Treatment of patients with thyroid cancer is usually successful, and most patients are cured of the disease. However, we do not have effective therapies for patients with invasive or metastatic thyroid cancer if the disease is not surgically resectable and does not concentrate radio-iodine. Conventional external beam radiotherapy and chemotherapy are of marginal benefit. In other types of cancer, new therapies are being developed that take advantage of our knowledge of cancer pathogenesis to interfere with the activity of specific oncoproteins believed to be important in disease causation. Because these approaches are being considered for thyroid cancer, I will briefly describe in this review examples of recent breakthroughs in medical therapy of certain hematological malignancies and some solid tumors using drugs that work in this fashion, focusing in particular on compounds that block the enzymatic activity of specific tyrosine kinase oncoproteins. It should be noted, however, that cancers commonly harbor mutations or other disruptions of many genes, each of which could conceivably play a role in disease pathogenesis. This makes the choice of molecular target a difficult and critical decision if these approaches are to succeed. Here I will argue that priority should be given to blocking the function of oncoproteins activated early in tumor development. We have a fairly good understanding of the genetic changes involved in thyroid cancer initiation, and hence these cancers may prove to be particularly well suited for oncoprotein-specific therapies.