Recent advances in managing differentiated thyroid cancer

The main clinical challenge in the management of thyroid cancer is to avoid over-treatment and over-diagnosis in patients with lower-risk disease while promptly identifying those patients with more advanced or high-risk disease requiring aggressive treatment. In recent years, novel clinical and molecular data have emerged, allowing the development of new staging systems, predictive and prognostic tools, and treatment approaches. There has been a notable shift toward more conservative management of low- and intermediate-risk patients, characterized by less extensive surgery, more selective use of radioisotopes (for both diagnostic and therapeutic purposes), and less intensive follow-up. Furthermore, the histologic classification; tumor, node, and metastasis (TNM) staging; and American Thyroid Association risk stratification systems have been refined, and this has increased the number of patients in the low- and intermediate-risk categories. There is now a need for new, prospective data to clarify how these changing practices will impact long-term outcomes of patients with thyroid cancer, and new follow-up strategies and biomarkers are still under investigation. On the other hand, patients with more advanced or high-risk disease have a broader portfolio of options in terms of treatments and therapeutic agents, including multitarget tyrosine kinase inhibitors, more selective BRAF or MEK inhibitors, combination therapies, and immunotherapy.


Introduction
The incidence of differentiated thyroid cancer (DTC) continues to rise worldwide, mostly because of the growing use of powerful diagnostic tools that permit the discovery of an increasing number of small papillary thyroid cancers (PTCs). Mortality from DTC, however, has changed minimally over the past five decades: the vast majority of patients with DTC have indolent tumors. The main clinical challenge is to avoid over-diagnosis in patients with low-risk disease or benign thyroid nodules while promptly identifying those patients with more advanced or high-risk tumors that require aggressive treatment approaches. In recent years, we have witnessed the emergence of new clinical and molecular data that have allowed the development of new staging systems, new predictive and prognostic tools, and new approaches to treatment.
Here, we will review the main advances regarding the staging and management of DTC.

Pathologic definition and staging systems
The fourth edition of the World Health Organization histologic classification of endocrine tumors, published in June 2017 1 , contains important revisions, which may have a broad impact on clinical practice. First, the follicular-derived neoplasms now include a new entity within the group of tumors with borderline histological features: the non-invasive follicular thyroid tumor with papillary nuclear features (NIFTP) 2,3 , an encapsulated follicular-variant PTC with no evidence of capsular or vascular invasion. NIFTPs may account for up to 20% of the tumors previously classified as PTCs. The word "cancer" has been eliminated from the definition of NIFTP to underline its excellent prognosis: no adverse outcomes (such as cancer-related death, distant or regional metastases, and structural or biochemical recurrence) have been recorded for these tumors. The explicitly stated objective is to reduce the intensity of treatment and follow-up and the psychological consequences of the diagnosis of cancer. This entity was first proposed by an international panel of pathologists and other professionals and has been endorsed by the American Thyroid Association (ATA). However, the seemingly excellent outcomes still need to be confirmed in long-term prospective studies. The current evidence is retrospective and of only moderate quality 2 . Other relevant changes include the identification of 15 PTC variants and the distinction of follicular thyroid cancers (FTCs) into three subgroups (minimally invasive FTCs with capsular invasion only, encapsulated angioinvasive FTCs, and widely invasive FTCs), which reflect the prognostic relevance of vascular invasion. Given its unique molecular profile 4 , the oncocytic variant is now considered a separate entity and is referred to as Hürthle cell carcinoma. Poorly differentiated thyroid cancer is also a separate entity, in accordance with the Turin criteria 5,6 .
The American Joint Cancer Committee/Union Internationale Contre le Cancer TNM staging system for thyroid cancer takes into account patient age, the size and extent of the primary thyroid tumor (T), lymph node involvement (N), and the presence of distant metastases (M). This system is aimed at predicting mortality (not recurrences). The eighth edition of this staging system contains important modifications. The age cutoff was raised from 45 to 55 years 7 , the presence of minimal extrathyroidal extension is no longer relevant for the T classification, and the tumor stages have been redistributed (Table 1 and Table 2). Regional lymph node metastases or the presence of gross extrathyroidal extension limited to strap muscle no longer mandate stage III but stage II. The presence of macroscopic invasion beyond strap muscle (subcutaneous tissue, larynx, trachea, esophagus in T4a tumors or carotid artery, prevertebral fascia or mediastinal vessels in T4b tumors) is an unfavorable prognostic factor. The overall goal of these modifications is to improve the accuracy of the system's prediction of mortality 8 , restricting the assignment of higher risk to a small subset of patients (5-10%) with tumors classified as stage III (pT4a, any N, M0) or stage IV (pT4b, any N, M0, and M1). Implementation of the eighth edition revisions is expected to result in the down-staging of approximately 30-40% of patients with thyroid cancer 9,10 .
In the 2015 edition of its practice guidelines, the ATA also revised its system for stratifying the risk of recurrent disease. The new variables considered include new pathologic features of the tumor (for example, histopathological variant, vascular invasion, number of metastatic lymph nodes, size of the largest metastatic lymph node, and the presence of extranodal extension) as well as its molecular characteristics (mutational status of BRAF and the TERT promoter 11 , when available) 12 . These changes are expected to allow more precise estimates of the likelihood of recurrence.
The dynamic risk classification process used during follow-up assigns patients to one of four subgroups and may be modified at each follow-up examination: responses to therapy are classified as excellent, biochemically incomplete, structurally incomplete, or indeterminate response.

