Malignancy rates and initial management of Thy3 thyroid nodules in a district general hospital: The ‘Reading’ experience

Abstract Background Ultrasound‐guided fine‐needle aspiration cytology is the gold standard for investigating thyroid nodules. Stratifying the Thy3 thyroid nodule risk of malignancy is essential for clinical decision‐making. According to the Royal College of Pathologists Guidance (2016), the rate of malignancy for Thy3a is 5–15% and for Thy3f 15–30%. Our aim was to investigate the malignancy rate and the initial management of Thy3 nodules in our institution. Methods A retrospective review was undertaken of 115 patients with Thy3 cytology results from thyroid fine‐needle aspirations performed between January 2015 and June 2020 at a single centre. A total of 90 out of 115 patients underwent surgery. Results Of the 90 patients, we had a 40% malignant rate (36/90). Specifically, 14 of 34 (41.1%) Thy3a and 22 of 56 (39.2%) Thy3f nodules were malignant. Of the malignant lesions, 52.7% (19/36) were follicular thyroid carcinoma. 58.8% (10/17) of male patients and 35.6% (26/73) of female patients had a malignant histology. Eighteen patients eventually needed a completion thyroidectomy. Conclusion Compared with national data, we showed a higher risk of malignancy in Thy3 nodules in our centre. Our study should encourage other centres to audit their own data. We propose setting up a national Thy3 registry as a basis to promote research in improving preoperative diagnosis of indeterminate thyroid nodules.


| INTRODUC TI ON
Thyroid cancer is the most common malignant endocrine tumour but constitutes less than 1% of all malignancies treated in the United Kingdom. 1 Although the incidence of thyroid cancer is increasing globally, the overall mortality from thyroid cancer has remained stable over many years. 2,3 Thyroid nodules are commonly seen in both, primary and secondary care. Diagnosis of benign versus malignant nodules can be challenging. Ultrasound plus or minus fine-needle aspiration (FNA) is the gold standard used for initial diagnosis of all thyroid nodules.
In the UK, thyroid nodules are assessed radiologically using the U or TI-RADS (Thyroid Imaging Reporting and Data System) classification to determine which nodules are benign and which are suspicious requiring a FNA. Despite this, there is still a proportion of patients with indeterminate cytology (Thy3). Quoted malignancy rates in the UK for Thy3a are 5-15% and 15-30% for Thy3f. 3 Thy3 cytology is indeterminate since follicular neoplasms require formal excision to assess capsular and/or vascular invasion within the tissue architecture to determine malignancy. 3,4 According to the statistical data resulting from the British Association of Endocrine and Thyroid Surgeons (BAETS) audit in 2012, over 50% of all thyroid surgery carried out in the UK was for benign disease. 2 British Thyroid Association (BTA) guidelines published in 2014 for managing Thy3 nodules suggested a repeat investigation by further ultrasound assessment with or without a repeat FNA for Thy3a nodules and diagnostic hemithyroidectomy for Thy3f lesions. 3 It can be challenging for patients to understand the ambiguity surrounding Thy3 nodules, the need for a diagnostic hemithyroidectomy, and the possibility of further treatment with surgery and radioactive iodine treatment. Therefore, preoperative counselling is key to help patients understand the risks and benefits of diagnostic thyroid surgery.
The aim of this study was to establish the risk of thyroid cancer for Thy3 nodules over five years (2015 to 2020) in our institution.
The secondary aim was to compare our initial management of all Thy3 nodules against the current BTA guidelines.

