Genetic Studies

Genetic studies in the USRT cohort have focused on the radiation-related cancers of the breast and thyroid for which we have adequate sample sizes for evaluation. Initial studies evaluated candidate single nucleotide polymorphisms (SNPs) in pathways that might influence response to radiation exposure, such as DNA repair and base excision repair. Later studies sought to replicate promising findings from whole genome scan studies and assess potential interaction between genetic variants and ionizing radiation.

Breast Cancer

The ATM missense mutations Ser49Cys and Phe858Leu were associated with significantly increased breast cancer risk (Stredrick et al, 2006), but did not modify the radiation-related breast cancer risk (Bhatti et al, 2008). Main effect associations were found with CASP8 D302H and ILIA A114S and breast cancer risk with the ILIA A114S variant significantly modifying the relationship of personal diagnostic radiation and breast cancer (pinteraction = 0.004) (Sigurdson et al, 2007). Among 55 candidate single nucleotide poloymorphisms (SNPs) in DNA repair pathways, 5 SNPs were found to modify the association between radiation exposure (either from occupational or personal diagnostic radiation) and breast cancer. These were WRN V114I, BRCA1 D652N, and three in the PRKDC gene (IVS15+6C>T, IVS34+39T>C, and IVS31-634C>A (none of these were in linkage disequilibrium) (Bhatti et al, 2008). Of six ERCC gene variants examined in the nucleotide excision repair pathway, only ERCC5 rs17655 showed a borderline main effect association with breast cancer risk; however, individuals with variants in both ERCC5 D1104H and ERCC2 K751Q were more susceptible to the effects of occupational radiation (p for interaction = 0.10 and 0.01, respectively) (Rajaraman et al, 2008). Since ionizing radiation and steroid hormones cause oxidative damage to DNA, 12 SNPs in 12 estrogen biosynthesis and metabolism genes were studied with CYP1B1 V432L modifying the radiation-related risk for both cumulative personal exposure and occupational dose (pinteraction = 0.03 and 0.06, respectively) (Sigurdson et al, 2009). In a separate analysis of oxidative stress and inflammatory pathway genes, an interaction between the rs5277 SNP in the PTGS2 gene and radiation-related breast cancer risk was observed (Schonfeld et al, 2010). Analyses of polymorphisms in the Vitamin D biosynthesis pathway are nearing completion. We are also examining interaction between oxidative SNPs and cigarette smoking among parous women. We plan to evaluate several SNPs and risk of benign breast disease, a condition associated with both radiation exposure and increased breast cancer risk

Recent results from whole genome scans of breast cancer have been used to inform choices of variants for our studies. In collaboration with the international Breast Cancer Association Consortium (BCAC), several new SNPs were found to be associated with breast cancer risk (BCAC 2006; Cox et al, 2007; Easton et al, 2007; Garcia-Closas et al, 2008; Ahmed et al, 2009; Thomas et al, 2009; Gaudet et al, 2009). Among eleven BCAC risk SNPs, the genotype-associated breast cancer risk was found to vary significantly by radiation dose for rs2107425 in the H19 gene(pinteraction=0.001), a maternally expressed imprinted mRNA that is closely involved in regulating the IGF2 gene, and which could exert its influence by this or by some other radiation-related pathway (Bhatti et al, 2008). In the most recent effort based on results from Thomas et al (2009) and Ahmed et al (2009), we found that the radiation-associated breast cancer risk varied significantly by linked markers at 5p12 (rs930395, rs10941679, rs2067980, and rs4415084) in the mitochondrial ribosomal protein S30 (MRPS30) gene (pinteraction=0.04). MRPS30 functions in the apoptosis pathway and is likely to be involved in pre-apoptotic events, but the exact function of this gene is unknown (Bhatti et al, 2010). The promising variants and data from BCAC are being used to assess interaction with known breast cancer risk factors such as reproductive history and body mass index, but thus far no strong interactions have been identified (Milne et al 2010).

Thyroid Cancer

In a case-control study of 167 validated papillary thyroid cancer cases and 491 age and gender frequency-matched controls, a borderline significant increasing risk was found for RET G691S (Ptrend = 0.05) and was especially pronounced among young women (Lönn et al, 2007). For women under 38 years (the median age at diagnosis), the odds ratios were 2.1 (95% confidence interval, 1.2-3.7) for those heterozygous for the RET G691S polymorphism and 3.7 (95% confidence interval, 1.1-11.8) for those who were homozygous (Ptrend = 0.001). These data provide limited evidence that TSHR- and RET-related genes are related to papillary thyroid cancer risk. During 2010, in collaboration with investigators from the University of Texas M.D. Anderson Cancer Center, we completed a pooled analysis of several thousand SNPs and thyroid cancer. Pathways studied included genomic integrity, immunity, hormone metabolism, diet and obesity, and chemical carcinogens (e.g. cigarette smoking and alcohol consumption). The genomic region encompassing 8q24, a known hotspot for several other cancers, was also assessed. Results from these studies will be available in 2011.