Management strategies
Current international guidelines advocate personalized decisionmaking-based on the risk of recurrence and disease-specific death-regarding the extent of surgery, the use of radioactive iodine (RAI) therapy, the intensity and length of follow-up, and the degree of thyroid-stimulating hormone (TSH) suppression.

Active surveillance
The 2015 ATA guidelines include active surveillance among the management options for small subcentimeter PTCs. In pivotal Japanese studies, this strategy appeared to be both safe and effective 13,14 : after 10 years, very few patients had experienced tumor growth (8%), and the development of lymph node metastases was even less common (4%). Age below 40 at diagnosis was an independent risk factor for disease progression 15 . In terms of cures, delayed surgical treatment of these tumors was as effective as immediate treatment 15 .
In a study conducted in the United States, 291 patients with cytologically suspicious or malignant thyroid nodules (Bethesda class V or VI) measuring 1.5 cm or less were managed with active surveillance for a median of two years 16 . The percentages of tumors displaying growth were 2.5% at two years and 12% at five years. Independent predictors of growth were age under 50 years and clinical judgment as "inappropriate for active surveillance" 16 . The latter label may be applied on the basis of nodule-related features (subcapsular location adjacent to the   Other observational clinical trials to evaluate the "active surveillance" approach in subcentimeter PTCs are underway in Korea and Israel (NCT02952612, NCT02938702, and NCT02609685). Also, there is a need for biomarkers that can identify those rare microcarcinomas that are likely to grow, so they can be promptly referred for surgery.

Individualized surgical approaches
According to the ATA guidelines 12 , thyroid lobectomy (TL) may be used for low-risk, intrathyroidal tumors up to 4 cm in size with no lesions in the contralateral lobe. Total thyroidectomy (TT) was previously considered the preferred approach for these tumors.
In a retrospective analysis of 52,173 cases in the Surveillance Epidemiology and End Results (SEER) database, TL for tumors measuring at least 1 cm was associated with small but statistically significant increases in the risks for recurrence (9.8% versus 7.7%) and mortality (2.9% versus 1.6%) compared with TT 18 . A recent retrospective analysis with a more extensive risk stratification found no such difference in terms of overall survival 19 , but, in another meta-analysis, the risk of recurrence after TL was significantly higher than that after TT (8.3% versus 4.4%; p<0.01) 20 . It is worth noting that tumor recurrence in the contralateral lobe has been observed in 5% of patients treated with TL 21 and that benign nodule relapse is reported in 20-50% 22,23 . Levothyroxine treatment has been reported to prevent benign nodule relapse, but the evidence for this effect is limited 24 .
TL offers several advantages over TT. First, the rate of side effects is lower with TL. It virtually eliminates the risks of permanent hypoparathyroidism and bilateral RLN palsy 25 and reduces the rates of permanent unilateral RLN palsy (0.6% in TL versus 1.3% in TT) 25 . Second, surgical hypothyroidism after TT requires lifelong levothyroxine (LT4) replacement therapy. The rate of hypothyroidism after TL varies from 23.6 to 47% 26,27 , and patients with a normal serum TSH level may not require LT4 treatment at all. TL and TT are also believed to have differential effects on the patient's quality of life. However, two relatively small studies have failed to detect any difference 28,29 and thus further investigation of this issue is needed.
The ATA guidelines do not advocate prophylactic central neck dissection for low-risk patients 12 despite the high frequency of subclinical lymph node metastasis in DTC 30 . In fact, microscopic lymph node metastases that are not clinically detected before surgery have a questionable role in patient outcome 31 , and central compartment neck dissection carries an increased rate of surgical complications, hypoparathyroidism in particular 32,33 . In a prospective study of patients with no preoperative evidence of lymph node metastasis who were randomly assigned to undergo TT alone or TT with central compartment neck dissection, no difference in outcomes was found after five years of follow-up 32 .

