Predicting the outcome and toxicity in radiotherapy of head and neck cancers (TIARA)

A total of 200 patients will be recruited for this prospective multi-center study. All the patients will be followed-up for 3 years to record LRC, OS, quality of life (QOL) and toxicity. Biomarkers exploited in this study include imaging biomarkers [magnetic resonance imaging (MRI), CT imaging, cone beam CT imaging (CBCT), positron emission tomography (PET) imaging, and hyperspectral imaging (HSI)], predictive assays for radiosensitivity and susceptibility (RILA, phosphorylated ATM, SNPs), and cancer biomarkers (cfDNA, tumour mutational spectrum, RNA expression). Predictive models will be developed to correlate the biomarker data with efficacy outcomes and toxicity. The predictive models created in this study can be used to individualize the RT treatments (total dose and fractionation) of HNSCC patients.

OBJECTIVES OF THE PROJECT

The overall aim is to carry out a prospective multi-center study on 200 head and neck squamous cell carcinoma (HNSCC) patients treated in curative intent with radiotherapy (RT) only or concomitant chemoradiotherapy (cCRT) with cisplatin.

Our hypothesis is that individual radiosensitivity can be detected early (1-3 weeks from the beginning of RT) by using a set of biomarkers.

The main objectives of the study are:

• recruit 200 HNSCC patients from four Finnish university hospitals, to collect biological samples, patient data and imaging data.
• identify markers predictive for individual radiosensitivity and health outcome, including:
  – imaging biomarkers [magnetic resonance imaging (MRI), CT imaging, cone beam CT (CBCT), positron emission tomography (PET), and hyperspectral imaging (HSI)]
  – markers defined by predictive assays for radiosensitivity and radiosusceptibility [lymphocyte apoptosis assay (RILA), phosphorylated ATM, single nucleotide polymorphisms (SNPs), gene expression]
  – cancer biomarkers [cell-free DNA, tumour mutational spectrum, RNA expression].
• determine radiation doses to nearby organs at risk (OAR) and the correlation with toxic reactions.
• develop predictive models correlating the biomarker data with treatment efficacy outcomes and toxicity outcomes. The predictive model created in this study can be used to individualize (total dose and fractionation) the RT treatments of HNSCC patients.
• develop recommendations for personalised RT. The use of predictive biomarkers would allow personalizing RT treatments, which has the ability to improve treatment outcome [locoregional control (LRC) and overall survival (OS)] and reduce treatment related toxicity.
• identify future research needs.

The project consortium consists of four university hospitals (KYS, TYKS, OYS and Tays), University of Eastern Finland, University of Oulu, and STUK – Radiation and Nuclear Safety Authority.