Combination of immune stimulatory strategies to promote anti-tumour immunity

CD8+ T cells can kill cancer cells but are tightly regulated by receptors that confer positive or negative signals. Members of the tumour necrosis factor receptor superfamily (TNFRSF) can improve the responses of antigen-specific CD8+ T cells by enhancing their survival, proliferation and differentiation into effector and memory T cells. However, the roles of different members of the TNFR superfamily on CD8+ T cell function has not been fully explored. The aims of this thesis were therefore to investigate the effects of stimulation through different TNFRSF members on augmenting CD8+ T cell responses, controlling tumour growth and CD8+ T cell differentiation into memory cells. Initially I compared the efficacy of agonist antibodies to TNFRSF members CD27, GITR, 4-1BB and OX40 for their ability to expand adoptively transferred gp100-specific pmel-1 tumour-reactive CD8+ T cells in vivo. Anti-CD27 was the most potent agonist and further combined with the TLR ligands PolyI:C and LPS and with inhibitors of the check-point receptor PD-1 to enhance the accumulation of T cells. For anti-tumour immunity, adoptive T cell transfer and anti-CD27 as a monotherapy was ineffective against a lethal dose of melanoma. However anti-CD27 synergised with PD-1/L1 blockade to confer long term protection dependent on co-transfer of tumour-reactive CD8+ T cells. This combination treatment increased the frequency of adoptively transferred cells, and their expression of effector molecules such as granzyme B, IFN-훾 and TNF-훼 and the transcription factor T-bet, which is associated with T cell effector function. Interestingly in a colon carcinoma model, while anti-CD27 as a monotherapy conferred protection to a minority of mice, this did not synergise with PD-1 blockade. I found that colon carcinoma cells express less PD-L1 compared with melanoma cells suggesting that the efficacy of PD-1 blockade may depend on local concentrations of PD-L1. I then investigated whether the rat anti-mouse CD27 mAb was still effective when converted to a syngeneic isoform to move towards clinical therapy. I found that anti-CD27 of the mouse IgG1 isotype was an effective agonist whereas, when converted to a mouse IgG2a form, CD27 positive cells were depleted. Finally, across all experiments I noted that adoptively transferred pmel1 CD8+ T cells did not persist after contraction and, in contrast to previous work, IL-2 and the mTORC inhibitor rapamycin did not lead to their increased maintenance. The affinity of pmel-1 cells for gp100 peptide is relatively low. To gain insight into whether the affinity of the TCR/peptide interaction influences memory CD8+ T cell generation, I made use of OT-1 TCR transgenic mice for which a range of altered peptide ligands of different affinities have been described. These data revealed that the magnitude of the primary CD8+ T cell response is dependent on both peptide affinity and density. However OT-1 CD8+ T cells differentiated into memory T cells and expanded equally after secondary stimulation following priming with either the low or high affinity peptide in the presence of anti-CD27. Moreover, the combination of anti-CD27 plus PD-1/PD-L1 delivered with either the low or high affinity peptide synergised for increased OT-1 CD8+ T cell expansion and anti-tumour immunity. Together my data show that anti-CD27 and PD-1/L1 blockade may be a particularly potent combination for enhancing low affinity CD8+ T cells specific for cancer cells