Application of Afirma Gene Sequencing Classifier in assessment of cytologically indeterminate Thyroid Nodules: A Middle Eastern Population Study

Summary: Introduction: Malignancy cannot be ruled out in thyroid nodules classified as Bethesda III or IV (indeterminate). This results in unwarranted surgeries, which are associated with poor quality of life, potential risk of complication, and increased healthcare costs. Afirma Gene Sequencing Classifier (GSC) categorizes indeterminate thyroid nodules into benign/suspicious, thereby optimizing selection of patients for hemi/total thyroidectomy. We aim to assess and report our experience in the performance of Afirma GSC in the assessment of cytologically indeterminate thyroid nodules in the Middle Eastern population. Materials and Methods: After approval from the Ethics community at American Hospital Dubai, retrospective identification of patients who had undergone thyroid FNA and Afirma GSC testing conducted between January 2019 - August 2023. The ultrasound and TI-RADs score were reviewed. For patients with a suspicious result, we retrieved surgical pathology outcomes. The decision for hemi/total thyroidectomy was made by the treating physicians based upon clinical, radiographic and patients' decisions parameters. Afirma results were correlated with sex, FNA category, ultrasound score and surgical pathology findings. Results: A total of 89 cases were included in the study, of which, most were females (76.4%, n=68). Mean age of the participants was 44.2±11.03 years. The mean size of the thyroid nodule was 2.07±1.05cm. Nearly half the cases were classified as TIRAD 4 on ultrasound (49%, N=44). Almost three-quarters of cases were classified as Bethesda III on FNA cytology (73%, n=65). Of the 89 cases, 57% (n=51) of cases were reported as benign on Afirma, while 43% were reported as suspicious (n=38). No statistically significant associations between sex, FNA diagnosis, ultrasound and Afirma were found (p>0.5). Sensitivity was calculated at 9.1%, while specificity was calculated at 100% (p>0.5). Positive predictive and negative predictive values were calculated at 100% and 41.2% respectively (p>0.5). Conclusions: We report high percentages of specificity and positive and negative predictive values. Although we report a low sensitivity (attributed to a smaller number of cases), our study proves that Afirma is indeed a useful tool in the diagnosis of cytologically indeterminate thyroid nodules and will further help stratify patients into surgical and non-surgical therapy groups.