Radioiodine remnant ablation: selective use
In the past, routine use of RAI ablation therapy after surgery was justified first by the need to eliminate residual normal thyroid tissue, to achieve an undetectable serum thyroglobulin (Tg) level. It also allowed the identification of persistent neoplastic tissue with a 131 I whole-body scan (WBS) and was likely to destroy any occult nests of neoplastic cells, thereby improving long-term outcomes. These indications have been questioned in recent years 12,34,35 . The ATA guidelines now recommend selective use of RAI based on individual risk 12 . Claiming uncertainties and ambiguities in the evidence, the European Association of Nuclear Medicine refused to endorse these recommendations 36 and noted that there are no prospective, controlled study data that allow us to identify the patients with low-risk DTC who may not benefit from RAI ablation. However, the Association did not emphasize that a treatment should be given only in patients in whom it may be beneficial and did not acknowledge that uncertainties persist concerning benefits of RAI administration in low-and intermediate-risk patients.
It is generally agreed that RAI has no role in the management of patients with intra-thyroidal microcarcinomas. In other low-and intermediate-risk patients, the decision to ablate can be based on individual prognostic factors and on the serum Tg level measured 6 weeks after surgery either on LT4 treatment with a sensitive assay or following recombinant human TSH (rhTSH) injections. An undetectable or a low serum Tg level at that time supports a decision to avoid RAI administration. When it is indicated, it should consist of the administration of 1.1 GBq following rhTSH injections 37 .
Two randomized clinical trials in Europe are enrolling low-risk patients and aim to obtain reliable data on the indications for post-operative RAI administration. In the ESTIMABL2 trial (NCT01837745) being conducted in France, 750 patients with a T1bN0,Nx tumor will be randomly assigned to post-operative ablation with an activity of 30 mCi after rhTSH stimulation or simple follow-up. The IoN trial in the UK (NCT01398085) has a similar design. The primary outcomes in the two studies are disease-free survival rates at 3 and 5 years, respectively.

Follow-up tools
The main tools used for the follow-up of DTC are neck US and serum Tg determination 38 . Undetectable serum Tg levels can reliably identify disease-free patients and have a negative predictive value close to 100%. In contrast, early minimally detectable levels have a low positive predictive value: the majority of patients with these findings remain free of structural disease during prolonged follow-up 39 . Tg trends over time-instead of absolute alues-should be monitored: declining levels are reassuring, whereas increases suggest the presence of growing thyroid tissue (normal or neoplastic) 40 . In the presence of Tg autoantibodies (TgAbs), serum Tg levels determined by immunometric assays may be falsely low. In these cases, management can be guided by the temporal trends in the TgAb titers themselves 41 . Novel biomarkers are emerging as replacements for serum Tg in these difficult cases, such as circulating microRNAs and other nucleic acids, but still need to be standardized and clinically validated 42-44 . The use of mass spectrometry for measuring serum Tg levels in the presence of TgAb also needs to be validated 45,46 . Sensitive Tg assays that can detect serum concentrations as low as 0.1 ng/mL are currently used and provide similar information on disease status as rhTSH-stimulated Tg obtained in the past with assays that had a sensitivity of 1 ng/mL. RhTSH-stimulated Tg levels are now measured during the follow-up only in those few patients with low but detectable serum Tg on LT4 treatment 47 . In these patients, a substantial increase in the stimulated Tg level may indicate the presence of neoplastic tissue.
Neck US provides useful information. PTC almost always spreads first to the cervical lymph nodes, where it can be identified sonographically using specific criteria 48 , thus eliminating the need for diagnostic 131 I WBS. US is more cost-effective, eliminates radiation exposure, and has no adverse effects. Surgical treatment of lymph node metastases is recommended for lesions with smaller diameters exceeding 10 mm (for lateral N1) or 8 mm (for central N1), and there is no need for discovering small N1 of only a few millimeters in diameter. However, US is notoriously operatordependent, and some findings are non-specific and classified as indeterminate. Lesions with such features display significantly lower rates of persistence and growth than those with more suspicious US characteristics 49 . Suspicious findings can be confirmed by US-guided FNA with cytologic assessment and assay of Tg in the needle-washout fluid 50 . Distant metastases are rare in patients with negative findings on neck US. However, in the presence of rising Tg levels or suspicious clinical features, second-line functional (diagnostic 131 I WBS and 18-fluorodeoxyglucose positron emission tomography scan) or cross-sectional (computed tomography or magnetic resonance imaging) imaging studies may be performed 51 .
Risk assessment is a dynamic process: the response to therapy is re-assessed on the basis of findings at each follow-up visit and expressed as excellent, indeterminate, biochemically incomplete, or structurally incomplete 12,39,52 . During long-term follow-up, even patients who initially displayed a high risk of persistent or recurrent disease can be re-classified as having lower-risk disease, and their follow-up program can be less intensive than originally planned. Lifelong surveillance is still recommended 12 . However, over 75% of recurrent lesions are identified within the first five years of follow-up, and late recurrence is very unlikely in low-or intermediate-risk patients with excellent responses to treatment 53 .
TSH stimulates the proliferation of normal and neoplastic thyrocytes 54 , and levothyroxine treatment significantly reduces DTC recurrence and cancer-related mortality 55 . The optimal TSH level is unclear. TSH suppression increases the risk for atrial fibrillation and osteoporosis in older patients and the risk of angina in patients with ischemic heart disease 56 . Suppressive therapy should take into account both the likelihood of complications and the risk of increasing tumor cell proliferation. For this reason, levothyroxine treatment is no longer recommended for low-and intermediate-risk patients with no evidence of disease. The goal in these cases is a serum TSH level within the normal range. Suppressive therapy is advocated only in patients with structural disease and no contraindications 12 .

