Abstract 4356: Finding the genetic determinants of response to radiotherapy

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Radiotherapy is very effective at local control of tumours but leads to adverse reactions in a minority of patients. This project aimed to predict which breast cancer patients will show radiation toxicity, allowing them to be offered alternative treatment (e.g. mastectomy). Previously, studies have focussed on single nucleotide polymorphisms (SNP) for association with adverse reactions, but there is growing evidence that genomic copy number variation (CNV) are frequent and significantly contribute to genetic variation and human diseases. The aim of this candidate gene association study was to determine copy number variants (CNVs) that influence radiation toxicity in cancer patients, and combine these with clinical indicators to contribute to a predictive algorithm. CNV data for 88 candidate genes were ranked based on CNV position, size and frequency. Paralogue ratio test (PRT) assays were designed for the top-ranked 13 genes. Typing of the CNVs in the analysed samples was performed using PRT method because it is cheap, simple and fast and requiring a small amount of genomic DNA. Copy number typing was carried out on 69 control samples (comprising 30 random DNA and 39 CEPH [Centre de ‘Etude du Polymorphisme Humain] specimens of three multigenerational reference families); and 401 samples from a breast cancer cohort (i.e. test samples) with good clinical characterisation and scored for late effects of radiotherapy. Two PRT assays in the DNA ligase 1 (LIG1) gene were used for typing. PRT amplification products were electrophoresed on agarose gel (PRT2 assay) and ABI 3031xl genetic analyser (PRT3 assay) respectively. Quantification of DNA bands was performed using Syngene and Genescan software respectively and statistical analysis in SPSS. The result of this study showed that 5.2 % of the CEPH samples and 0.3% of the breast cancer cohorts showed consistent evidence of raised copy numbers which suggest duplication of LIG1 gene in those samples. The result also showed moderate correlation between the two assays used (r=0.56). The findings from this study have re-inforced the need for future research to focus on further characterisation of this studied CNV either by using more PRT over a wider region around LIG1 or an array or sequence based approach. Future work would also attempt to establish whether cells from the breast cancer patients show changed expression of LIG1. Lastly, CNV typing would be repeated in further cohorts, though with a frequency of 0.3%, many more cohorts in the range of thousands might be required. With multiple patients with the variant, then it might be possible to detect significant differences in radiation toxicity phenotypes. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4356. doi:1538-7445.AM2012-4356