Studying the B and plasma cell responses to the immune checkpoint blockade treatment of cancer

Abstract

Immune checkpoint blockade (ICB) therapies targeting PD-1 and CTLA-4 have revolutionised cancer treatment. However, their efficacy is not completely understood, especially regarding their effects on anti-tumour B and plasma cell responses, which are now regarded as important components of the adaptive immune response to cancer. Additionally, the utility of ICB therapies is limited by the immune-related adverse events (IrAEs) they induce – toxic autoimmune conditions that affect various organs throughout the body. As with ICB efficacy, it is not entirely understood how ICB therapy leads to the activation of autoreactive B cells, which may drive IrAE pathology. Considering these gaps in understanding, this thesis aims to study how the B and plasma cell responses to cancer are affected by combination anti-PD-1 + anti-CTLA ICB therapy, predominantly using a mouse model of ICB-sensitive melanoma (YUMMER1.7). Firstly, the intra-tumoural B cell response was interrogated by flow cytometry. Whilst their frequency was slightly higher in ICB-treated mice, intra-tumoural B cells comprised a small portion of the CD45+ compartment. In Mock- and ICB-treated mice, intra-tumoural B cells were mostly IgD- and did not express the activation markers CD80 and CD86. Intra-tumoural germinal centre frequency was also similar between treatment groups, with IHC analysis of tumour samples showing no evidence of tertiary lymphoid structure formation. Finally, preliminary V(D)J sequencing data of IgD- B cells from an untreated YUMMER1.7 tumour suggested that there was minimal clonal expansion and low levels of somatic hypermutation within the class-switched B cell population. Next, the B and plasma cell response within the tumour-draining lymph node (TdLN) of mice was assessed by comparison with non-draining lymph node (NdLN) tissue from the same mice. Tumour-induced TdLN germinal centre and plasma cell responses were detected. Surprisingly, however, combination ICB therapy appeared to decrease germinal centre response magnitude but significantly increase plasma cell response magnitude by ~10-fold. The follicular T cells compartment (Tfh and Tfr cells) was also expanded in the TdLN of ICB-treated animals. By utilising genetically engineered mouse models, the TdLN humoral response was investigated further. Using the AgRSR mouse, it was shown that B cells, which received antigen stimulation before tumour cell injection, contributed to the TdLN humoral response. Furthermore, in IL-6Rα KO mice, the TdLN germinal centre and plasma cell responses of ICB-treated mice were diminished without any loss in anti-tumour efficacy. Finally, scRNA-seq of the TdLN germinal centre and plasma cell response in Mock- and ICB-treated animals was performed. Surprisingly, the expression of the immune-modulator Cd52 distinguished between B and plasma cell clusters that were phenotypically similar but spatially separate on the UMAP plot. CD52-low naïve B cells downregulated IGHM, indicative of an anergic phenotype. CD52-low germinal centre B cells displayed elevated Bach2 and Il21r expression, suggesting they may hold a competitive advantage within a germinal centre reaction. Interestingly, cells from ICB-treated mice appeared to preferentially display the CD52-low germinal centre phenotype compared to cells from Mock-treated animals. In the V(D)J analysis, the clonality of cells in plasma cell clusters from ICB-treated mice was greater compared to equivalent cells from Mock-treated mice. However, the opposite was true for cells within germinal centre clusters. Finally, antibodies from the top expanded clones from each mouse were cloned to explore their reactivity. Of the clones that displayed binding, all bound to intracellular antigens within the YUMMER1.7 cell line, but also in other tumour cell lines (YUMM1.7 and MC38). This complemented the AgRSR mouse data and further suggests that the TdLN humoral immune response is directed against conserved self-antigens. In summary, this thesis has performed the first characterisation of how the TdLN B and plasma cell responses are affected by combination anti-PD-1 + anti-CTLA-4 ICB therapy using an orthotopic mouse model of melanoma. The data suggests that the TdLN serves as a hub for the activation of B cells responding to antigens not exclusive to the YUMMER1.7 tumour cell line. Interestingly, whilst ICB therapy reduces the magnitude and clonality of the TdLN germinal centre response, it significantly increases the magnitude and clonality of the plasma cell response. Further work is required to validate fully the autoreactivity of the TdLN response, but the findings from this thesis suggest the response is directed against self-antigens. This could have important implications for how ICB therapy leads to the activation of autoreactive B and plasma cells that drive IrAE pathology. Blockade of IL-6 signalling could be a strategy to mitigate this potentially pathogenic humoral response. Finally, this thesis provided preliminary data to suggest that CD52 plays an important role in germinal centre and plasma cell responses. Further work is required to see if these findings are replicated across different tumour models and other immunological settings.