Introduction:

Aggressive non-Hodgkin's lymphoma (aNHL) with secondary central nervous system (sCNS) involvement has a poor prognosis. Studies have reported a response to induction treatment as low as 35%, leaving less than half of patients eligible for autologous stem cell transplant (ASCT). Outcomes of patients in these clinical scenarios are dismal and treatment is ill-defined. Small case series suggest chimeric antigen receptor (CAR)-T cell therapy may play a role in the management of relapsed/refractory (R/R) B-cell lymphoma (BCL) with sCNS involvement, but follow-up is limited and response duration is uncertain. Allogeneic hematopoietic stem cell transplant (alloHCT) offers a durable remission for a subset of patients with R/R systemic aNHL primarily mediated through a graft versus lymphoma (GVL) effect, but it is unclear if GVL properties include the immune-privileged CNS. The present study aims to describe outcomes of a cohort of patients with R/R aggressive B- and T-cell NHL with sCNS involvement who underwent alloHCT at a single academic institution.

Methods:

This is a retrospective analysis that includes all patients with R/R aNHL with sCNS involvement who underwent alloHCT at the University of California, Los Angeles from 2005-2020. The UCLA Institutional Board Review approved this study. Relevant clinical data was extracted from medical records. Hematopoietic cell transplantation comorbidity index (HCT-CI) and time to neutrophil and platelet engraftment were measured according to Center for International Blood and Marrow Transplant Research criteria.

Results:

Ten patients were included (3 females, 7 males). Histologic subtypes included anaplastic BCL (1), mantle cell lymphoma (1), blastic natural killer-cell lymphoma (1), peripheral T-cell lymphoma, not otherwise specified (1), primary mediastinal BCL (1), and diffuse large B-cell lymphoma (DLBCL) (non-germinal center=3, germinal center-like=2). Two DLBCL patients had histologic transformed lymphoma (follicular lymphoma =1, chronic lymphocytic lymphoma = 1). Four patients had sCNS involvement at the time of initial diagnosis or during frontline treatment; the remaining 6 patients developed sCNS lymphoma at relapse. sCNS lymphoma was identified in the parenchymal (n=4), leptomeningeal (n=3), or both (n=3) compartments.

The median age at the time of alloHCT was 49.5 (range 28-68), and 1 patient was ˃ 60. At the time of alloHCT, 1 patient had residual disease in the CNS and the remaining 9 patients were in a complete remission. Eight patients received ˃ 3 prior lines of therapy, and 3 patients failed prior ASCT. HCT-CI scores were 0 (n=1), 1 (n=2), 2 (n=3), 3 (n=1), and unknown (n = 3). Donor types included 10/10 matched related (3), 10/10 matched unrelated (4), 9/10 mismatched related (1), and double umbilical cord blood (2). Graft source was peripheral blood in 8 patients and cord blood in 2 patients. Conditioning regimens included myeloablative, reduced intensity and non-myeloablative in 6, 3 and 1 patient(s), respectively. Six patients received total body irradiation-containing conditioning. The average time to neutrophil engraftment was 18 days (range 11-29), and the average time to platelet engraftment was 26.5 days (range 18-59). One patient had primary graft failure.

Of the 6 patients with day 100 disease reassessment (CR at time of alloHCT=5, PR in CNS at time of alloHCT=1), all were in CR. With a median follow-up of 341 days, 2 patients relapsed (CNS=1 and systemic = 1), and 6 patients died. Cause of death included infection (n=3), lymphoma (n=1), primary graft failure (n=1), and organizing pneumonia (n=1). Six patients developed acute graft versus host disease (GVHD) (grade 1-2, n=1; grade 3, n=5), and 4 patients developed chronic GVHD (score 1-2, n= 2; score 3, n=2). The median overall and progression-free survival for the entire cohort was 341 days (range 23-4825) and 268.5 days (range 23-4825), respectively.

Conclusions:

AlloHCT in patients with R/R aNHL with sCNS involvement is feasible and may provide a durable response in a subset of patients. Four patients remained alive and free of disease one year post-alloHCT and one patient converted from a PR to a CR in the CNS 100 days post-alloHCT, suggesting effective donor immune surveillance within the CNS. Transplant-related morbidity and mortality limits the widespread application of this therapeutic modality and less toxic approaches are urgently needed.

Disclosures

No relevant conflicts of interest to declare.

Sign in via your Institution