BACKGROUND
Ultrasound is frequently used to evaluate thyroid nodules. While the appearance of the nodule on ultrasound is used to determine whether nodules should be biopsied, the decision regarding surgery or monitoring is usually made on the basis of subsequent biopsy results with molecular testing when indicated. Currently guidelines recommend using molecular testing if biopsy results are indeterminate. The 2 major commercially available molecular tests are Veracyte Afirma Genomic Sequencing Classifier (GSC) and CBLPath ThyroSeq v3. Patients with benign results on molecular testing can be monitored, while patients with abnormal results will usually undergo surgery. However, molecular testing is expensive and the authors wanted to determine if adding the information on ultrasound characteristics would affect the utility of molecular testing.

THE FULL ARTICLE TITLE
Hu TX et al 2022 The Effect modification of ultrasound risk classification on molecular testing in predicting the risk of malignancy in cytologically indeterminate thyroid nodules. Thyroid. Epub 2022 May 25. PMID: 35611970.

SUMMARY OF THE STUDY
The study group included all patients (343) with indeterminate thyroid nodule cytology (Bethesda classes III and IV) at a major medical center from July 2017 to April 2020. There were 375 indeterminate nodules that were randomly assigned to molecular testing with either Afirma GSC (RNA test) or ThyroSeq v3 (DNA-RNA test). The original study was performed to determine whether either test was superior to the other and found them to be similar. Since the pathologists were blinded to the results of ultrasound when reading the cytology, this group of patients could also be used to study whether there was a difference in utility of molecular testing results depending upon the ultrasound findings.

Of the 375 nodules studied, 162 were found to be suspicious by molecular testing and 82% (133) ended Of the 375 nodules studied, 162 were found to be suspicious by molecular testing and 82% (133) ended up going to surgery. There was a 63.9% rate of cancer or NIFTP based on the pathology of the nodules removed by surgery. The 29 nodules not removed were either monitored with surveillance per patient preference (20), had surgery deferred due to other illnesses (2), were lost to follow up (6), or had surgery with no histologic diagnosis (1). Only 32 of 245 nodules found not to be suspicious on molecular testing were resected and 7 were found to be cancers (2.9%).

ATA ultrasound risk classification was then used to stratify the nodules into benign (<1% risk of cancer), very low suspicion (<3% risk of cancer), low suspicion (5–10% risk of cancer), intermediate suspicion (10–20% risk of cancer), and high suspicion (70–90% risk of cancer) with 1,13,157,172 and 32 nodules respectively. The rate of cancer in ATA low/intermediate suspicious nodules who went to surgery (343) with positive molecular testing was 56% and 67% for Afirma and ThyroSeq respectively and was 3.8% and 2.1% respectively when molecular testing was negative. Of the 19 patients with ATA high suspicion who underwent surgery, all had positive molecular markers except for one patient who was Afirma benign, confirmed benign at surgery. Among the other 18 with positive molecular testing, 80-88% were found to have cancer on pathology depending upon the molecular test used.

WHAT ARE THE IMPLICATIONS OF THIS STUDY?
The finding of very low cancer rates in patients with low/ intermediate ATA (ultrasound) nodules, indeterminate cytology and negative molecular testing confirms that diagnostic surgery may be avoided. However, in patients with highly suspicious ultrasound findings, the addition of molecular testing adds little to the outcome and avoiding this extra cost may be warranted. We should realize that this study is from a single center with experienced ultrasonologists and cytologists and may not be expandable to a larger, more diverse population.

— Marjorie Safran, MD

Table of Contents | PDF File for Saving and Printing