| MATERIAL S AND ME THODS
This study is a retrospective observational review of all thyroid nodules with a Thy3 cytology result investigated by ENT and endocrine departments between January 2015 and June 2020. It was conducted in a district general hospital by obtaining information from cytological reports in the pathology database. The study was registered with the clinical audit department and approved in the departmental clinical governance meeting. Since it was a retrospective audit of cytological performance with no intervention, ethical committee approval was not required.
All Thy3 nodules were discussed in a regional thyroid multidisciplinary team (MDT) meeting (Oxford University Hospitals) comprising of an oncologist, a histopathologist with specialist interest in thyroid pathology and a thyroid surgeon. Only Thy3 nodules that had a preoperative thyroid ultrasound and surgical histology were included in the study. Exclusion criteria included paucity of information on the electronic patient record (EPR) with regard to ultrasound, fine-needle aspiration cytology or histopathology. The MDT recommendation, treatment options and risk of malignancy were discussed with each patient. This allowed the patient to make an informed choice for further management. In our practice, surgery was also offered to patients whose nodule cytology had been downgraded from Thy3 to Thy2 but measured 40 mm or greater. Additional operative criteria included compressive multinodular goitres, nodule growth on previous surveillance ultrasound imagings and/or patient preference.
During the study period, we obtained 115 patient records with The data set was then analysed by going through the thyroid MDT discussion on EPR to determine which results were downgraded or upgraded. All statistical analysis was performed using Microsoft Excel. Demographic characteristics were recorded for each group. The final histopathological results for each group were compared with the initial FNA results to determine the 'rate of malignancy'. We acknowledge that our results would have an incomplete data bias due to 10 patients being excluded.

| RE SULTS
In this study, 115 patient records were analysed. A total of 55 (47.8%) were Thy3a and 60 (52.2%) were Thy3f (Figure 1). Each cytology result was discussed in the regional thyroid MDT. Of the 55 Thy3a results, 11 (20%) were downgraded to Thy2, two (3.6%) were upgraded to Thy4 cytology, and 34 (61.8%) patients underwent surgery. Similarly, of the 60 Thy3f results, two (3.3%) were downgraded to Thy2 whereas 56 (93.3%) went ahead with surgery. A total of 10 patients had missing data (eight Thy3a and two Thy3f) and these were excluded. Overall, 13 patients were downgraded to Thy2 at MDT. Of these, 10 were discharged with no follow-up as they previously had ultrasound surveillance of the nodule without evidence of any growth and three patients opted to proceed with surgery (patient concern). Table 1a and Table 1b shows the breakdown of ultrasound gradings for the Thy3 nodules in our study. 58.8% (10/17) of male patients and 35.6% (26/73) of female patients had a malignant histology ( Table 2). The average age (in years) of patients with a malignant histology was 49 and 56.2 in the Thy3a and Thy3f groups, respectively (Table 2). Overall rate of malignancy in our study for those who underwent surgery for Thy3 lesions was 40% (36/90). The rate of malignancy in the Thy3a patients was 41.1% and 39.2% in Thy3f patients ( Figure 1). All the three Thy3a patients who were downgraded at thyroid MDT but chose to undergo surgery had benign histology. The ratio of follicular thyroid carcinoma to papillary thyroid carcinoma in the Thy3f group was 11:10 compared to 8:6 in the Thy3a group. The histological outcomes for both groups have been summarized in Table 3. In our study, three patients had an incidental papillary microcarcinoma on the background of a benign goitre. 18 of the 36 patients (50%) who had a malignant histology after primary surgery underwent completion surgery (Table 4).

| DISCUSS ION
The main objective of this study was to analyse the malignancy rates in Thy3 nodules over a five year period in our institution.
The most recent guidelines on the management of thyroid cancer were issued by the British Thyroid Association in 2014. 5  Textural Analysis on DW-MRI, 13 Real Time Elastography 14 and Conventional US + Elastography. 15 Of these Afirma GEC showed highest sensitivity (95.5%) 11 while Textural Analysis on DW-MRI had highest specificity (96%). 16 The use of molecular biomarkers has been included in the most recent iteration of the American Thyroid Association management guidelines. 10

| Study limitations
Since data collection was retrospective, only those patients with histological outcome were included. Other limitations included a small cohort of 90 patients. However, most of the cytology specimens were reported or supervised by a pathologist with specialist thyroid interest, and hence, the inter-observer variability was low.

ACK N OWLED G EM ENTS
The authors would like to thank Amanda Babington, Dr. Asha Rupani and Dr. Sukhvinder Ghataura from the department of Pathology, Royal Berkshire Foundation Trust Hospital for their contribution to data acquisition.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request from the authors.