| Summary: | Thyroid cancer is the most common malignancy of the endocrine system. The majority of thyroid cancers derive from the follicular cells, being designated as non-medullary thyroid carcinomas (NMTC). NMTC also occurs in a familial form, entitled familial non-medullary thyroid carcinoma (FNMTC), representing 3-9% of all thyroid cancers. FNMTC is currently defined by the diagnosis of two or more first degree relatives with differentiated thyroid cancer of follicular cell origin. The family members frequently present benign lesions of the thyroid, such as multinodular goiter (MNG). Several FNMTC susceptibility genes have been reported, such as NKX2.2, FOXE1 and DICER1, but these are mutated only in a small fraction of families. Thus, the molecular basis of FNMTC is still mostly unknown, being regarded as a genetically heterogeneous disease. Recently, germline truncating mutations in DNA repair-related genes have been described in cases of thyroid cancer. In addition, activating mutations in BRAF and RAS oncogenes have been reported to be involved in familial thyroid tumour progression. In order to further clarify the molecular basis of FNMTC, the main goal of this project was to identify novel susceptibility gene(s) for this disease. To achieve this aim, three different approaches were used: the study of a very representative family of our cohort (six affected members with NMTC) through wholeexome sequencing (WES); the analysis of 94 genes associated with hereditary cancer predisposition in 48 probands from FNMTC families, through next-generation sequencing (NGS), using a commercial panel (Trusight Cancer Kit); and the analysis of the candidate genes telomerase reverse transcriptase (TERT) and eukaryotic translation initiation factor 1A X-linked (EIF1AX), selected based on previous evidence for their involvement in familial and/or sporadic thyroid cancer. The analysis of the FNMTC family with six-affected members through WES generated above 300,000 variants for each sample. Significant variants were selected through bioinformatics analysis, sets of filters, and validated using Sanger sequencing. In silico prediction, literature and database search of gene function and expression, were used to select potentially pathogenic variants, which was followed by analysis of their segregation with the disease in the family. The variant (c.701C>T, p.Thr234Met) in SPRY4 gene was prioritised for functional studies. To disclose the contribution of this variant to thyroid tumourigenesis, functional studies were performed using three cell models (NIH/3T3, PCCL3 and TPC-1). Overall, mutant SPRY4, compared to the wild-type, induced an increase in cell proliferation/viability, colony formation and in phosphorylation levels of proteins involved in MAPK/ERK and PI3K/AKT pathways. These results are in accordance with the well-established role of these signalling mechanisms in thyroid tumour development. Overall, these data suggested, for the first time, a role for SPRY4 in familial thyroid cancer initiation. In the NGS analysis of 94 genes in the 48 probands from FNMTC families, a total of 20,160 variants were identified. In silico analysis of NGS data unveiled 47 likely pathogenic germline variants in genes involved in DNA repair (33 variants) and in other hereditary cancer predisposing genes (14 variants). From these variants, only 18 segregated with FNMTC in 13 families, of which 15 variants were in DNA repair genes (APC, ATM, CHEK2, ERCC2, BRCA2, ERCC4, FANCA, FANCD2, FANCF, BRIP1 and PALB2), two in DICER1, and one in RHBDF2. These results reinforced the relevance of DNA repair genes and DICER1 in FNMTC aetiology and extended the present knowledge, by suggesting CHEK2, ERCC4, FANCA, FANCD2, FANCF, PALB2, BRIP1 and RHBDF2 as susceptibility genes for this disease. The main mechanisms involved in DNA repair, which may be altered according to this study, include the repair of double-strand breaks by homologous recombination (CHEK2, ATM, BRIP1, BRCA2, FANCD2 and PALB2 genes) and by DNA interstrand crosslink repair (FANCA and FANCF genes); the other mechanisms include repair of single-strand breaks by nucleotide excision repair (ERCC4 and ERCC2 genes) and by base excision repair (APC gene). In the approach that involved the direct study of candidate genes in FNMTC, using Sanger sequencing, no potentially pathogenic germline variants were identified in the TERT promoter in the 75 FNMTC families’ probands. However, in 54 familial thyroid tumours studied, we identified mutations in TERT promoter (9%), BRAF (41%), and RAS (7%), but no mutations were identified in EIF1AX. TERT-positive samples were also positive for BRAF, and this co-occurrence was statistically significant (p=0.008). In addition, TERT mutations in concomitance with BRAF mutations, had a significant correlation with more advanced tumour stages (T4) (p=0.020). This study showed that TERT promoter mutations are not frequently involved in FNMTC aetiology, but rather implicated in tumour progression and aggressiveness, when coexisting with BRAF mutations. Overall, we have identified novel susceptibility genes that are likely to participate in FNMTC tumourigenesis: SPRY4 seems to explain thyroid cancer aetiology in the FNMTC family studied and DNA repair genes may also be involved in FNMTC initiation. The oncogenic role of RAS and BRAF mutations in familial thyroid tumour progression was confirmed, and coexistent mutations in TERT promoter and BRAF were, for the first time, implicated in the progression and aggressiveness of FNMTC. This study improved the present knowledge of the genetic basis of FNMTC and further supported that this is a genetically heterogeneous disease. The identification of these genes involved in the initiation and progression of FNMTC, if supported by future studies in other cohorts, may allow families with this disease to undergo early diagnosis, and improve the clinical management of these patients.
|
|---|