Treatment of distant metastases
RAI ( 131 I) is currently the first-line treatment for distant metastases that are RAI-avid. Traditionally, thyroid hormone withdrawal has been the preferred method of preparation for this type of patient because it is associated with higher neoplastic tissue uptake and slower clearance of RAI than that achieved with rhTSH preparation. Observational data have suggested that rhTSH preparation may also be effective in terms of response to treatment in these patients 57 , but the evidence in support of this conclusion is currently insufficient to recommend rhTSH use in patients with metastatic DTC 12 . The activity of RAI for the treatment of distant metastases can be calculated dosimetrically or a fixed empiric dose can be used. A large retrospective study of DTC patients with distant metastases found similar overall survival with the two approaches 58 after adjustments for age and tumor burden 59 .

Radioactive iodine-refractory disease
RAI-refractory DTC is defined as persistent neoplastic tissue that does not take up RAI; disease characterized by heterogeneous RAI uptake (that is, some lesions are RAI-avid and others are not); or disease that progresses after RAI treatment despite RAI uptake 12 . Specific molecular profiles are more likely to result in RAI-refractory disease 60,61 . Even in the presence of distant metastases, most patients have asymptomatic, slowly progressive disease. RAI-refractory patients should benefit from local treatments (surgery, external beam radiotherapy, or thermal ablation, depending on the site of the lesion and local expertise) if they have symptoms or a high risk of local complications.
When disease progression occurs at multiple sites in patients with target lesions of more than 1-2 cm in diameter, treatment with tyrosine kinase inhibitors (TKIs) should be considered.  NCT03244956). Immunotherapy also appears to be a promising approach to thyroid cancer, alone or in association with other drugs 70 (NCT02390739 and NCT03181100). These agents have yet to be approved by regulatory agencies, such as the U.S. Food and Drug Administration and the European Medicines Agency. Enrollment in clinical trials for patients with progressive metastatic disease should be considered and encouraged in order to improve both clinical case outcomes and medical knowledge in the field 71 .

Conclusions
In recent years, DTC treatment has become considerably more conservative, with less extensive surgery (or no surgery at all), reduced use of radioisotopes, and less intensive follow-up of low-and intermediate-risk patients. Furthermore, the systems used to histologically classify and stage DTCs have recently been refined, along with the ATA scheme for estimating their risk of recurrence, and more patients are now considered to be at low or intermediate risk. There is a need for new, prospective data to clarify how these changing practices will impact the long-term outcome of these patients. On the other hand, patients with more advanced or high-risk disease now have a broader portfolio of treatment options, including multitarget TKI therapy, more selective BRAF inhibitors, combination therapies, and immunotherapy. However, the indications for each, their optimal starting times and dosing schedules, and their long-term safety profiles remain to be clarified in coming years.

Competing interests
The authors declare that they have no competing interests.

Grant information
The author(s) declared that no grants were involved in supporting this work.

Open Peer Review Current Referee Status:
Editorial Note on the Review Process are commissioned from members of the prestigious and are edited as a F1000 Faculty Reviews F1000 Faculty service to readers. In order to make these reviews as comprehensive and accessible as possible, the referees provide input before publication and only the final, revised version is published. The referees who approved the final version are listed with their names and affiliations but without their reports on earlier versions (any comments will already have been addressed in the published version).
The referees who approved this article are:

Version 1
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com Ear, nose and throat (ENT) Department, NHS Lothian, Edinburgh, UK

Iain Nixon
No competing interests were disclosed.