Long-term follow up of nivolumab/ipilimumab-primed immunotransplant in patients with Relapsed/Refractory DLBCL

Abstract Introduction: Patients with relapsed/refractory (r/r) diffuse large B-cell lymphoma (DLBCL) have poor outcomes, particularly those with progression after autologous stem cell transplant (ASCT) or chimeric antigen receptor T-cells (CAR-T). Although certain subsets of DLBCL may respond to checkpoint blocking antibodies, PD1/CTLA4 dual checkpoint blockade (DCB) has been ineffective in patients with r/r DLBCL with a notable 0% complete remission (CR) rate [Ansell et al 2016]. Homeostatic T-cell expansion and activation achieved by T-cell transfer into lymphodepleted recipients, as is standard in CAR-T therapy or ASCT, is the result of higher levels of “available” trophic cytokines including IL7 and IL15. However, this rapid expansion is coupled with homeostatic upregulation of maturation/exhaustion checkpoints including PD-1 and CTLA-4, leading to homeostatic inhibition which, in turn, can be prevented by DCB, increasing anti-tumor T-cell toxicity [Marshall et al 2019]. Here, we report long term follow-up of 6 patients enrolled on a phase Ib/II study of nivolumab/ipilimumab-primed “immunotransplant” (IT) for r/r DLBCL (NCT03305445). Methods: Six patients with progressive DLBCL following at least one line of standard chemoimmunotherapy were enrolled in Phase Ib of the trial between 2018 and 2020. On day -54, patients received rituximab 375mg/m2, followed by two cycles of DCB (ipilimumab 1mg/kg and nivolumab 3mg/kg q3weeks, starting on day -53), followed by rituximab on day -11 and peripheral blood T-cell harvest. Patients underwent lymphodepletion on day -5 to -3 with fludarabine 30mg/m2/d and cyclophosphamide 500mg/m2/d, followed by autologous T-cell transfer of DCB-primed T cells (IT) on day 0. This was followed by two additional cycles of DCB and maintenance nivolumab 240mg q14days for up to 2 years, intolerance, or disease progression. Results: In long-term follow-up, 2 of 6 patients are in durable CR without receiving subsequent systemic therapy at > 80 months post-IT. Patients received a median of 3.5 prior lines of therapy; 33.3% had received anti-CD19 CAR-T therapy and 33.3% had received prior ASCT. All 6 patients received at least one cycle of DCB. Overall response rate (ORR) was 50% (16.6% CR, 33.3% PR). One patient had progressive disease requiring salvage chemotherapy after one cycle of DCB and was taken off protocol prior to IT. Five patients (83.3%) received at least 2 cycles of DCB and IT. One patient died of unknown cause 1 month after IT (grade 5 TEAE) prior to response assessment. A 74-year-old female with non-germinal center (non-GC) DLBCL who had received 5 prior lines of therapy including ASCT achieved a PR at 1 month post-IT and a CR at 9 months post-IT and remains in durable remission at 82 months. A 47-year-old female with GC-type DLBCL with 4 prior lines of systemic therapy achieved a PR at 2 months post-IT. At 5 months post-IT, she developed localized progression vs pseudo-progression; maintenance nivolumab was stopped. Subsequent scans showed improving systemic tumor regression with continued growth of a left pelvic lesion. She received left pelvic radiotherapy 35 months post-IT and has subsequently remained in CR without additional systemic therapy at 80 months post-IT. A 58-year-old male with non-GC DLBCL with 5 prior lines of therapy including anti-CD19 CAR-T therapy, achieved a PR at 1.5 months post-IT, but had progression at 4 months post-IT. A 61-year-old male with leg-type DLBCL who had progressed after anti-CD19 CAR-T had progression at 2 months post-IT. Treatment emergent adverse events (TEAE) occurred in 100% of patients; grade ≥3 events occurred in 83.3% of patients and were generally related to lymphodepleting chemotherapy. The most common adverse events were fatigue (100%, grade ≥3 16.7%), fever (83.3%, 0%), neutropenia (83.3%, 83.3%), anemia (66.7%, 50%), anorexia (66.7%, 0%), dyspnea (66.7%, 0%), constipation (66.7%, 0%), diarrhea (50%, 0%), and vomiting (50%, 0%). No patients discontinued treatment due to adverse events. Conclusions: Although DCB has been shown to be minimally effective in r/r DLBCL, DCB-primed immunotransplant, a multi-modal approach to augment responsiveness and overcome T-cell exhaustion, led to durable remissions in heavily pre-treated DLBCL patients without subsequent systemic therapy. Further investigation of DCB-primed immunotransplant may uncover efficacy for common tumor types in which DCB has proven ineffective.