Anti-PD-1 therapy can possibly reverse CAR T cells exhaustion in DLBCL

Exhaustion describes a condition in which T cells are unable to perform any function. It has been characterized by chronic infections and cancers.1 Various features within the tumour microenvironment led to dysfunction of T infiltrating lymphocytes, such as antigen persistence, nutrient deprivation and/or PDCD1 demethylation.2 This gene codes for the protein PD-1, whose expression is the most known feature of exhaustion. Targeting PD-1 with nivolumab or pembrolizumab led to impressive clinical success in the context of various metastatic tumours and Hodgkin lymphoma.3 Even though single-agent anti-PD1 showed low and transient antitumor activity in R/R DLBCL,4 with the exception of patients with HIV-related lymphoma, that may benefit of a long-term disease control.5 About 60 per cent of patients treated with axicabtagene ciloleucel for a relapsing or refractory diffuse large B cell lymphoma (R/R DLBCL) do not achieve a persistent response.6 Massive research effort has been undertaken to dissect CAR T cell failure. Among other causes, absence/low expansion, functionality or persistence of those engineered T cells had been related to exhaustion features. These characteristics have been described among harvested T cells, CAR T cells from the final product and/or circulating CAR T cells, themselves linked to tumour burden, tumour microenvironment and/or systemic inflammation.7 In this issue, Gazeau et al. report on their experience with nivolumab in that context. With that drug, they have been trying to rescue 11 patients that did not achieve a complete remission 30 days after axicabtagene ciloleucel for a R/R DLBCL. They treated patients in 3 different situations. The “primary refractory” group included patients who showed stable or progressive disease at initial evaluation, the “booster” group included patients who achieved a partial response and were belief to require additional therapy, and the “salvage” group included patients who progressed after achieving a partial response and therefore required salvage therapy. Nivolumab was administrated every 2 weeks, and efficacy was evaluated by PET/CT after four injections. Regarding safety, four mild respiratory tract infections were reported, and no CRS/ICANS was observed. Four patients achieved a complete response after 2 months of treatment, two patients were in the situation of partial response at D30 and two others were experiencing relapsing disease after a partial response. A second expansion of circulating CAR T cells was observed, after a single injection of nivolumab, which argues on reversion of CAR T cell exhaustion: Cmax was 0.035 G/L, lower that the initial expansion but following a long period when the CAR T cells were undetectable. Moreover, the secretion of pro-inflammatory cytokines is probably reflecting the functionality of those CAR T cells. Another group recently reported three patients with early progression or relapse, after being treated with 4-1BB costimulated CD19 CAR T cells, who responded to pembrolizumab.8 Ten out of the 12 patients treated in this study experienced a second expansion of circulating CAR T cell. Interestingly, pretreatment biopsy from seven out of nine patients had positive PD1 ligand expression (PD-L1). Other groups described that PD-L1 expression by DLBCL tumour cells is associated with activated B cell genotype, as well as poor outcomes.9 Taken together, those clinical data suggest a potential therapeutic benefit of the reversion of CAR T cell exhaustion with anti-PD-1 therapy. Ongoing clinical trials will help to determine the best candidates and modalities for such salvage.