Some patients with intermediate thyroid FNA cytology wil have a chance of cancer:

• AUS/FLUS – Thy3 (predicted probability of cancer 5-15%).
• FN/SFN – Thy3 (predicted probability of cancer 15-30%).
• SMC – Thy4 (predicted probability of cancer 60-75%)

The current BTA/ATA guidelines suggest a diagnostic lobectomy for a biopsy desingated 'Thy3' with completion of thyroidectomy if histology is positive for cancer as 10–40% will prove to have malignant potential. A thyroidectomy is recommended for Thy4 lesions.

This study aims to answer whether an FNA can distinguish between a benign nodule and cancer preoperatively, to allow planning of definitive surgical management. The revised American Thyroid Association’s guidelines suggest that units consider a 'mutational panel' for nodules with indeterminate FNA cytology to help guide clinical management.

• From April 2007 to April 2009, 1056 consecutive residual FNA samples from 762 patients with indeterminate cytology diagnoses were prospectively tested for mutations.
• Cytological diagnosis was established before molecular testing.
• A retrospective analysis was performed with correlation between the results of cytological evaluation, histological diagnosis and molecular testing

Results:

• 967 FNA samples adequate for mutational analysis.
• 479 thyroidectomies provided a histopathological diagnosis for 513 FNA samples.
• Total of 85 mutations were detected.
• Molecular testing has a PPV of 87–95% for predicting thyroid cancer.
• RAS mutation (85% risk of malignancy).
• BRAF V600E, RET/ PTC, PAX8/PPAR mutations (associated with malignancy in close to 100% of nodules)s

Imperial Centre for Endocrinolgoy felt that this was:

• Easy, reproducible, does not prolong medical procedures.
• Definitive surgical care in the first surgical procedure: 121/479 patients in this study, ie 25.3%.
• The group does not actually propose no surgery vs total (definitive) thyroidectomy, but provides an algorithm to save patients a repeat surgical procedure vs those that will only undergo diagnostic lobectomy.
• For this study, each FNA sample was considered independently. In practice patients with indeterminate cytology have repeated FNA, which may yield different result.
• Histopathologist not blinded to the results of molecular testing.
• Paper does not make clear how many patients had diagnostic vs total vs cpmpletion of thyroidectomy and implies that all 479 operations were lobectomies.
• 51 patients actually had SMC (Thy4) cytology, which should have been total thyroidectomy according both to BTA and ATA guidelines, but seem to have had diagnostic lobectomy in this cohort – why?
• Negative predictive value of molecular testing: In this paper FN/SFN NPV 14%, SMC NPV 28%, best reported has been 6%. However, for these markers to change practice, ie avoid surgery overall, the NPV needs to be almost 0, since only then can you say to a patient with indeterminate/Thy3 cytology that they do not require surgery overall. Most patients would not accept a conservative approach with a 6% risk that they have a cancer.