**3. Genetic factors associated with sporadic DTC**

#### **3.1 Findings from association studies**

Sporadic DTC, which represents the majority of all NMTC, is considered as a complex disease caused by multiple environmental and genetic risk factors. Common polymorphisms in low-penetrance genes or altering their expression are hypothesized to play an important role in sporadic DTC. The numerous association studies conducted in the general population used a case–control design that compares the distribution of the polymorphisms in a population of affected versus unaffected individuals. Two types of association studies can be performed: the candidate gene approach that focuses on a limited number of polymorphisms located in genes select based on *a priori* knowledge of their biological function and the genome-wide association studies (GWAS) approach that consists in screening the association between the disease and the genetic variants along the genome without prior hypothesis.

The candidate gene approach usually focuses on functional or tags polymorphisms located in specific genes selected based on their potential functional impact on the gene product or on its potential interaction with known environmental/lifestyle risk factors for the disease. Candidate genes selected for analysis in association studies on DTC are mostly involved in the following biological pathways: DNA repair, cell cycle regulation, and apoptosis, xenobiotic metabolism, thyroid function, MAPK pathway, immune response and inflammation, and obesity [65–67].

Figlioli et al. [67] proposed an exhaustive review of polymorphisms investigated in association studies published before September 2013. They reviewed 100 original articles and five meta-analyses and reported 91 significant SNPs (over 316 analyzed) from 127 genes. They also conducted a meta-analysis on 46 SNPs, which were reported by at least two studies, and reported 13 significant SNPs, of which six are located in the coding sequence of candidate genes: *ADPRT* (rs1136410; p.Val762Ala), *BRCA1* (rs16942; p.Lys1183Arg), *XRCC7* (rs7830743; p.Ile3434Thr), *TP53* (rs1042522; p.Pro72Arg), *MTHFR* (rs1801133; p.Ala222Val), *RET* (rs1800862; p.Ser836Ser), one is intronic (rs4658973 in *WDR3*) and six SNPs are located in intergenic regions highlighted by GWAS. Therefore, many candidate genes and polymorphisms were considered in association with DTC but only a few were properly replicated.

With the completion of the human genome sequencing in 2003, GWAS involving hundreds of thousands to millions of SNPs across the human genome became more and more common. GWAS are usually composed with one discovery phase that aims to analyze a large number of variants, followed by a replication phase consisting of validation of the most significant variants in an independent sample. Since 2009, seven GWAS were published on NMTC risk, of which six were conducted in individuals of European ancestry [68–73] and one was conducted in individuals of Asian ancestry from Korea [74] (**Table 2**). The latest GWAS conducted by Truong et al. [73] also included a small discovery sample of individuals of Oceanian ancestry but with no replication set. Among the GWAS conducted in the European ancestry population, one study focused on radiation-related PTC, with cases recruited in Belarus and aged 0−18 years old at the time of the Chernobyl accident [69]. All these GWAS were based on a relatively low number of cases compared to other cancers GWAS and the largest study included 3,001 cases is a meta-analysis of several studies with no replication phase [72].


*NMTC: non-medullary thyroid carcinoma, DTC: differentiated thyroid carcinoma, PTC: papillary thyroid carcinoma, TSH: thyroid stimulating hormone*

#### **Table 2.**

*Details on published genome-wide association studies on NMTC risk.*

**Tables 3** and **4** summarize the significant and suggestive loci highlighted by these seven GWAS. All these variants are located in intronic or intergenic regions, except rs6793295 which is a missense variant in *LRRC34* at 3q26.2. The loci that were replicated in independent studies are shown in **Figure 2**. The most robust associations, *i.e.* the ones that were reported by several GWAS and independent sample sets, are for variants at 9q22.33, 14q13.3, 2q35, and 8p12.

