Background: Although the current gold standard for diagnosing thyroid nodule malignancy is ultrasound-guided fine-needle aspiration (FNA) cytology, about 20–25% of cytological evaluations are considered indeterminate for malignancy. This limitation has led to the emergence of next-generation sequencing panels, for example, ThyroSeq v3 (TSv3), which recognize highly diagnostic genetic mutations of common thyroid carcinomas in FNA samples and classify them as test-negative or test-positive, helping optimize treatment for indeterminate thyroid nodules (ITNs). Our goals were to evaluate the benign call rate (BCR) of TSv3 and assess its diagnostic performance and clinical utility while highlighting the points of consideration for a public Canadian institution. Methods: This is a single-center study conducted at the Royal Victoria Hospital (McGill University Health Centre) in Montreal, Canada, between January and February 2019. Patients were offered TSv3 following the McGill algorithm for ITN workup, a novel protocol developed at our institution to select only diagnostic surgery candidates to minimize waste of public resources, considering the single-payer health care system. Patient demographics, cytopathology results, TSv3 data, treatment plan, and final histopathology result were reviewed. Results: A total of 50 ITNs underwent TSv3 testing; molecular analysis yielded 20 (40%) “positive” results and 24 (48%) “negative” results. Six (12%) results were classified as “currently negative” or “negative but limited.” “Currently negative” results indicate a low-risk mutation that alone is insufficient for development of a malignant lesion. “Negative but limited” results indicate a sample that is nondiagnostic for malignancy due to low cell count. BCR was calculated as (“negative” and “currently negative”)/total, resulting in a BCR of 58%. Twenty-three (46%) patients were scheduled for surgery and 27 (54%) patients continued with surveillance. Ninety-one percent (20 of 22) of the resected target nodules were malignant on final pathology. Conclusions: TSv3 proved beneficial in classifying ITNs as positive or negative, avoiding surgery in the latter cases. We found a lower reduction rate in surgery and BCR than the previously published studies, which is attributable to the criteria of the McGill algorithm. In the Canadian public health care system, preventing unnecessary surgery represents significant cost savings for the provincial government while also improving patient quality of life.

Background: Molecular testing of thyroid nodules is a diagnostic tool used to better understand the nature of thyroid nodules. The aim of this study is to better comprehend the relationship between specific mutations and aggressive behavior of the tumour as demonstrated on postoperative pathological analysis. Methods: A retrospective chart review of 103 cases was performed. Included were patients who had undergone molecular testing using a panel that tests for 9 mutations (ThyGenX®) and were found to have malignant tumours. The following gene alterations were found pre-operatively in the nodules: BRAF V600E ( n = 32), BRAF K601E ( n = 4), NRAS ( n = 11), HRAS ( n = 4), KRAS ( n = 3), RET/PTC1 rearrangement ( n = 1), TERT promoter ( n = 2), PAX8-PPARγ rearrangement ( n = 1), and 45 cases where no mutation was detected. Aggressive behavior was defined by extra-thyroidal extension (ETE), lymph node metastasis (LN+), and the following variants of papillary thyroid carcinoma: tall cell, solid, diffuse sclerosing, columnar cell and hobnail. Chi-squared testing was performed to compare groups. Results: The group with BRAF V600E, RET/PTC1 rearrangement, and TERT promoter mutations was associated with ETE 37.1%, and LN+ 45.7% of the time compared to 4.3 and 13.0% in the group with other mutations, and 4.4 and 4.4% in the group with no mutations ( p-value 0.02, p-value < 0.001, p-value 0.006). In addition, the BRAF V600E, RET/PTC1 rearrangement, and TERT mutations group demonstrated tall cell variants (17.1%), columnar cell variants (5.7%), and hobnail variants (3%). The other mutations group demonstrated columnar cell variants (4.3%), and the no mutations group demonstrated solid variants (2.2%). Conclusions: In this study, BRAF V600E, RET/PTC1 rearrangement, and TERT mutations were associated with aggressive behaving thyroid malignancies as defined above. Molecular testing may be a useful method to anticipate aggressive tumour types and therefore assist in planning the extent and timing of surgery.

Background: Medullary thyroid cancer (MTC) cells are capable of secreting various tumor markers including calcitonin and carcinoembyronic antigen (CEA). The purpose of this study is to determine whether abnormal CEA levels may be used as a tumor marker to predict the severity of disease in MTC. Methods: A retrospective analysis was completed for 33 patients with MTC who had preoperative serum CEA levels. Univariate and multivariate analyses were used to quantify the relationship between serum CEA levels and tumor stage and prognosis. Results: On multivariate analysis, elevated preoperative CEA levels were significantly associated with the size and stage of tumor, distant metastasis, decreased biochemical cure, and mortality. There was a significant association between tumor size greater than 37 mm and elevated CEA levels (> 271 ng/ml). There was also a positive correlation with increased cancer stage (> 377 ng/ml), distant metastasis (> 405 ng/ml), and contralateral compartment location of lymph node metastasis (> 162 ng/ml). When pre-operative CEA levels are > 500 ng/ml, patient mortality was 67%. Conclusion: In this study, both pre-operative calcitonin and CEA levels were significantly correlated with the extent of disease in MTC. While calcitonin has a linear relationship with disease progression, abnormal CEA levels were a better indicator of advanced disease. CEA levels > 271 ng/ml are significant for advanced tumor size and staging, metastasis to the central compartment, and decreased chance of biochemical cure. CEA levels greater than 500 ng/ml are associated with significant patient mortality.