Breaking Bad: Researchers Find Mutations that Turn Thyroid Cancers Deadly

Illustration of human neck and head with thyroid gland highlighted and tumor growing on gland.

A small percentage of thyroid tumors appear to turn aggressive after acquiring specific mutations.

Most thyroid cancers are slow growing and nonthreatening, but a small percentage are aggressive and lethal. Some of these virulent thyroid cancers are classified as anaplastic (undifferentiated) and are nearly always fatal within a year. However, non-anaplastic tumors — which make up approximately 98% of all thyroid cancers — occasionally become deadly as well. Until now, it has been a mystery what causes these seemingly harmless growths to turn malignant.

Memorial Sloan Kettering researchers have identified several genes that when mutated appear to convert these thyroid cancers into tumors that can spread to the lungs or bones. The discovery, reported in Clinical Cancer Research, could enable earlier diagnosis and alter treatment decisions. It also could further the development of drugs that block the effects of these mutations and slow or stop tumors’ growth.

The study, led by head and neck surgeon Ian Ganly, involved examining 57 tumor samples from MSK patients who died from non-anaplastic thyroid cancer. The samples were analyzed using MSK-IMPACT™, a powerful genomic sequencing test that looks for mutations in more than 450 genes known to play a role in various cancers.

The researchers compared mutations in these fatal non-anaplastic tumors against mutations already known to be present in nonfatal tumors and in anaplastic tumors. They were looking specifically for mutations that seem to turn the innocuous non-anaplastic cancers into deadly ones.

Telltale Mutations

We were interested to see … how the genetic profile of patients who died differed from the other thyroid cancer patients.
Ian Ganly head and neck surgeon

“We were interested to see what the fingerprint of death looks like — how the genetic profile of patients who died differed from the other thyroid cancer patients,” Dr. Ganly says.

The research team found mutations in two genes that were not known to be associated with thyroid cancer: MED12 and RBM10. They also found a high prevalence of mutations in a gene called TERT. Although TERT mutations were known to be numerous in anaplastic tumors, their high incidence in aggressive non-anaplastic tumors suggests that they play an especially important role in thyroid cancer virulence.

Dr. Ganly says the lethal non-anaplastic tumors probably start out in a slow-growing state and then at some point acquire the additional mutations that turn them deadly. “It’s likely a progression model similar to what is seen in colorectal or head and neck cancers, where the tumors acquire mutations and become more aggressive,” he explains. “But we need to explore this further in a mouse model to be sure.”

Benefiting Diagnosis and Treatment

The implications of this finding are twofold. Patients diagnosed with non-anaplastic thyroid cancer whose tumors carry these mutations may need more aggressive treatment. While doctors at MSK and elsewhere are choosing to monitor many non-anaplastic tumors rather than remove them, this approach could be wrong when these mutations are found.

In addition, knowing that these mutations play an important role in thyroid cancer’s aggressiveness will guide the development of drugs. Dr. Ganly says there are drugs in early testing against these mutations in other cancers, but they need to be explored in a mouse model harboring some or all of these mutations relating to aggressive thyroid tumors.

He emphasizes the importance of the unique resources available to MSK researchers that enabled him and his colleagues to make this finding.

“We have a comprehensive database for all patients with thyroid cancer treated here, including all the pathology and clinical details, and how each patient fared,” he says. “We also have an extensive collection of tumor samples. The combination of those two things allowed us to do something that no one else at any other institution has been able to do.”

This work was supported in part by NIH grant P30-CA008748 and the LesLois Shaw Foundation.