Post-Transplant Nivolumab Plus Unselected Autologous Lymphocytes in Truly Refractory Hodgkin Lymphoma Patients Is Highly Effective and Responses Are Mediated By NK Cells

Background

In the relapsed/refractory Hodgkin Lymphoma (HL) setting (RHL), the achievement of complete remission (CR) is fundamental in order to proceed with allogeneic stem cell transplantation (HSCT) consolidation. Immune checkpoint inhibitors (CI) have demonstrated clinical activity in patients relapsing after ASCT, although only 20% CR rate is observed. In order to improve the efficacy of CI in RHL, we previously reported a salvage treatment based on early CI therapy following post-autologous stem cell transplantation (ASCT) with the reinfusion of unselected autologous lymphocytes infusions (ALI). The rationale of this approach is based on the achievement of a disease debulking through ASCT conditioning. Early administration of CI (nivolumab) is performed on a minimal disease burden, while reinfusions of ALI reduce post ASCT immune depression eliciting CI efficacy.

Aims

The aim of this study was to evaluate the efficacy of the procedure in terms of CR rates and overall survival (OS) after an extended follow up. Biological endpoint was to investigate the lymphocyte subpopulations involved in the mechanism of response and investigate the role of NK-cells mediating anti tumor responses in HL.

Methods

Patient were eligible to this study after failure to respond to first line treatment (i.e. positive PET2 or PET6 scan) and underwent autologous lymphocyte apheresis upon enrollment. All patients received conventional 2nd line chemo followed by 3rd line chemo-immunotherapy with Brentuximab-Vedotin (BV) in non-responding patients. Patient failing to achieve CR after BV were enrolled in the treatment arm and proceeded to ASCT + CI and ALI, whereas patients responding to 2nd- or 3rd treatment received ASCT followed by ALI alone, as a control cohort (figure 1A).

NK cell degranulation assay was evaluated by staining for CD107a (LAMP-1) expression on healthy-donors derived NK cells exposed to HL cell lines (L428 and L250), in the presence or absence of monoclonal antibodies targeting inhibitory or activating receptors. Surface and intracellular staining was performed prior to flow cytometry and analysis was performed on the CD3-/CD56+ gated population.

Results

Twenty-one patients with RHL (median age 32 years; range 18-65) underwent lymphocyte apheresis after failure of 1st line chemo and then proceeded to salvage therapy. Thirteen patients failed to respond to 2nd and 3rd line therapy and thus received early post-transplant CI supported with four ALI. Eight patients responded to 2nd line chemo (n=6) or 3rd line BV (n=2) and received ASCT followed by ALI alone.

No adverse events were recorded, specifically, no immune-related toxicity was observed.

All patients receiving ALI + CI (treated patients) achieved negative PET scan CR. Eight of them received HSCT consolidation. One patient refusing HSCT after achieving CR in this study relapsed and responded to CI re treatment followed by HSCT.

All patients in the treatment arm are alive and disease-free after a median follow-up of 32 months (95% CI 26.4-51.0). Median disease-free survival (DFS) was not reached in treatment arm (Fig 1B).

Four of the patients in the control arm relapsed (50%). Three of them responded to 3rd line therapy and proceeded to HSCT. One patient achieved CR with conventional CI treatment but refused HSCT and the other relapsing patient died because of progressive lymphoma after failure to achieve CR with conventional single agent CI. Median DFS in the control arm was 22 months (Fig 1B).

NK cells showed higher expansion and a faster maturation in the treatment arm, compared to the control arm (p<0.05).

As expected, HL cells lines do not express HLA class I and II, but express activating NK cells receptors (aNKG2D, aNKp30, aNKp46, aDNAM-1). In vitro, the activation of each of those receptors was able to trigger NK cells degranulation. Similarly, PD-L1 and PD-L2 blockade on HL cell lines significantly increased NK cell degranulation (p<0.05). Taken together, those observations support the role of NK cells in this setting.

Conclusions

Our data show very high anti-tumor activity of ALI + CI for RHL, allowing most patients to proceed to HSCT. RHL patients in the treatment arm had an excellent outcome if compared to the control arm, which included patient who did respond to conventional treatment.

Biological data suggests that this approach may accelerate NK cell development/maturation and suggest that NK cells may mediate response to CI in HL.