Salivary DNA methylation derived estimates of biological aging, cellular frequency and protein expression as predictors of oral mucositis severity and survival in head and neck cancer patients

Abstract

Background

Oral mucositis is a painful and debilitating condition that occurs in the majority of head and neck cancer patients receiving radiation and/or chemotherapy. While some patient and treatment related factors are known to contribute to the incidence and severity of disease, reliable biomarkers remain elusive. In the following study, we investigated the association of salivary DNA methylation derived biological aging, cellular frequency and protein concentration measures with the severity of oral mucositis and overall survival in a cohort of head and neck cancer (HNC) patients (n = 103).

Methods

DNA methylation profiling was performed on saliva samples obtained prior to treatment. Biological aging measures included Horvath2, PhenoAge, FitAge and GrimAge, and cellular frequency included epithelial and specific immune cell populations.

Results

Severe mucositis (i.e. grade 3 or 4) occurred in nearly half of patients. For malignant HNC patients (n = 84), every 1-SD increase in GrimAge was associated with 2.62-times risk of severe mucositis (95 % CI: 1.38, 5.57), while a 1-SD increase in monocyte frequency was associated with a decreased risk (OR [95 %CI]: 0.40 [0.18, 0.80]). Over a median follow-up of 53 months, 39 of 103 participants died. Six protein scores (TNFSF14, GCSF, MATN3, GDF8, nCDase, TNF-β) were associated with survival at q < 0.15.

Conclusion

We provide evidence that the risk-related biological aging measure GrimAge may be a useful predictor of mucositis severity in HNC patients. Salivary monocyte frequency may be protective against mucositis, and this measure could be used as a predictive biomarker while also providing clues into the pathobiology of the disease.