Medullary thyroid carcinoma (MTC) is a malignant tumor originating from thyroid C-cells that can occur either in sporadic (70–80%) or hereditary (20–30%) form. In this study we aimed to identify recurrent copy number alterations (CNA) that might be related to the pathogenesis or progression of MTC. We used Affymetrix SNP array 6.0 on MTC and paired-blood samples to identify CNA using PennCNV and Genotyping Console software. The algorithms identified recurrent copy number gains in chromosomes 15q, 10q, 14q and 22q in MTC, whereas 4q cumulated losses. Coding genes were identified within CNA regions. The quantitative PCR analysis performed in an independent series of MTCs (n = 51) confirmed focal recurrent copy number gains encompassing the DLK1 (14q32.2) and AIFM3 (22q11.21) genes. Immunohistochemistry confirmed AIFM3 and DLK1 expression in MTC cases, while no expression was found in normal thyroid tissues and few MTC samples were found with normal copy numbers. The functional relevance of CNA was also assessed by in silico analysis. CNA status correlated with protein expression (DLK1, p = 0.01), tumor size (DLK1, p = 0.04) and AJCC staging (AIFM3p = 0.01 and DLK1p = 0.05). These data provide a novel insight into MTC biology, and suggest a common CNA landscape, regardless of if it is sporadic or hereditary MTC.

Context: Our group described a p.G533C RET gene mutation in a large family with multiple endocrine neoplasia type 2 syndrome. Clinical heterogeneity, primarily associated with the presence of lymph node metastases, was observed among the p.G533C carriers. Objective: The aim of this study was to use single-nucleotide polymorphism-array technology to identify copy number variations (CNVs), which are present in the constitutional DNA and associated with the established clinical and pathological features of aggressive medullary thyroid carcinoma (MTC), primarily the presence of lymph node metastasis. Design: Fifteen p.G533C carriers with MTC were chosen for the initial screening. The subjects were divided into two groups according the presence (n = 8) or absence (n = 7) of lymph node metastasis. Peripheral blood DNA was independently hybridized using a genome-wide single-nucleotide polymorphism Array 6.0 platform. The results were analyzed using both Genotyping Console and PennCNV software. To identify the possible candidate regions associated with the presence of lymph node metastasis, cases (metastatic MTC) were compared with controls (nonmetastatic MTC). The identified CNVs were validated by quantitative PCR in an extended cohort (n = 32). Results: Using two different algorithms, we identified nine CNV regions that may contribute to susceptibility to lymph node metastasis. The validation step confirmed that a CNV loss impacting the FMN2 gene was potentially associated with a greater predisposition to lymph node metastasis in this family (P = .0179). Finally, we sought to investigate whether the development of lymph node metastasis might not depend on a single CNV but rather a combination of various CNVs. These analyses defined a CNV pattern related to a more aggressive phenotype in this family, with CNV deletions being enriched in the metastatic group (P = .0057). Conclusion: Although hereditable specific RET mutations are important to determine cancer risk, germline CNVs in disease-affected individuals may predispose them to MTC aggressiveness.