Supporting data for "The Role of Nerve-Cancer Crosstalk in Head and Neck Squamous Cell Carcinoma and Adjuvant Immunotherapy"

Head and neck squamous cell carcinoma (HNSCC) has a high prevalence worldwide and a poor 5-year survival, posing a significant health burden. In recent years, the intricate relationship between nerves and cancer has been recognized to exert significant influences on the development and metastasis of HNSCC. A notable aspect of this interaction is perineural invasion (PNI), which involves the infiltration of cancer cells into peripheral nerves. The presence of PNI has been extensively investigated as a prognostic factor in HNSCC. Despite these findings, the precise mechanisms governing nerve-cancer crosstalk in HNSCC remain elusive.

The aim of this study is to investigate the role of nerve-cancer crosstalk in HNSCC and the adjuvant immunotherapy, namely, 1) to quantitatively estimate the predictive significance of PNI in HNSCC survival outcomes; 2) to investigate the predictive significance of neural-related gene transcriptome alteration in HNSCC prognosis and the benefit of immune checkpoint inhibitor therapy; and 3) to investigate molecular mechanisms involving nerve-cancer-crosstalk of regulating tumor aggressiveness and impairing anti-tumor immunity.

Firstly, a systematic review and meta-analysis was performed and 74 studies comprising 27,559 patients were included to analyze the prognostic role of PNI in HNSCC. The analysis revealed that PNI has a cumulative occurrent rate of 30% in HNSCC. PNI+ HNSCC patients exhibited worse overall survival, disease specific survival, disease-free survival, local recurrence, locoregional recurrence, locoregional relapse free survival, distant metastasis and distant metastasis free survival compared to those PNI– patients.

Secondly, a bioinformatic analysis was performed based on the transcriptome data of 546 HNSCC patients from The Cancer Genome Atlas database. Through this analysis, four neural-related hub genes (ITGA5, PYGM, GNG7 and ATP2A3) were identified to construct neural-related gene risk score (NRGRS). NRGRS-high patients had a worse overall survival than NRGRS-low patients. Tumors with high NRGRS were more likely to have high infiltration of naive CD4+ T cells, M0, M2 macrophages and resting mast cells, indicative of suppressive immune responses and reduced efficacy of immunotherapy.

Thirdly, a trigeminal ganglion neuron and OSCC cell coculture system was established in vitro and the surgical lingual denervation was performed in an immunocompetent mouse OSCC model to investigate the mechanism of nerve-cancer crosstalk. Notably, coculturing with neurons not only enhanced the proliferation, migration and invasion but also upregulated TGFβ-Smad2 signaling and PD-L1 expression of tumor cells, which were reversed by the inhibition of TGFβ. Lingual denervation slowed down tumor growth and improved the efficacy of anti-PD-1 treatment through decreasing TGFβ and PD-L1 expression and increasing CD8+ T cell infiltration and the expression of IFNγ and TNFα in the OSCC model.

In conclusion, the existing evidence strongly supports PNI as a critical prognostic factor for worse survival in HNSCC patients and the neural-related hub gene transcriptome alteration demonstrates their potential as predictors of both survival and immunotherapy response. Additionally, denervation of OSCC inhibits tumor growth and improves anti-PD-1 immunotherapy through downregulating TGFβ signaling and activating CD8+ T-cell activities. This study open avenues for developing new medications or strategies targeting nerve-cancer crosstalk for HNSCC management, particularly in optimizing immunotherapy for patients.