Whole-exome ctDNA sequencing reveals intratumoral heterogeneity and drivers of relapse in locally advanced head and neck cancer

Abstract

Background: The mechanisms underlying tumor evolution and treatment resistance in locally advanced head and neck squamous cell carcinoma (LA-HNSCC) are not fully understood. This study used whole-exome sequencing (WES) of paired tumor tissue and circulating tumor DNA (ctDNA) to analyze intratumoral heterogeneity (ITH) and clonal dynamics at diagnosis and relapse.

Patients and methods: Between January 2017 and September 2022, we enrolled 152 patients with LA-HNSCC treated with radical chemoradiotherapy. Tumor tissue and plasma samples were collected at baseline and at relapse. WES was carried out on DNA extracted from tissue, plasma, and germline samples. Oncogenic variants were analyzed to evaluate tumor evolution and ITH. Concordance, pathway alterations, and associations with survival were examined.

Results: Out of the 152 patients selected, 81 were excluded based on clinical criteria. Of the 71 remaining patients, 37 had the complete set of samples of adequate quality for analysis. The most frequent mutations involved TP53 (48%), KMT2D (34%), and NOTCH1 (34%) at diagnosis. Pathogenic germline variants, including actionable mutations in BRCA2, CHK2, and KIT, were identified in 13.5% of cases. Concordance between tissue and plasma was low (median 11.8% at baseline, 12.4% at relapse), with up to 67% of oncogenic variants found only in ctDNA. Longitudinal analysis revealed limited overlap (10.3%) between baseline and relapse ctDNA, with some emerging resistance-associated mutations in PI3K-mTORC2 and ATM-CHK2 pathways. Alterations in IL6-JAK-STAT3 and RHO GTPase pathways were enriched in locoregional relapses (adjusted P < 0.001). Mutations in MMP15 and PTPRE were linked to shorter locoregional progression-free survival, whereas PDE2A mutations were associated with prolonged progression-free survival.

Conclusions: WES of ctDNA uncovers extensive ITH and identifies molecular drivers of resistance not captured by tissue biopsies. These findings support the use of plasma ctDNA analysis to monitor tumor evolution and personalize follow-up in LA-HNSCC, especially given the anatomical complexity that often limits repeated tissue sampling.

Keywords: LA-HNSCC; biomarkers; ctDNA; intratumor heterogeneity; precision medicine.