Abstract

Background Information on the Year in Review Lectures of the American Thyroid Association
The Year in Review lectures were presented on October 31, 2024 at the Annual Meeting of the American Thyroid Association, in Chicago, Illinois. The session consisted of three lectures that highlighted selected published research advances in the thyroid field in the past year under the general categories: (1) basic research, presented by Douglas Forrest, (2) clinical research, presented by Jennifer Sipos, and (3) surgical research, presented by Elizabeth Grubbs. Each presentation considered several notable publications in the field but for reasons of time limitations, these were necessarily selective. Each summary therefore represents a personal view but is intended to reflect some of the major advances in the broad and active field of thyroid research. Although only a small selection of studies could be presented at the session (references annotated by * in the accompanying tables) several additional publications that were considered have been included in the tables in this summary article. Each of the authors conducted a search of the published literature in their respective fields from September 2023 to October 2024 (up to the date of the lecture) and included searches of relevant electronic databases and journals. The authors also consulted numerous colleagues.
Basic Thyroid Research
The thyroid gland
The genetic controls underlying the function of the thyroid gland and how disruption of these controls leads to thyroid disorders remain incompletely understood. Two studies from Japan and the United States reported convergent findings about the genetic basis of conditions described variously as compensated hypothyroidism, 1 or as resistance to thyrotropin (resistance to TSH) 2 (Table 1). These are dominantly inherited disorders with elevated TSH but thyroxine (T4) within normal ranges. Remarkably, mutations were found in a non-coding region of chromosome 15q in an Alu repetitive element. The mutations resided in a short tandem repeat sequence and resulted in an open chromatin structure creating the potential for recruitment of transcription factors at the site. The mutations derepressed a micro-RNA locus 35 kb downstream from the mutation that could potentially result in interference with the expression of genes for thyroid function. 2 These mutations were estimated to account for approximately 13% of cases of these thyroid conditions. Given that the majority of the mammalian genome is non-coding, the findings point to the importance of further investigation of non-coding sequences as determinants of thyroid gland function.
Thyroid hormone action
Thyroid hormone (triiodothyronine [T3]) receptors are nuclear receptor transcription factors. Despite decades of study in vitro surprisingly little is known of the receptor-regulated chromatin sites (i.e., enhancers) underlying T3-regulated gene expression in most tissues in vivo, including in the pituitary gland, a key target organ. Altered pituitary function often accompanies thyroid disorders, as reflected, for example, in changes in growth hormone and TSH levels. A genome-wide study 3 investigated T3-regulated responses in the mouse pituitary gland at the critical level of chromatin regulation. The study revealed a genome-wide map of receptor binding sites and indicated how T3 opens chromatin and modifies histones at specific populations of sites. The authors proposed a “poised receptor” model in which persistent association of receptors with these sites, regardless of T3 levels, permits sensitive, reversible regulation of chromatin and gene expression in response to T3 fluctuations. There remains much scope to broaden the investigation of T3 action at this crucial level of chromatin responses in other target tissues.
Growing numbers of studies are taking advantage of model systems derived from induced pluripotent stem cells (iPSC) in a “differentiation in a dish” approach to investigate thyroid hormone functions. A study of human iPSC-derived hepatic organoids 4 has indicated a role for type 2 deiodinase (encoded by DIO2), a T3-generating enzyme, in the maturation of hepatocyte-like cells in culture. A prominent peak of DIO2 expression was detected during organoid maturation, resembling an embryonic peak reported in mouse liver in vivo, and supporting a conserved role for DIO2 in early programming of human hepatic differentiation. Transcriptome analyses indicated that DIO2 promotes hepatocyte gene expression and the proportion of hepatocyte-like cells in mature organoids. The findings indicate the importance of deiodinase enzymes in mediating thyroid hormone action within tissues and the value of iPSC-derived models in yielding insights into T3 action in human tissue differentiation.
Highlighted Articles for “Basic Thyroid Research”
Article featured in the talk.
Highlighted Articles for Clinical Thyroid Research
Article featured in the talk.
RCT, rearranged during transfection; DTC, differentiated thyroid cancer; FNA, fine-needle aspiration; TSH, thyrotropin; ATA, American Thyroid Association; RAI, radioactive iodine.
Highlighted Articles for Surgical Thyroid Research
Article featured in the talk.
Thyroid cancer
Papillary thyroid cancer is a common endocrine cancer and up to 15% of cases exhibit aggressive characteristics. The combination of two genetic changes, namely, in the BRAF gene (BRAF kinase) and in the promoter of the TERT (telomerase reverse transcriptase) gene has previously been associated with aggressive forms of this cancer. TERT maintains telomere length during cell division but normally becomes inactive upon differentiation. Promoter mutations abnormally activate TERT expression, bypassing controls that shorten telomeres as proliferation diminishes. A study 5 has now reported a third alteration, a distinct single-nucleotide variation (SNV) in the TERT promoter that further elevates promoter activity and is associated with more aggressive papillary thyroid cancer. In combination with the previously recognized alterations in the TERT promoter and BRAF, the additional SNV was associated with higher cancer recurrence in this study. The authors propose that testing for all three changes may refine the prognosis for aggressive papillary thyroid cancer.
