Abstract 423P
Background
Intense multimodality therapy leads to treatment associated morbidity in patients, thereby generating a need for de-intensification strategies. Individualized treatment for these patients is hindered due to lack of biomarkers that can identify patients that are at low risk of recurrence.
Methods
We applied an unbiased approach (weighted gene correlation network analysis (WGCNA)) that allows detection of autocorrelated gene sets on 4 independent cohorts of HPV+ HNSCC including a novel UNC cohort of 56 patients that we sequenced and a prospective clinical trial study.
Results
22 consensus transcriptional modules were identified by selecting genes that grouped together in WGCNA analyses from all 4 cohorts. Interestingly, only 1 module intrinsically divided HPV+ HNSCC into 2 subtypes, displaying different mutational profiles, mutational signatures, HPV gene expression patterns, HPV integration status, and patient survival. GSEA revealed that this module was enriched in NF-kB genes and strongly associated with NF-κB signaling, confirming our prior findings that defined HPV+ HNSCC subtypes by presence or absence of NF-kB regulators, TRAF3 or CYLD defects and by the NF-κB activity classifier. NF-kB-driven intrinsic tumor characteristics contribute to increased sensitivity of NF-kB active HPV+ head and neck tumors to radiation, providing patients survival benefits. TRAF3 or CYLD deletion dramatically increased radiation sensitivity of HPV+ head and neck cancer cells. The survival benefit of HPV+ vs. HPV-negative HNSCC is attributable to the subtype of HNSCC with highly active NF-κB, suggesting that this subtype is more responsive to treatment. Our studies also revealed a strong inverse relationship between NF-kB activity and NRF2 signaling and increased NRF2 activity is associated with radiation therapy resistance in many cancers, including HPV-negative HNSCC; however, HPV-positive HNSCCs rarely harbor NRF2 activating mutations.
Conclusions
Risk stratifying subtypes of HPV+ HNSCC will serve as the foundation for new or de-intensified therapies. Tumor subclassification based on high or low NF-κB and NRF2 signaling may help guide selection of therapeutic intensity for patients with HPV+ HNSCC.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
NIH.
Disclosure
All authors have declared no conflicts of interest.