Thyroid malignancies are the most common type of endocrine tumors. Of the various histologic subtypes, anaplastic thyroid carcinoma (ATC) represents a subset of all cases but is responsible for a significant proportion of thyroid cancer-related mortality. Indeed, ATC is regarded as one of the more aggressive and hard to treat forms of cancer. To date, there is a paucity of relevant model systems to critically evaluate how the signature genetic abnormalities detected in human ATC contribute to disease pathogenesis. Mutational activation of the BRAF protooncogene is detected in approximately 40% of papillary thyroid carcinoma (PTC) and in 25% of ATC. Moreover, in ATC, mutated BRAF is frequently found in combination with gain-of-function mutations in the p110 catalytic subunit of PI3′-Kinase (PIK3CA) or loss-of-function alterations in either the p53 (TP53) or PTEN tumor suppressors. Using mice with conditional, thyrocyte-specific expression of BRAF^V600E, we previously developed a model of PTC. However, as in humans, BRAF^V600E-induced mouse PTC is indolent and does not lead to rapid development of end-stage disease. Here, we use mice carrying a conditional allele of PIK3CA to demonstrate that, although mutationally activated PIK3CA^H1047R is unable to drive transformation on its own, when combined with BRAF^V600E in thyrocytes, this leads to development of lethal ATC in mice. Combined, these data demonstrate that the BRAF^V600E cooperates with either PIK3CA^H1074R or with silencing of the tumor-suppressor PTEN, to promote development of anaplastic thyroid carcinoma.

Implications: This genetically relevant mouse model of ATC will be an invaluable platform for preclinical testing of pathway-targeted therapies for the prevention and treatment of thyroid carcinoma. Mol Cancer Res; 12(7); 979–86. ©2014 AACR.