Disclosures: Figge reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
September 08, 2021
2 min read
Save

Molecular testing most accurately predicts thyroid cancer

Disclosures: Figge reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Molecular testing was the best predictor for malignancy in indeterminate thyroid nodules when compared with clinical factors, according to findings from a post hoc analysis published in Thyroid.

In analysis of data from the ThyroSeq v3 clinical validity trial, neither the American Thyroid Association nor the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) scoring systems were associated with cancer or noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP). Molecular testing also more accurately predicted thyroid cancer or NIFTP than age and Bethesda category.

neck thyroid
Source: Adobe Stock

“The results of this study demonstrate that among nodules selected for fine-needle aspiration that yield indeterminate cytology, molecular testing is much more informative than all other examined parameters, which do not contribute to improved cancer/NIFTP risk stratification in such nodules,” James J. Figge, MD, MBA, an endocrinologist with St. Peter’s Health Partners and Trinity Health in Rensselaer, New York, and colleagues wrote. “The only possible exception to this is male sex, which showed a trend towards association with thyroid cancer/NIFTP risk in addition to molecular testing results.”

The ThyroSeq v3 clinical validity trial was an analysis of 257 thyroid nodules from adults aged 18 years or older with fine-needle aspiration results yielding an indeterminate cytology at 10 study centers. The ThyroSeq v3 genomic classifier was used to conduct molecular testing of all nodules. For the post hoc analysis, researchers obtained age, sex, nodule size by ultrasound and Bethesda category from the original trial. Data on history of radiation exposure, family history of thyroid cancer and ultrasound pattern were obtained post hoc. Ultrasound images were reviewed at each study center to assign ATA and TI-RADS risk categories for each nodule. Researchers used univariate logistic regression to identify covariates most strongly associated with thyroid cancer. These covariates were further analyzed in multivariate logistic regression.

There were 232 participants with 257 thyroid nodules included in the analysis (80% women; median age, 53 years). Sixty percent of nodules were classified as Bethesda III, 36% were categorized as Bethesda IV and 4% were placed into the Bethesda V category.

In univariate logistic regression, sex, age, Bethesda category and ThyroSeq v3 score were associated with thyroid cancer and/or NIFTP probability (P < .05 for all). ThyroSeq v3 result was the strongest predictor of all the parameters.

“We observed that in indeterminate cytology nodules, the ATA and TI-RADS ultrasound patterns, although demonstrating modest positive trends, were not further associated with malignancy/NIFTP independently of molecular testing,” the researchers wrote. “Whether these trends would become statistically significant with a larger sample size would be an important question for further study.”

In a multivariate logistic regression model, researchers examined cancer risk while incorporating sex, age, Bethesda category and ATA ultrasound pattern. The model had a C-index of 0.653 (R2 = 0.108). Adding the ThyroSeq v3 results to the model increased the C-index to 0.888 (R2 = 0.572). After ThyroSeq v3 was added, age, Bethesda category and ATA ultrasound pattern became nonsignificant predictors within the model.

A multivariate model with only sex and ThyroSeq v3 result included as covariates had a C-index of 0.889 (R2 = 0.588). ThyroSeq v3 had the greatest influence on thyroid cancer predictability in the model (P < .0001), with sex having a weak contribution.

The researchers wrote that future studies are needed to investigate the interaction of specific types of molecular alterations and to validate the proposed clinical prediction nomograms against independent data in clinical studies.