With great interest we read the article by Kleiman et al that addressed the putative added value of preoperative BRAF(V600E) analysis to the initial surgical strategy. Because the basic surgical strategy for indeterminate thyroid nodules differs between many US centers and those in Europe, we propose that the data from Kleiman et al might lead to an opposite conclusion regarding the use of preoperative BRAF testing in the European setting. To substantiate this hypothesis, we have used data from the Netherlands Cancer Registry.
As pointed out by Kleiman et al, the standard surgical approach to indeterminate thyroid nodules in the United States is to perform a total thyroidectomy on all patients with Bethesda category V nodules and on all nodules with “worrisome cytologic features” (eg, nuclear grooves and pseudoinclusions). Because 12 of the 13 patients with BRAF mutations had already undergone total thyroidectomy as the initial procedure based on their cytology findings, routine preoperative BRAF testing would have altered the surgery in only 1 patient.
In contrast to the situation in many US centers, in the Netherlands, as well as in large parts of Europe, a diagnostic hemithyroidectomy is performed routinely on all patients with indeterminate fine-needle aspiration results. If the final histology reveals malignancy, a completion thyroidectomy is performed as second-stage surgery.
In the Netherlands, 459 patients were diagnosed with differentiated thyroid cancer in 2011. In the flowchart in their study (Fig. 1), Kleiman et al did not differentiate between Bethesda category III/IV and Bethesda category V nodules. Nonetheless, we could extract this from the text and tables. In short, 36 of the indeterminate results (12%) were Bethesda category V; of those 36 results, 26 (72%) were malignant, and of those 26, 11 (31%) were BRAF positive. When we apply these results to our nationwide data, using the same distribution of relative ratios as found by Kleiman et al, we would have had 42 patients with Bethesda category V nodules. Of those 42 patients, 30 (71%) would have had a malignancy, 13 of which (31%) would comprise a BRAF(V600E) mutation. Hence, by performing preoperative BRAF(V600E) analysis in the group of patients with Bethesda category V nodules, these 13 patients of the 42 patients examined (31%) would have benefited because their initial surgical treatment would have been changed to total thyroidectomy, and a 2-stage procedure could have been avoided. This would involve a significant benefit, including a shortened period of uncertainty for the patient, the avoidance of a second surgery, and a significant reduction in the time between diagnosis and final treatment. Of course, we realize that our analysis is not statistically flawless. For practical reasons, we have relied on the assumption that cytology scores in our country are not significantly different from those in the United States. This assumption is supported by 2 other studies describing comparable ratios of indeterminate cytology results and BRAF(V600E) positivity.[2, 3]
Subsequently, we performed a rough estimate as to whether the BRAF(V600E) mutation analysis could be cost-effective. If only Bethesda category V nodules are tested, 42 BRAF(V600E) tests have to be performed to detect 13 BRAF(V600E) mutations. A calculation based on the costs of hemithyroidectomy and total thyroidectomy in our institution demonstrated that this would not lead to a significant increase in costs.
Based on the notion that the Dutch approach is comparable to the strategy used in large parts of Europe, the introduction of preoperative BRAF testing on Bethesda category V nodules could make a difference. A considerable number of patients might be spared unnecessary 2-stage surgery without increasing the total costs of the treatment. We therefore propose that the impact of preoperative BRAF(V600E) testing of indeterminate thyroid nodules on initial surgical management is predominantly dependent on the routine initial surgical strategy that is adhered to.