Allogeneic Stem Cell Transplantation Provides Durable Remission in Patients with Primary Mediastinal Large B-Cell Lymphoma

Background: Primary mediastinal large B-cell lymphoma (PMBCL) is a subset of aggressive B-cell non-Hodgkin lymphoma (B-NHL) with distinct biological and clinical features. Although most patients are cured with frontline chemoimmunotherapy with or without radiation therapy (RT), relapsed or refractory (rel/ref) PMBCL is much harder to control. Standard treatment of rel/ref PMBCL is similar to other aggressive B-NHLs, including salvage therapy and autologous (auto) stem cell transplantation (SCT) in chemosensitive patients. Recently, immunotherapy with PD-1 blockade and chimeric antigen receptor modified T-cells has proven to be effective in rel/ref PMBCL. Despite this, allogeneic (allo) SCT retains an important potential role as it has curative potential for patients with advanced aggressive B-NHLs. However, there are scant modern data on alloSCT outcomes in patients with PMBCL, limited to case reports or small series. We therefore performed a multicenter retrospective study to evaluate alloSCT outcomes in patients with rel/ref PMBCL.

Methods: We retrospectively studied consecutive patients with rel/ref PMBCL who underwent alloSCT at Fred Hutchinson Cancer Center, Dana-Farber Cancer Institute, Massachusetts General Hospital, or City of Hope between 1/2000 and 5/2014. Baseline and transplant characteristics are reported descriptively. Progression-free survival (PFS) and overall survival (OS) were calculated using the Kaplan-Meier method. Incidence of relapse and non-relapse mortality were calculated using competing risks methods.

Results: 28 patients with rel/ref PMBCL underwent alloSCT at participating institutions during the study period. Among these patients, median age at SCT was 36 years, 54% were female, median number of prior therapies was 4 (range, 2-7), 57% were refractory to frontline therapy, 86% received prior RT, and 71% had prior autoSCT. At alloSCT, 1 (4%) patient was in complete response (CR), 21 (75%) were in partial response (PR), and 6 (21%) were refractory to pre-alloSCT therapy (18 patients were assessed with PET). Most patients (86%) received reduced intensity conditioning, most commonly fludarabine/melphalan +/- ATG or Zevalin (25%), fludarabine/TBI200 (21%), or fludarabine/busulfan (14%). GVHD prophylaxis most frequently consisted of a calcineurin inhibitor (CNI) with mycophenolate mofetil (12, 43%), CNI with sirolimus +/- methotrexate (8, 29%), or CNI with MTX (4, 14%). 15 (54%) patients had a matched (8/8) related donor, 8 (29%) had a matched unrelated donor, 2 had a mismatched unrelated donor (7/8), and 3 had umbilical cord donors. All patients received peripheral blood stem cell grafts except for the 3 cord recipients.

The median follow-up time in survivors was 5.0 (range 0.5-14.0) years. The 2 year PFS and OS in the cohort were 39% and 45%, respectively, while non-relapse mortality (NRM) and cumulative incidence of relapse (CIR) were 32% and 29%, respectively. The 5-year PFS, OS, NRM, and CIR were 34%, 45%, 32%, and 33%, respectively. The cumulative incidence of grade II-IV and III-IV acute GVHD were 39% and 4% at day 100, while the incidence of chronic GVHD at 1 year was 21% (18% extensive). Among patients in CR/PR at the time of alloSCT, the 2-year PFS and OS were 50% and 58%, respectively, as compared to a 2-year PFS and OS of 0% in patients who were refractory at the time of alloSCT (p=0.046 for PFS, p=0.014 for OS).

One patient received post-alloSCT lenalidomide as maintenance therapy and remained in ongoing CR. Of the 9 patients who relapsed after alloSCT, 3 out of 4 patients exhibited a response to immunosuppression taper, while 4 out of 5 patients responded to subsequent systemic therapy. 2 patients underwent a donor lymphocyte infusion (DLI) and both developed subsequent GVHD - 1 patient had a CR documented 64 days after DLI while the other had continued disease progression. In the 9 patients who relapsed after alloSCT, the 2-year OS was 33%.

Conclusions: AlloSCT can produce durable remissions in a subset of patients with heavily treated, rel/ref PMBCL. Patients with refractory disease at alloSCT had dismal outcomes. Despite the expanding treatment options available for these patients, alloSCT should be considered in the management of patients with rel/ref PMBCL who are sensitive to salvage therapy.

Figure 1A PFS and OS After AlloSCT in Patients with Rel/Ref PMBCL

Figure 1B PFS in Patients with Sensitive versus Refractory PMBCL at AlloSCT

View largeDownload slide

View largeDownload slide

Close modal