Our understanding of the genetic abnormalities associated with thyroid cancers has grown significantly over the past decade. Thus, papillary thyroid cancers (PTC) have non-overlapping activating mutations genes encoding the growth factor receptors RET or NTRK, the three isoforms of RAS, or of BRAF, which altogether are found in ~70% of cases. In addition, mutations of effectors in the phosphoinositol-3-kinase (PI3K) pathway are present at various stages of the disease, but in particular in advanced thyroid cancers. There is also a strong genotype/phenotype relationship in thyroid cancers, with specific mutations associated with particular histological variants, and more importantly, with the biological behaviour and prognosis of the cancers. The BRAF mutation, in particular, is associated with invasive PTC, which more often become radioactive iodine refractory. The relevance of many of the genotypephenotype correlations first identified through association studies in human specimens is now supported by mouse models, in which oncogene activation through transgenic or gene targeting approaches recapitulates the main features of various forms of the disease. It is now legitimate to ask whether this information can be leveraged to make clinical decisions, and improve the care of patients with the disease. For example, FNA of thyroid nodules discriminates between benign and malignant thyroid cancers in most instances. Several groups have shown that immunohistochemical or genetic assays can refine the accuracy of cytopathological diagnoses. Most oncogenic mutations in thyroid cancer result in constitutively active kinases, which provide tractable targets for pharmacological inhibition. This is based on the assumption that cancers will remain dependent on the activity of the initiating oncoprotein for viability, despite accumulating further genetic damage during the course of their evolution. As a proof-of-principle, the multi-kinase inhibitor AZD6474, which inhibits RET, EGFR and KDR, showed promising evidence of activity in a phase 3 clinical trial of patients with medullary thyroid cancer. However, it is not clear that the beneficial effects of this agent are due to its inhibition of RET kinase, or to effects on other kinases. Several clinical trials incorporating patients with papillary, follicular and Hurthle cell carcinoma with multi kinase inhibitors have also been reported. Two studies with sorafenib, an inhibitor of KDR, RET and possibly RAF, showed a high percentage of patients with disease stabilization and some partial responses, particularly in PTC. Although sorafenib was developed as a RAF inhibitor, it is not clear if it is sufficiently potent to block the activity of mutant BRAF in vivo. Indeed, it has been proposed that the activity of sorafenib, as well as that of other multi kinase inhibitors such as axitinib and motesanib, may be due to their antiangiogenic activity, since they are all inhibitors of KDR, the receptor for vascular endothelial growth factor. Despite the promising results of these early trials, many important questions remain. It is not clear if the favorable responses will prove to be durable. Moreover, no study so far has been designed to examine impact on mortality. In view of these uncertainties, these drugs should be considered only for patients with rapidly progressing metastatic disease, ideally as part of a clinical trial. Although these compounds are quite well tolerated, they do have significant side effects, and many participants in the trials required dose reductions, drug holidays, and in some cases were unable to continue on the study. In other tumor types, the type of genetic mutation harboured by the cancer predicts response to therapy with specific kinase inhibitors. Despite this, most trials of thyroid cancers derived from follicular cells were not designed to examine whether specific oncogenic mutations might determine treatment outcome, a shortcoming that will hopefully be corrected as this promising research field continues to develop.
Generously supported by the Clinical Endocrinology Trust.