BCG enriches T_reg cells

A study suggesting that BCG therapy affects levels of regulatory T (T_reg) cells has shed light on immune regulatory responses to BCG.

BCG therapy represents one of the earliest immunotherapies and it is an effective treatment in a majority of patients. However, 20–30% of patients will develop recurrent disease that requires surgery. Thus, identifying targets to optimize BCG therapy could improve outcomes.

Studies have shown that patients with higher numbers of T cells than myeloid-derived suppressor cells respond better to BCG therapy. However, as T cells include various subpopulations, Derré and colleagues investigated levels of conventional regulatory T (cT_reg) cells and PD-L1⁺ T_reg cells in peripheral blood mononuclear cells (PBMCs) of healthy donors and patients with urothelial cancer, as well as in the urine of patients during BCG therapy. PD-L1⁺ T_reg cells are a novel subset that are different from cT_reg cells as they do not express FOXP3.

blocking PD-1–PD-L1 interaction might increase immune-related BCG activity

A small, but significant, increase in cT_reg cell frequency was noted in PBMCs of patients with urothelial cancer compared with healthy donors, and cT_reg cells were also enriched in the urine of patients undergoing BCG therapy. However, no differences were noted between preinstillation and postinstillation samples, suggesting a BCG-independent mechanism. Further investigation into the presence of the regulatory PD-L1⁺ CD4 T cell subset found that PD-L1⁺ T_reg cells were not clearly detectable in the blood of healthy donors or patients with urothelial cancer, but were found at high (variable) levels in the urine of patients during BCG treatment. In contrast to the effect of BCG on cT_reg cells, a higher frequency of PD-L1⁺ T_reg cells was detected in post-BCG-instillation urine than in pretreatment samples.

In vitro co-culture was performed to investigate PD-L1⁺ T_reg induction. Culture of T24 bladder tumour cells alone induced only low levels of PD-L1⁺ T_reg cells and BCG alone also had limited ability. However, combined T24 and BCG had a strong synergistic effect, inducing high levels of PD-L1⁺ T_reg cells. The team then tested the effect of PD-1–PD-L1 blockade in an autologous T cell proliferation assay in the presence of PD-L1⁺ CD4 T cells. In this study, CD4 T cells containing, or not containing, PD-L1-expressing cells (BCG and T24-stimulated or not stimulated, respectively) were sorted and co-cultured with autologous labelled total T cells with or without antibodies blocking PD-1–PD-L1 interaction. Antagonism of PD-1–PD-L1 had no effect on T cell proliferation when co-cultured with untreated CD4 T cells, but significantly increased CD4 and CD8 T cell proliferation when co-cultured in the presence of stimulated CD4 T cells. These data suggest that blocking PD-1–PD-L1 interaction might increase immune-related BCG activity.

The use of PD-L1–PD-1 axis blockade has revolutionized the treatment of bladder cancer. These data suggest that, in addition to PD-L1 expression on tumour cells or antigen-presenting cells, T cells themselves might be an important source of PD-L1 during BCG therapy. “A 2014 study showed that the level of PD-L1 expression on immune cells, but surprisingly not on tumour cells, was associated with clinical response to PD-L1 blockade in muscle-invasive bladder cancer patients,” commented Derré. “Thus, it could be interesting to study which immune cells express (or overexpress) PD-L1 upon BCG treatment in vivo.” Trials investigating the combination of BCG and checkpoint blockade are already underway, and this study provides further rationale for such trials to improve bladder cancer outcomes.