Why Did the Rapid Increase of Reflex Molecular Testing of Indeterminate Fine Needle Aspiration Cytologies in the USA Not Impact Thyroidectomy Rates? What Is the Lesson to Be Learned?

In this issue of Thyroid, Huang et al. report their analysis of the relationship between reflex use of molecular tests (MTs) for indeterminate thyroid nodule cytologies and thyroidectomy rates. Because of lack of cytology results, MTs and thyroidectomy rates were evaluated as a percentage of all fine needle aspirations (FNAs).

The authors used retrospective investigation of health care claims information with interrupted time series analysis to compare the preintervention trend of thyroidectomy rates with outcome changes and postintervention trends in relation to the publication of the validation studies for molecular FNA tests, approval of reimbursement codes, and the publication of the 2015 American Thyroid Association (ATA) thyroid cancer and thyroid nodule guideless that recommended de-escalations for several aspects of thyroid cancer diagnosis and treatment. ^1,2 Interestingly, this study by Huang et al. reports an unchanged stable decline of thyroidectomy rates post-2015 ATA guidelines and thus a lack of post-2015 deceleration of thyroidectomy rates that contrasts with acceleration of MTs post-2015.

This lack of impact of the rapid increase of MTs of indeterminate FNA cytologies in the United States on thyroidectomy rates is not unexpected and consistent with several previously reported findings.

Soon after the availability of MT, several studies reported that MT had a very limited impact on surgical decision making for indeterminate thyroid nodule cytologies (for references see Huang et al. ¹ ). Also, the validation studies for all three MTs for indeterminate thyroid nodules currently used in the United States focused on the evaluation of the negative predictive value (NPV) and the resulting benign call rate (BCR) of the respective MT. The respective NPV and BCR calculations were based on the assumption that patients with test negative nodules avoid surgery because of the MT result and that all indeterminate nodules would undergo diagnostic surgery without MT.

However, this has never been the case in clinical practice as it would have resulted in overtreatment. Average resection rates for atypia of undetermined significance/follicular lesion of undetermined significance (AUS/FLUS) and for follicular neoplasm/suspected follicular neoplasm (FN/SFN) nodules are 39% and 70%, ³ and in our setting, before introduction of MT were even 21% for AUS/FLUS and 56% for FN/SFN respectively. ⁴ Moreover, the real clinical impact of published BCRs still need to be determined by long-term follow-up studies for nonoperated patients with indeterminate thyroid nodule cytologies.

An increase in indeterminate diagnoses has been reported in some centers after implementation of MTs. As this may partly offset any positive effect of MTs, high cytology performance needs to be maintained and monitored after implementation of MT.

Repeat FNAs of AUS/FLUS nodules can lead to a definitive diagnosis for half of the repeat FNAs and is, therefore, a respective guideline recommendation. ² Restriction of MT to two separate Bethesda III biopsies has been shown to reduce the rate of diagnostic surgery for histologically benign nodules while missing only rare low-risk tumors. ^5,6

The important diagnostic challenges with rat sarcoma virus (RAS) mutations have previously been addressed in an editorial in Thyroid. ⁷ RAS mutations (NRAS/HRAS/KRAS) are the most frequent finding with any molecular FNA test and lead to classification of the respective nodule as neoplasia. However, several systematic reviews showed that the risk for a RAS-positive nodule to be thyroid cancer is highly variable. Therefore, resection of all RAS-positive nodules may result in an increase in resections and overtreatment of benign nodules.

Thus, clinical decision and complementary instead of binary use of RAS findings are required to determine whether the life of the respective patient will likely be impacted by a RAS-positive neoplasia (with no further detected mutation), that is, if the patient will die with it instead of it. Thyroid cancers detected for indeterminate FNAs are typically low-risk thyroid cancers, and observation for indeterminate FNA cytologies has been reported as a feasible option and is a reasonable management strategy according to the recent Bethesda guideline update. ⁸

Most important, all MT validation studies did not report details for the upstream malignancy risk stratification, especially the quality and results of ultrasound malignancy risk stratification and malignancy risks and resection rates for Bethesda categories. Also, the selection of patients based on history and clinical findings is of importance but not reported in MT validation studies.

Only two recent studies began to evaluate the additional impact of MTs as part of an optimized integrated interdisciplinary thyroid nodule diagnostic pathway by also evaluating the further upstream components of thyroid nodule malignancy risk stratification after their respective optimization. These studies reported that the added value of MTs is highest in ATA low- and intermediate-suspicion/TR4 nodules and adds little to the malignancy risk stratification of ATA high-suspicion/TR5 nodules. ^5,9

Previous cost analyses have not addressed the above topics as indispensable parts of an optimized integrated thyroid nodule diagnostic pathway. It is, therefore, noteworthy that a 700 USD MT focused on thyroid nodules with ultrasound ATA low/intermediate risk, ACR-TIRADS 3/4 in an optimized integrated diagnostic pathway has a similar NPV as reflex MT of all ITNs with a >3000 USD MT. ^5,9 This can possibly be further optimized by restriction of MTs to repeat ITNs and by not performing MTs if it will not affect management. ^2,10

So, what is the lesson to be learned from the reported lack of impact of the rapid increase of MTs of indeterminate FNA cytologies in the United States on thyroidectomy rates? Like for other tumors, the diagnosis of thyroid tumors is increasingly transitioning to a complementary or symbiotic histological and molecular classification with increasing importance for treatment stratification. However, together with previous findings summarized above, the Huang study is a further reminder that MT is unable to compensate for upstream deficiencies in a thyroid nodule diagnostic pathway.

The disconnect between rapid increase in MT utilization and thyroidectomy rates demonstrated by Huang et al. re-emphasizes that MT of indeterminate FNA cytologies needs to be appropriately focused on situations where it can make a clinically important difference and it needs to be integrated into an optimized integrated interdisciplinary thyroid nodule diagnostic pathway with expertise and input from all disciplines involved as emphasized by the recent ETA thyroid nodule guideline and textbooks. ¹⁰

Author's Contribution

R.P. wrote the editorial.