15P - TIGIT+ CD8+ T cells limit the efficacy of PD-L1 blockade plus chemoradiotherapy in MSS locally advanced rectal cancer via NECTIN2-TIGIT interplay

Background

PD-1/PD-L1 blockade combined with chemoradiotherapy putatively benefited patients with dMMR/MSI-H colorectal cancer. However, the effect of PD-L1 blockade plus chemoradiotherapy for pMMR/MSS rectal cancer remains unclear, and the mechanisms influencing therapeutic response need further exploration.

Methods

A phase II, open-labeled, study was conducted to evaluate the efficacy and safety of adding Atezolizumab (PD-L1 blockade) to neoadjuvant chemoradiotherapy in MSS locally advanced rectal cancer (LARC). All patients received long-course radiotherapy plus capecitabine followed by three 21-day cycles Atezolizumab and TME surgery. Single cell RNA sequencing was performed on surgical resection samples.

Results

Twelve patients were recruited with 100% R0 resection. 50% (n=6) of MPR rate was observed, 3 of these patients achieved pCR. Interestingly, the proportion of T cells in the non-MPR group was significantly higher than in the MPR group, particularly TIGIT+ CD8+ T cells. However, further analysis indicated that the T cells in the non-MPR group were functionally exhausted. Compared to the MPR group, the interaction between NECTIN2+ cancer-associated fibroblasts and TIGIT+ CD8+ T cells was significantly enhanced in the non-MPR group. Mechanistically, the NECTIN2-TIGIT interaction limit the cytotoxic response by CD8+ T cells through inhibition of cyclic GMP-AMP synthase (cGAS)-STING pathway. Moreover, addition of TIGIT blockade to PD-L1 blockade plus chemoradiotherapy significantly enhances the therapeutic efficacy in MSS rectal cancer in vitro and in vivo.

Conclusions

NECTIN2-TIGIT serves as a crucial immune checkpoint for the efficacy of PD-L1 blockade plus chemoradiotherapy in MSS locally advanced rectal cancer, and the supplementary application of TIGIT blockade presents a rational approach for an immunotherapeutic combination strategy.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

J. Xu.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.