Thyroid Cancer Genomics May Redefine Diagnosis, Treatment

Alexander M. Castellino, PhD

October 23, 2014

The genomic landscape of papillary thyroid cancer (PTC), the most common form of thyroid cancer, has just been redefined with respect to its somatic molecular alterations.

In a new study published October 23 in Cell, the analysis identifies new mutations, expands on information of alterations already known, and provides insights into what the future may hold for the diagnosis and clinical management of patient with thyroid nodules and PTCs.

"This understanding of the genomic landscape of thyroid cancer will refine how it's classified and improve molecular diagnosis. This will help us separate those patients who need aggressive treatment from those whose tumor is never likely to grow or spread," Thomas J. Giordano, MD, PhD, professor of pathology at the University of Michigan Medical School, Ann Arbor, said in a university press release.

TCGA Research Network

The Cancer Genome Analysis (TCGA), a project funded by the National Institutes of Health, has led the way in redefining molecular landscapes for several cancers, including breast, lung, ovary, and colon cancers.

In this report, an analysis was undertaken in the largest collection of PTCs from patients primarily without radiation exposure, which is a known PTC risk factor. Tumor samples and matched germline DNA were available from 496 patients and included various histologic types: classical type, follicular variant, tall-cell variant, and some uncommon types of PTC.

The first level of analysis provided mutational information by using different "platforms" (see figure below). At this level, several new mutated genes were identified, including EIF1AX, PPM1D, and CHEK2. Mutations were also identified in MAPK-related genes, BRAF, NRAS, HRAS, and KRAS. Mutations in tumor suppressor genes (eg, TP53, RB1, PTEN), numerous gene fusions (eg, RET and BRAF fusions) and TERT were in the newly expanded mutational landscape of PTC.

Identifying Driver Mutations

Unlike many other cancers, the thyroid genome is considered relatively quiet because it harbors fewer mutations and copy number alterations. This meant that the researchers could perform some integrated bioinformatic analyses not possible with more complex cancers.

Using data from multiple genomic platforms, TCGA researchers could define in greater detail mutational "drivers." In this analysis, researchers identified drivers in all but 3.5% of cases — down from 25% of tumors without apparent driving alterations.

The drivers were divided into two main groups: "BRAFV600E-like" mutations and "RAS-like" mutations. Using a 71-gene signature, a BRAFV600E-RAS score (BRS: –1 to +1) was developed, which showed a distinct separation of BRAFV600E-like from RAS-like tumors (see figure in the article's Graphical Abstract).

A thyroid differentiation score (TDS), developed on the basis of expression of thyroid metabolism and function genes, indicated a correlation across all the tumors analyzed. RAS-like tumors were uniformly highly differentiated, while BRAFV600E tumors displayed significant variation in differentiation. Currently, all BRAF-mutant PTCs are considered to be a homogeneous clinicopathologic group. However, the TDS indicates otherwise.

Ramifications for PTC Classification

"We will no longer be able to talk of BRAFV600E as a single tumor type. Indeed, the granular level at which we have teased out some of these mutations tells us that tumors with the most commonly understood and studied mutation in thyroid cancer — BRAFV600E — are not as homogeneous as is commonly believed," Dr Giordano said.

"PTC is too nonspecific a diagnosis," said Ronald J. Koenig, MD, PhD, also from the University of Michigan Medical School but not associated with the study. "It is not easy to predict how a panel of mutations will make a cancer more aggressive. That's where TCGA researchers have helped," he told Medscape Medical News.

This comprehensive analysis informed TCGA researchers that thyroid cancer, especially the follicular variant of PTC, begs to be reclassified according to molecular characteristics.

This would allow endocrinologists and oncologists to identify slow-growing, highly differentiated tumors from aggressive, less-differentiated tumors, Dr Giordano told Medscape Medical News.

Indeed, classical-type PTCs are rich in RET rearrangements, follicular-variant PTCs have RAS-like molecular features, and the aggressive tall-cell variants are enriched in BRAFV600E.

Designing clinical trials to evaluate targeted agents can benefit from these observations and the underlying genetics can help inform trial results, Dr Giordano said.

But although TCGA network has re-examined the molecular landscape across several tumor types, the role of the pathologist is not going to be passé, Dr Koenig told Medscape Medical News. The two are going to complement each other, he added.

Implications for Clinical Practice

In the clinical management of PTC, genetic testing is not routine, Dr Koenig told Medscape Medical News: "Genetic testing is done in less than 5% of patients."

In patients with thyroid nodules whose biopsy specimens are indeterminate, a gene signature can help assess the risk for malignancy and therefore can aid in clinical decision making, he said. There is no standard of care for integrating genetics into the clinical management of patients with PTC, Dr Koenig said.

According to Dr Koenig, "Most patients are cured by surgery or radioactive iodine [RAI] treatment." Differentiation is important if RAI therapy is to work, Dr Giordano explained to Medscape Medical News. Undifferentiated carcinomas and some less differentiated PTCs have lost their avidity for RAI. Indeed, RAI-resistant PTCs are enriched in BRAFV600E mutations, he added.

Although Dr Koenig agrees that TCGA has expanded understanding of the molecular defects across PTC, he indicated that most patients with BRAFV600E> actually do well with the current standard of care. TCGA has helped us understand how thyroid cancers with one bad gene can become aggressive when they accumulate additional mutations, he added.

In this regard, a separate but recent study published online October 20 in the Journal of Clinical Oncology found that the prevalence of BRAFV600E in 2000 patients with PTC was 48.5%. More significantly, PTC recurred in 20.9% of these patients. The investigators concluded that patients with BRAFV600E PTC were at an 82% higher risk for PTC recurrence compared with patients who did not have this mutation.

Those researchers indicated that PTC recurred even in patients with low-risk disease and suggested that "management of these patients is highly controversial."

Indeed, TCGA researchers would suggest that molecular testing of these patients may help in the risk stratification and clinical management of these patients.

"Our study will take molecular testing to the next level," Dr Giordano told Medscape Medical News. The expanded mutational landscape of thyroid cancer will improve the preoperative evaluation of thyroid nodules and better inform surgical decisions by helping to distinguish patients with benign nodules from those with thyroid cancer, he added.

This study was funded by several grants from the National Institutes of Health. Multiple authors identified relationships with industry.

Cell. 2014;159:676-690. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: