ICHNO-ECHNO 2022 - Abstract Book

S3

ICHNO-ECHNO 2022

Materials and Methods We collected 32 spatially distinct biopsies from 13 resected primary oral cavity tumors (median 3 biopsies/tumor). The biopsies were snap-frozen at Amsterdam UMC (the Netherlands) or at the University Hospital Parma (Italy). None of the patients received treatment before surgery. On the DNA from all 32 biopsies, we performed lcWGS for copy number alterations (CNAs) and target-enrichment sequencing using a panel of 35 cancer genes which are frequently altered in HNSCC. Results Data from lcWGS demonstrated minor variations in all CNA profiles. The results of the target enrichment sequencing showed that all biopsies from each patient shared most somatic mutations. However, different mutations were observed in the biopsies of 5 out of 11 tumors. In these apparent heterogeneous tumors, extra mutations were observed in 1 to 3 genes with variant allele frequencies between 0.02 and 0.27. Conclusion By analyzing multiple spatially distinct biopsies from 11 oral cavity tumors we confirmed the presence of intratumor genetic heterogeneity with lcWGS and target enrichment sequencing. It remains to be determined whether this indeed explains the discordancy between mutational profiles between paired primary tumors and LRs after chemoradiotherapy. 1 Beatson West of Scotland Cancer Centre, Clinical Oncology, Glasgow, United Kingdom; 2 University of Glasgow, Institute of Cancer Sciences, Glasgow, United Kingdom; 3 Beatson West of Scotland Cancer Centre, Therapy Radiography, Glasgow, United Kingdom; 4 Glasgow Experimental Cancer Medicine Centre, Clinical Research Facility, Glasgow, United Kingdom; 5 University of Glasgow, Institute of Cancer Sciences , Glasgow, United Kingdom; 6 University of Glasgow, CRUK Clinical Trials Unit, Glasgow, United Kingdom; 7 University of Glasgow, Institute of Cancer Sciences, Glasgow, United Kingdom Purpose or Objective Poorer loco-regional control drives survival differences between good and poor prognosis oropharyngeal squamous cell cancer (OPSCC), suggesting radio-resistance is a significant factor. There are no established biomarkers which identify resistant tumours at baseline or during radiotherapy (RT) which would then allow stratification into alternative or intensified treatment strategies. Amino acid metabolism plays a central role in the response of tumours to RT and underpins radioresistance mechanisms. Radiation generates reactive oxygen species (ROS) which damages DNA and triggers cell death. Cancer cells use metabolism in two key ways to combat and survive radiation, firstly, by using amino acids to synthesise antioxidants (e.g. glutathione) to detoxify ROS. Secondly, by using amino acids and other nutrients to make nucleotides to repair damaged DNA, allowing tumour cells to survive. The aim of this study was to assess the feasibility of measuring baseline circulating metabolites in patients with poor prognosis OPSCC and to correlate these with response to RT. Materials and Methods Patients with intermediate and high risk OPSCC scheduled for radical (chemo) RT were recruited to this study. Disease control at 2 years and pattern of any relapse were noted. Patients who were alive and disease free were classed as responders to RT, those who had relapsed disease, non-responders. Whole blood samples were collected at baseline (i.e. pre-treatment). Polar metabolites were extracted and analysed on a high-performance liquid Chromatography mass spectrometry (LCMS) system. Peak area data was used for relative quantification and comparative analysis between responders and non-responders. Results Forty patients with a minimum of 2 years follow up were included. Multivariate statistical analyses were performed on the LCMS data to identify a metabolic signature in radio-resistant individuals. The data were examined by non-supervised principal components analysis (PCA) and partial least squares-discrimination analysis (PLS-DA). The results of multivariate statistical analysis were visualized and integrated with volcano plots to identify metabolites with significant difference between the survivors and deceased groups. A total of 152 metabolites showed significantly different abundance between responder and non-responder groups, see figure 1 for examples. OC-0008 Predicting radiotherapy resistance using circulating metabolites in poor prognosis OPSCC C. Paterson 1 , A. Huerta Uribe 2 , L. Hay 3 , A. Duffton 3 , L. Devlin 3 , M. MacLeod 4 , J. Evans 5 , E. Soulis 6 , O. Maddocks 7