Some papillary thyroid cancers display familial clustering, but the genetic basis is poorly understood. A new study 6 has reported germline variants in several genes encoding telomere-binding proteins (POT1, TINF2, ACD) in familial papillary thyroid cancer. Individuals displayed ultra-long telomeres that in some cases became longer in successive generations and also a tendency to display syndromic cancer with additional cancers including melanoma and lymphoma. The papillary thyroid tumors in this cohort carried somatic mutations in BRAF (V600E) but no telomere-maintenance changes (such as TERT promoter changes). The authors suggest that telomere-lengthening germline variants predispose to a syndromic subtype of papillary thyroid cancer without a need for somatic changes that maintain telomeres. These two studies 5,6 highlight the role of changes, whether somatic or germline, that interfere with telomere-shortening controls in thyroid tumorigenesis.
Potential interventions based on knowledge of thyroid hormone action
Advances in understanding the basic molecular mechanisms underlying biological processes are a prerequisite for devising new or improved therapeutic approaches to disease. This potential is being realized in a number of ways across the field of thyroid research. A recent study 7 based on knowledge of T3 actions in neural differentiation, investigated if T3 can modify the outcome of medulloblastoma, a brain cancer, mostly occurring in childhood. The authors tested whether T3 can stimulate terminal differentiation and accordingly, suppress proliferation of neural precursor cells that form the tumor. T3 dissociated a corepressor from the T3 receptor, allowing the expression of NeuroD1, a neural differentiation factor that reduced tumor proliferation. T3 treatment improved the survival of mice carrying transplanted medulloblastoma cells, promoting differentiation rather than proliferation of these cells. The authors suggest that manipulating T3 signaling as a form of “differentiation therapy” might offer a future means of suppressing the progression of this cancer.
Clinical Thyroidology
Artificial intelligence
Artificial intelligence (AI in the field of radiology provides an enticing opportunity to examine sonographic images and improve malignancy risk stratification of thyroid nodules. Ha et al. 8 constructed and validated a deep-learning-based model for thyroid cancer diagnosis from 19,711 ultrasound images of thyroid nodules with a final biopsy or histological diagnosis (Table 2). This tool was then used to train radiologists with differing levels of experience. They found a significant improvement in diagnostic performance when physicians of all levels of experience were trained with the AI tool. Importantly, this AI training also dramatically improved the interobserver agreement from moderate (kappa value 0.472) to substantial (kappa value 0.723). The authors conclude that while AI is not practical for daily clinical evaluation of thyroid nodules due to the significant time required to select images for analysis, the tool offers an important mechanism for training clinicians to interpret sonographic examinations.
Autoimmunity
The determination of when to discontinue thionamide therapy for Graves’ disease (GD) is an important clinical challenge. Postulating that addressing the underlying autoimmune component of the disease may provide an opportunity for earlier GD remission, Xie et al. 9 conducted a prospective, open-label randomized clinical trial of low-dose methotrexate (MTX) in combination with methimazole (MMI) compared with MMI alone. They found that the discontinuation rate of the MTX group (55.6%) was significantly higher than those taking MMI alone (38.9%). In addition, they noted that those taking MTX had higher rates of negative TRAb titers (87.5%) compared with those taking MMI alone (33.4%) at study end. Importantly, MTX was well tolerated without any serious adverse events. This first-of-its-kind study offers appealing improvements in some important clinical endpoints of those with GD. It is important to note, however, that this cohort was followed for 18 months, and a longer study is needed to establish the durability of their MTX-induced remissions.
Thyroid cancer
Understanding which of the many multikinase and targeted therapies to utilize as a first-line treatment for progressive medullary thyroid cancer (MTC) is a notable clinical challenge. The Phase 3 trial by Hadoux et al. 10 specifically addresses this question by randomizing 291 patients with advanced, rearranged during transfection (RET) mutant MTC to selpercatinib or the physician’s choice of cabozantinib or vandetanib (control). At a median follow-up of 12 months, they found that median progression-free survival (PFS) was not reached in the selpercatinib group and was 16.8 months in the control group (hazard ratio for disease progression or death, 0.28; confidence interval [CI], 0.16–0.48; p < 0.001). This lack of a median PFS for the selpercatinib group indicates that the treatment has prevented disease progression in a significant proportion of the patient population during the trial and is suggestive of a positive treatment effect. The PFS at 12 months was 86.8% in the selpercatinib group and 65.7% in the control group. Dose reductions were significantly less likely in the selpercatinib group (38.9%) compared with the control group (77.3%). This important study offers unique insights into the efficacy of specific therapies for progressive thyroid cancer through its randomized, prospective, and head-to-head comparison of multiple agents.
