Currently, patients who have thyroid nodules with indeterminate cytology findings from fine-needle aspiration (FNA) biopsy typically undergo thyroidectomy. While this has lowered the rate of mortality in patients developing thyroid cancer, only 20% to 30% of these indeterminate nodules are cancerous,1 exposing many healthy patients to unnecessary surgical procedures.
By employing several platforms for molecular testing of thyroid nodules, including using a combination of next generation sequencing (NGS) and messenger RNA (mRNA) expression profiling, the possibility of reducing the number of unnecessary surgeries may become standard practice. Clinicians can look to advanced mutational analysis to more accurately discriminate between benign and malignant indeterminate thyroid nodules. However, neither mutational testing nor mRNA expression profiling can be relied on alone to classify all indeterminate nodules, in part, because not all genetic alterations in thyroid cancer are currently known.
Following approval of ThyGenX (Interpace Diagnostics, Parsippany, NJ), a next generation oncogene sequencing panel to identify mutations in 8 genes associated with thyroid carcinomas, by the New York State Department of Health, the company has initiated a multicenter study throughout New York to provide further validation of the improved risk stratification to improve patient referral for surgery with dual ThyGenX/ThyramiR testing.
Now, the accuracy of molecular testing may be improved when different testing platforms are combined, such as concurrent assessment of results from both mutational testing and microRNA (miRNA) profiling.2,3 These tests use known genetic mutations to confirm nodules that are cancerous and miRNA profiles to verify those that are benign. Such coordinated testing is becoming more widely available given approval of these technologies by state regulatory agencies and acceptance of coverage by Medicare and private insurance companies.
“Molecular profiling that combines a rule-out test and rule-in test using one biopsy sample can improve rates of diagnosis and appropriate therapy based on whether this is a high likelihood of cancer or not,” Alexander Shifrin, MD, surgical director of the Center for Thyroid, Parathyroid and Adrenal Diseases at Jersey Shore University Medical Center, Neptune, New Jersey, conducted research verifying the value of the dual testing,3 told EndocrineWeb. “This actually results in fewer surgeries for diagnostic purposes in patients with benign disease and more surgeries for patients with specific diagnoses.”
About half of all people will develop a thyroid nodule by age 60, and despite improved riskbased classification of thyroid nodules using FNA biopsy, up to 20% of cases cannot be classified as benign or malignant based on cytological examination alone.1 Several different molecular testing technologies have recently been developed to overcome the limitations of FNA biopsy without resorting to surgical resection and the associated risks this procedure presents in cases of benign nodules.
Qualitative mutational testing can identify gene mutations and fusion genes, which are known to be associated with thyroid cancer at a high positive predictive value (PPV) that nodules with these mutations are highly likely to be cancerous. Since not all cancer-causing mutations are known, however, such tests provide little information for thyroid nodules without known mutations. Molecular testing methods with a high negative predictive value (NPV) are required to definitively exclude thyroid nodules from a cancer diagnosis.
Expression profiling of miRNAs offers one way to test thyroid nodules with a high NPV. This is valuable since miRNAs, which are small, highly conserved, non-coding RNA molecules, are highly dysregulated in a variety of cancers.2 Furthermore, miRNAs regulate key biological pathways involved in the development of thyroid cancer, including cell cycle progression, differentiation, proliferation, and survival. Therefore, profiling the expression levels of cancer-associated miRNAs in biopsies can provide additional information regarding the diagnosis of indeterminate nodules. Such testing can be used as a complement to mutational testing in order to identify cancerous nodules that lack known oncogenic mutations.
In examining the predictive value of so-called “multiplatform” testing of indeterminate modules that included both mutational and miRNA analysis to discriminate between benign and malignant cases.3 This combined molecular testing technique had sensitivity of 89% and specificity of 85%.3 This increased identification of true benign samples over RNA expression testing and significantly decreased the rate of unnecessary surgeries.
Combined molecular testing for thyroid nodules, although promising, can be improved even more to increase the sensitivity and specificity of these tests, according to Dr. Shifrin. “There may be some mutations that have not been discovered that could be added to these analyses to improve specificity and sensitivity,” he told EndocrineWeb.
Joshua D. Safer, MD, an associate professor of medicine and molecular medicine at Boston University School of Medicine in Boston, Massachusetts, and a member of the EndocrineWeb editorial board commented on the need for further validation of many commercially available molecular testing platforms.
“The molecular methods to date have not been well validated by independent researchers,” he said. “Good external validation will help separate the more successful among these modalities and provide us guidance regarding which will have the best cost to benefit ratio.”