Genomics May Inform Malignancy Risk of Indeterminate Thyroid Nodules

Relying on molecular "rule-in" testing of thyroid nodules to better assess for cancer risk improves predictive value, based on a systematic review of the literature.

With Whitney S. Goldner, MD, and Susan Mandel, MD, MPH

Management of thyroid nodules in which the cytology is neither clearly benign nor malignant continually challenges clinicians. Now, based on results of a systematic review of studies examining molecular variants and their risks for malignancy have found some genetic patterns are indicative of a higher positive predictive value (PPV) than others.1  

The findings, published in Thyroid, will lead to better cancer prediction and inform clinicians regarding the independent contribution of a genomic profile in assessing the recommended course of treatment for patients who present with uncertain thyroid nodules, say the authors.

Certain genetic codes can tell clinician that the thyroid node is likely to be cancerous.

Ascertaining Risk of Cancer Boon for Patients with Uncertain Thyroid Nodules 

Historically, gene panels have been used to assess predisposition to hereditary cancers by testing multiple susceptibility genes and/or variants. In recent years, genetic panels have been used as part of solid tumor malignancy testing to assess somatic alterations. One example, the use of targeted variant panels for thyroid nodules that are not conclusively malignant or benign.

The initial studies included single genomic alterations, followed by panels of several genes, then numerous alterations among many genes.2

In this systematic review, "the individual gene alternations had different levels of thyroid cancer risk," said lead author Whitney S. Goldner, MD, professor of medicine in the Division of Diabetes, Endocrinology, and Metabolism at the University of Nebraska Medical Center and Nebraska Medicine in Lincoln. "So, they really should be considered individually, not as part of a big panel."

In selecting appropriate studies, the researchers searched the medical literature from 2008 through 2018 for research that provided genetic data from preoperative fine-needle aspiration (FNA) specimens on cytologically indeterminate thyroid nodules; all patients subsequently underwent surgical resection.

Dr. Goldner's team found that one alteration, BRAFV600E, produced the highest predictive value, at 98%, while the remaining four alterations reported were at least 10 times, had lower PPV, ranging from 38% to 55%.1

The new findings will be ''not so much a surprise as clarity" for clinicians, Dr. Goldner said.

As we know, if a nodule is not clearly benign or malignant, falling in that indeterminate range, the risk of malignancy is the crucial question in deciding the right course of action for the patient. "That is where additional molecular analysis has been used," Dr. Goldner told EndocrineWeb.

"What we did was actually look at previously published studies to combine all the individual mutation analyses. Rather than looking at the panel, we looked at patients’ DNA variations and RNA fusions that were reported," she said. This approach allowed the researchers to zero in on mutations or alterations suggesting the greatest risk for malignancy.

The researchers found 540 studies in their search and supplemented their data with 18 others from bibliographies or personal files. Of those, 61 studies met the researchers' criteria, which reflected more than 4,600 indeterminate thyroid nodules.1

"About 26% of the nodules showed some sort of genetic alterations, of which the ones that had BRAF v600E demonstrated the highest rate of cancer, at 98%,'' she said.

Among the review limitations, the authors noted that while the nearly one in four nodules showed positive for at least one gene variant and/or fusion, the sequence-changing variants made up the majority of the genetic matter (94%) of the aberrations found. But about 50% of these cases only had information on genes involved.1

The investigators did not have information on the specific amino acid or resultant protein changes. For that reason, only a gene-level PPV could be calculated. According to Dr. Golden, this was crucial because they believe variants in the same gene are associated with different PPVs.

Gene Variability in Individual Thyroid Nodules May Offer Positive Predictive Value

The researchers evaluated the positive predictive value of all genomic alterations that were reported at least 10 times; however, only five genetic alterations reached this threshold.1

These genetic variants included:

  • BRAF V600Ewith a PPV of 98% (CI 96-99%)
  • PAX 8/PPARG, 55% (CI, 34-78%)
  • HRASQ61R, 45%, (CI 22-72%)
  • BRAFK601E ,42%,  (CI 19-68%)
  • NRAS Q61R, 38% (CI 23-55%).

When the researchers excluded the most common BRAF V600E, they found the pooled PPV for all the other molecular variations and fusions fell to 47%. Multiple variants existing within the same nodule were found in about 1% of nodules with a cumulative PPV of 77%.1

The authors concluded: the prospect that a genomic variation is sufficient to predict cancer depends on the specific gene variant or fusion.1

There were a few limitations to their review that offers caution regarding the potential use of this approach, at least for now, the researchers acknowledged. Most obvious was the small sample size of most variants and fusion pairs identified in the indeterminate nodules, which definitely limits the confidence in individual PPV estimates.

We need to study gene variants consistently and in more detail, say the researchers. "The individual gene alterations had different levels of thyroid cancer risk, so really need to be considered individually," Dr. Goldner said.

"As for the clinician, probably the take home message is—look at the individual variant or fusion in each patient to help predict the risk of malignancy," she said.

Genetic Changes Can Be Useful in Diagnosing Indeterminant Thyroid Tumors

"This study represents excellent work to comprehensively summarize the association of different single and combinations of genetic mutations or fusions with thyroid cancer while recognizing the study limitations," said Susan Mandel, MD, MPH, professor of medicine and radiology at the Perelman School of Medicine at the University of Pennsylvania and director of clinical endocrinology and diabetes at University of Pennsylvania Hospital System in Philadelphia.

"As molecular testing becomes more widely adopted for thyroid nodule diagnosis, we need to better understand the methodologies of the mutational analysis and the predictive value of genetic alterations," she told EndocrineWeb; she was not involved in the study and so agreed to offer her perspective.  

"For the clinician, paying attention to the specific sequence variant or the fusion and taking that individual risk of malignancy into account when trying to decide how to interpret that finding" is important, Dr. Goldner said.

Indeterminate thyroid nodules present a tough clinical challenge, she said. "We know they don’t look benign but yet are not clearly malignant." So the hardest question is what to do, if anything since suggesting that the patient face surgery is not without risks. There is good reason to avoid surgery given the chance that it may reveal that the nodule is benign, making the procedure unnecessary and requiring that the patient take long-term thyroid hormones, which brings with it a whole set of other health risks. ''These [avoidable treatments] are significant hurdles that people want to avoid."

She added that this challenge is only getting greater given how frequent thyroid nodules are being diagnosed.3 "By age 50 or 60 years, about half of people will have thyroid nodules," Dr. Goldner said. Many of these nodules are detected incidentally, often when a patient has had a scan of the neck or chest area when being checked for shortness of breath or other medical evaluations.

In the paper, Dr. Goldner said that they raised the point that "American Thyroid Association guidelines identify BRAFV600E, RET/PTC, and PAX8/PPARG as having high enough PPV (> 95%) to be considered 'rule-in' tests."4

However, their review shows a similarly high PPV for BRAFV600E and the combined RET/PTC fusions (at 100%) but much lower PPVs for PAX8/PPARG—at 55%.1 So, while the risks of malignancy are increased with the other variants, the Dr. Goldner suggests that the risks were not considered high enough to meet the threshold of the rule-in tests.1

While Dr. Goldner has no financial conflicts other coauthors are employees or consultant for Veracyte, Inc., which makes a molecular test for indeterminate thyroid nodules. No funding was received for the report. Dr. Mandel has no relevant disclosures.

Continue Reading:
Preferred Surgical Approach for Indeterminate Thyroid Nodules
close X