoma: The Timing of Relapse after Autologous Transplant and Primary Refractory Disease Do Not Impact the Survival Outcomes

The present study retrospectively analyzed a cohort of consecutive relapsed/refractory Hodgkin lymphoma (RR-HL) patients receiving allogeneic stem cell transplantation (alloSCT) in a single center with a thiotepa-based conditioning to understand whether the timing of relapse after autologous stem cell transplant (ASCT) or the primary refractory disease would change the survival.

Of 247 patients with HL referred to our division from 2001 to 2014, 109 had RR-HL (64 primary refractory). After receiving salvage treatment, 62 patients (57%) underwent alloSCT with a thiotepa-based conditioning; 4 patients (4%) received fludarabine-melphalan conditioning in a phase II study, 43 patients (39%) did not receive alloSCT for progressive disease (31), advanced age (>65 years, 3 patients) or for achieving a complete response after a third-line chemotherapy consolidated by ASCT (9).

This study analyzes the outcomes of the 62 consecutive patients allografted with thiotepa-based conditioning, which consisted of thiotepa, fludarabine, and cyclophosphamide (TFC) for patients with an HLA identical sibling donor, TFC plus anti-thymocyte globulin for matched unrelated (MUD) donors, TFC plus alemtuzumab and 2-Gray (Gy) total body irradiation (TBI) for T-deplete haploidentical alloSCT, or TFC plus 2-Gy TBI and post-transplant cyclophosphamide for T-replete haploidentical alloSCT. Multivariate analysis of alloSCT outcomes included as covariates the pre-transplant disease status (CR vs PR vs resistant), donor (HLA identical, MUD, or T-deplete or T-replete haploidentical), primary refractory disease (yes vs no) and timing of relapse after ASCT (<12 mos after ASCT vs >12 mos vs no ASCT).

Patients had a median of 33 years at alloSCT, 76% of them had a primary refractory disease at diagnosis. 74% of patients relapsed <12 months after ASCT, 15% relapsed >12 months after ASCT, 11% underwent alloSCT without previous ASCT. At alloSCT, 25% had resistant disease whereas 75% were in partial (31%) or complete response (44%) after the last salvage treatment. Donors were HLA identical siblings (42%), MUD (29%), or haploidentical (21% T-deplete, 8% T-replete).

Median follow-up was 5.4 years. Three- and 5-years OS was 61 and 59%, PFS and relapse incidence were 46% and 38% at both 3 and 5 years. Non-relapse mortality (NRM) was 10% at 100 days, 17% at 1 year and for the entire follow-up. In multivariate analysis, the timing of relapse after ASCT and primary refractory disease did not impact the transplant outcomes. OS was reduced by resistant disease at alloSCT (HR=4.01, CI95% 1.34-11.97, p=0.012) and by T-depleted haploidentical transplant (HR=3.81, CI 95% 1.36-10.66, p=0.010). PFS and relapse incidence were impacted only by resistant disease (HR=5.54, CI 95% 2.13-14.37, p<0.001, and HR=6.75, CI 95% 2.07-21.96, p=0.001, respectively). NRM was significantly impacted only by the use of T-depleted haploidentical grafts (HR=7.63, CI95% 1.07-54.31, p=0.042).

In conclusion, the pre-transplant disease status and not the timing of relapse after ASCT or primary refractory disease, impacts OS, PFS and relapse of RR-HL patients allografted with thiotepa-based conditioning. An optimal response before alloSCT is critical to maximize the long-term benefit of alloSCT. In the era of novel agents this can be a realistic goal for the majority of patients.