555 Challenges in predicting ICI response in HPV+ HNSCC: insights into the tumor microenvironment

Background

The relationship between HPV status and immune checkpoint inhibitor (ICI) response in head and neck squamous cell carcinoma (HNSCC) is not fully understood. Previously, we showed that immune-inflamed TMEs are associated with ICI response in HPV- HNSCC.¹ However, HPV+ tumors have distinct biological characteristics that influence the immune landscape. Therefore, our objective was to characterize the tumor microenvironment (TME) in HPV+ HNSCC relative to ICI response.

Methods

Comparative analysis of RNA-seq data from 74 HNSCC patients (40 HPV+ and 34 HPV-) receiving anti-PD-1 therapy and two public cohorts, TCGA-HNSC (n = 504)² and GSE127165 (n = 112)³ was performed. Functional gene expression signature (FGES) analysis⁴ and Kassandra cellular deconvolution⁵ were performed. T and B cell receptor (TCR, BCR) repertoires were reconstructed using MiXCR.6 GATK Pathseq⁷ and Vi-Fi⁸ were used to identify viral reads and viral-host chimeric reads, respectively.

Results

Cellular deconvolution of 594 HPV- and 164 HPV+ samples indicated HPV+ samples had higher immune (NK, CD8+ T, CD4+ T, and B) cell content, while HPV- samples were enriched with stromal components (fibroblasts and endothelial cells, p < 0.001). Congruent with these findings, HPV+ samples had stronger checkpoint inhibition, antitumor cytokine, and MHCII FGES and weaker CAF, extracellular matrix, and angiogenesis FGES (p < 0.001) compared to HPV- samples. Moreover, HPV+ tumors contained larger and more diverse TCR and BCR repertoires (p < 0.001). Pre-treatment HPV+ samples from two nivolumab-treated cohorts with 25 responders and 14 nonresponders were analyzed, but no associations were observed between ICI response and immune and stromal cell content in the TME, analyzed FGES, TCR and BCR diversity or clonality, viral transcript expression, or host genome integration. Interestingly, longitudinal analysis of 40 pre- and 37 post-treatment samples showed higher levels of EMT, CAF, macrophage, and checkpoint inhibition FGES in ICI responder and nonresponder post-treatment samples (p < 0.05), suggesting these TME dynamics were unrelated to outcomes.

Conclusions

Our study highlighted the disparity in underlying mechanisms of ICI response between HPV- and HPV+ HNSCC patients. While factors like an immune-enriched TME and robust expression of checkpoint inhibition FGES are reportedly influential in HPV- cases, these and other factors were not associated with therapy response in HPV+ HNSCC patients. These results are likely due to the predominantly immune-enriched TME characteristic of HPV+ HNSCC. Our findings underscored the importance of considering HPV status when investigating the molecular determinants of ICI response in HNSCC.

References

N. Kotlov et al. ‘Abstract 1441: Predictors of treatment response in a preoperative window of opportunity trial of nivolumab in resectable squamous cell carcinoma of the head and neck,’ Cancer Res., Jul. 2021; 81(13_Supplement): 1441–1441, doi: 10.1158/1538-7445.AM2021-1441. The Cancer Genome Atlas Network, ‘Comprehensive genomic characterization of head and neck squamous cell carcinomas,’ Nature, Jan. 2015; 517(7536): 576–582, doi: 10.1038/nature14129. Y. Wu et al. ‘Circular RNA circCORO1C promotes laryngeal squamous cell carcinoma progression by modulating the let-7c-5p/PBX3 axis,’ Mol. Cancer, Jun. 2020;19(1): 99, doi: 10.1186/s12943-020-01215-4. A. Bagaev et al. ‘Conserved pan-cancer microenvironment subtypes predict response to immunotherapy,’ Cancer Cell, Jun. 2021; 39(6), 845–865.e7, doi: 10.1016/j.ccell.2021.04.014. A. Zaitsev et al. ‘Precise reconstruction of the TME using bulk RNA-seq and a machine learning algorithm trained on artificial transcriptomes,’ Cancer Cell, Aug. 2022; 40(8): 879–894.e16, doi: 10.1016/j.ccell.2022.07.006. D. A. Bolotin et al. ‘MiXCR: software for comprehensive adaptive immunity profiling,’ Nat. Methods, May 2015;12(5): 380–381, doi: 10.1038/nmeth.3364. M. A. Walker et al. ‘GATK PathSeq: a customizable computational tool for the discovery and identification of microbial sequences in libraries from eukaryotic hosts,’ Bioinformatics, Dec. 2018; 34(24): 4287–4289, doi: 10.1093/bioinformatics/bty501. N. D. Nguyen, V. Deshpande, J. Luebeck, P. S. Mischel, and V. Bafna, ‘ViFi: accurate detection of viral integration and mRNA fusion reveals indiscriminate and unregulated transcription in proximal genomic regions in cervical cancer,’ Nucleic Acids Res., Apr. 2018; 46(7): 3309–3325, , doi: 10.1093/nar/gky180.

Ethics Approval

The cohort from Thomas Jefferson University was collected under ClinicalTrials.gov identifiers NCT03238365 and NCT03854032.