Gene Classifier Identifies Inconclusive Thyroid Nodules

Nancy A. Melville

June 26, 2012

June 25, 2012 (Houston, Texas) — A recently developed gene expression classifier shows promise for distinguishing between benign and malignant thyroid nodules when fine-needle aspiration yields inconclusive cytologic findings, potentially sparing patients unnecessary surgery, according to research presented here at ENDO 2012: The Endocrine Society 94th Annual Meeting.

The findings were also published online June 25 in the New England Journal of Medicine.

As many as 15% to 30% of thyroid nodules are still not clearly identifiable as either benign or malignant after fine-needle aspiration, leaving clinicians no choice but to refer the patient to diagnostic surgery, noted Bryan R. Haugen, MD, professor of medicine and pathology and head of the Division of Endocrinology, Metabolism & Diabetes at the University of Colorado, Denver.

Most nodules turn out to be benign, meaning patients in such cases typically are not only placed at risk for surgical complications but also unnecessarily left requiring levothyroxine replacement for life.

"These data confirm the critical need to improve the preoperative diagnostic evaluation for patients with indeterminate cytologic findings on fine-needle aspiration," the authors of the new study write.

Using a novel diagnostic test capable of measuring the expression of 167 genes (Afirma Gene Expression Classifier, Veracyte), researchers conducted a 19-month, prospective study involving 49 clinical sites, 3789 patients, and 4812 fine-needle aspirates from thyroid nodules that were 1 cm or larger and required evaluation.

There were 577 nodules that were considered cytologically indeterminate; after inclusion criteria were met, 265 with corresponding histopathological specimens from excised lesions were available to be evaluated for the study.

Among the 265 indeterminate nodules, the histopathologies indicated that 85 were malignant; the gene expression classifier was accurate in identifying 78 of the malignant nodules as being suspicious (92% sensitivity [95% confidence interval (CI), 84% - 97%], 52% specificity [95% CI, 44% - 59%]).

The negative predictive values for atypia, or follicular lesion, of undetermined clinical significance were 95%; for a follicular neoplasm or lesion suspicious for a follicular neoplasm, the predictive value was 94%; and for suspicious cytologic findings, it was 85%.

There were 7 false-negatives, of which 1 was a Hurthle-cell carcinoma and 6 were papillary thyroid carcinomas. An evaluation of molecular markers showed that 6 of the 7 aspirates had a paucity of thyroid follicular cells, indicating that the nodule represented an insufficient sample.

The cost of the test is about $3200; however, it is covered by Medicare and some other insurance providers, Dr. Haugen noted.

"We certainly are all aware of cost issues in terms of US expenditure," he acknowledged. "There has been a cost-effective analysis that does show there can be a savings in addition to the lower risk not going to surgery."

Dr. Haugen added that no particular genes have been singled out as definitively predictive of a benign or malignant nodule. "As [another expert] once told me, it's not so much one of the genes being a better predictor than another, it's all of the genes working together in an algorithm."

J. Larry Jameson, MD, PhD, the Robert G. Dunlop Professor of Medicine and executive vice president at the University of Pennsylvania for the Health System, and dean of the Perelman School of Medicine at the University of Pennsylvania in Philadelphia, noted that although the gene expression test will seemingly indeed prevent some unnecessary surgeries, some important caveats should be considered.

"It bears emphasizing that this test does not have sufficient specificity to inform decision making for samples with clear-cut cytologic results (ie, benign or malignant)," said Dr. Jameson.

He added that with approximately 75,000 operations for cytologically indeterminate nodules performed in the United States each year, the gene expression classifier could potentially reduce surgery for nodules with indeterminate cytology by at least one third, or about 25,000 operations per year, resulting in substantial savings.

A concern, however, would be the 5% to 10% of nodules classified as benign (false-negatives) that are likely to be malignant.

"Because this group is at high risk for cancer, it might be reasonable to repeat the fine-needle aspiration biopsy or perform a diagnostic hemithyroidectomy even when the gene expression classifier indicates a benign profile," Dr. Jameson noted.

Leonard Wartofsky, MD, MPH, MACP, past president of the Endocrine Society and the American Thyroid Association, agreed that the false-negative potential represents an important caveat.

"I think the [false-negative] issue touches on the key problem for the clinician: Once you get a negative gene classifier back, what do you do with the patient?" said Dr. Wartofsky, who is chairman of the Department of Medicine at MedStar Washington Hospital Center in Washington, DC.

Patients could still have problems related to a larger nodule, ranging from cosmetic issues to pressure that affects breathing, so knowing the nodule is benign is good, but it does not necessarily avoid the need for surgery, he told Medscape Medical News.

"It could result in less surgery, so perhaps a total thyroidectomy, for instance, could now just be a lumpectomy, if you know it's likely to be benign."

"If there were certain suspicious characteristics on ultrasound that suggested a malignancy in spite of the negative gene classifier, however, I would likely refer my patient for surgery anyway."

Veracyte's main competition is a molecular test made by Asuragen that predicts the opposite — malignancy, with about a 70% to 80% predictive value — so clinicians in fact currently have at least 2 choices for indeterminate nodules.

"The one problem facing clinicians right now is which test do they do: the Veracyte assay that tells them that if it's likely to be benign, or the Asuragen molecular test that tells them a positive marker would be cancer?" Dr. Wartofsky said.

"With each test costing several thousand dollars, using both tests would be a significant burden in terms of healthcare costs, even though it would be the most specific."

Dr. Wartofsky also noted that additional systems are likely on the way.

"I think [the Veracyte test] is just one of what will be an increasing number of approaches to diagnosis by molecular analysis in the coming age of what has been called personalized medicine, where you try to identify a disease by genetic or molecular analysis," he said.

"The Veracyte classifier is, meanwhile, a useful adjunct that will avoid a lot of surgeries and modify a lot of surgeries, likely resulting in fewer complications, but it's not the final answer in the management."

The study received research funding from Veracyte. Dr. Jameson's disclosures include that he is on the advisory board for Quest Diagnostics. Dr. Wartofsky has disclosed no relevant financial relationships.

N Engl J Med. Published online June 25, 2012. Full text

ENDO 2012: The Endocrine Society 94th Annual Meeting. Presented June 25, 2012.


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.