Thyroid function
Although thyroid function changes with age, most laboratories do not differentiate age-specific reference intervals (RIs) beyond childhood. Establishing appropriate RIs for adults is particularly important because of the high rate of subclinical thyroid dysfunction, and ultimately, unnecessary use of levothyroxine supplementation in older adults. Jansen et al. 11 sought to establish age-specific RIs for TSH and free T4 (FT4) throughout life using big data from 13 Dutch laboratories, which included 7.6 million TSH values and 2.2 million FT4 results. They determined that the upper reference limit (URL) in children was higher than the adult range until 12–14 years of age, after which it aligned with the adult range. Similarly, the FT4 lower reference limits were higher in early childhood and decreased toward adulthood after age 10–12 years. In adults, the TSH URL increased after the age of 50 years in women and 60 years in men. The FT4 URL increased for both sexes after the age of 70 years. Furthermore, they determined that using an age-specific RI resulted in a significant decrease in the diagnosis of subclinical hypothyroidism (SH) in older adults. This large study provides valuable insights into appropriate age categories for RIs across the age continuum and demonstrates that using these intervals will reduce the number of diagnoses of SH.
Thyroid nodules
Sample preparation from a fine-needle aspiration (FNA) of a thyroid nodule may be performed with either conventional smears (CS) or liquid-based cytology (LBS). While each method has its strengths and limitations, few studies have examined systematically the diagnostic performance of each technique, and controversy remains over the optimal approach to FNA specimen preparation. In a single-center study by Lee et al., 12 they compared the diagnostic performance of 11,438 CS samples and 23,968 LBS specimens. There was a significantly higher nondiagnostic cytology rate with CS compared with LBS. Furthermore, the sensitivity of FNA increased from 71% with CS to 82% with LBS. This study is the largest of its kind and provides valuable insights into the benefits of LBS compared with CS.
Pediatrics
Recognizing that most pediatric thyroid cancers are associated with an excellent long-term prognosis, recent guidelines have recommended more constraints in the use of radioactive iodine (RAI). Two studies examined the outcomes of pediatric thyroid cancer patients who did not receive RAI. The first analysis by Bojarsky et al. 13 compared outcomes of 95 patients with low-risk differentiated thyroid cancer (DTC), 50 (53%) of whom were treated with RAI and 45 (47%) were followed without RAI. They found no significant difference in one-year remission rates between those treated with and without RAI. Similarly, Castellanos et al. 14 described their experience with individuals harboring initially low-, intermediate-, and high-risk papillary thyroid cancer, the majority of whom remained disease-free without further therapy. Importantly, this latter cohort expanded the potential indications to avoid RAI as they included patients treated with lobectomy alone (21%) and those with potentially more aggressive clinical features like lymph node metastases (57%). Additional long-term, multicenter studies are needed to further characterize the clinical features of disease that may be successfully treated without RAI.
Pregnancy
The rates of progression to overt hypothyroidism from SH and hypothyroxinemia (HT) diagnosed during pregnancy have not been specifically examined. A secondary pooled analysis of two multicenter treatment trials for individuals with SH or HT diagnosed between 8 and 20 weeks’ gestation was performed by Varner et al. 15 This study examined those individuals randomized to the placebo group at 1 and 5 years after delivery for the development of overt hypothyroidism. They found that 23.1% of those with SH developed hypothyroidism during the five-year follow-up after delivery, compared with 5% of those with HT. Progression to hypothyroidism was more common in those with a TSH >10 µIu/mL. At the multivariable analysis of the combined groups, those with a thyroid peroxidase antibody (TPOAb) titer over 50 IU/mL had higher rates of hypothyroidism at year 1 or 5 (OR 7.16, CI 4.09–12.56). This study emphasizes the importance of longitudinal evaluation of thyroid function studies in patients diagnosed with SH or HT during pregnancy, particularly when the TSH or TPOAb titer is high.
Surgical Thyroidology
Non-surgical treatment of thyroid disease
Retreatment after thermal ablation of benign thyroid nodules is becoming pertinent as technical inefficiency from first time treatment is becoming increasingly observed. Benardi et al. 16 provided an initial multicenter retrospective cohort study evaluating 135 patients who underwent retreatment with thermal ablation, through radiofrequency ablation (RFA) or laser ablation (LA), for benign thyroid nodules whose volume decreased by less than 50% after initial treatment (considered technique inefficiency) (Table 3). Retreatment led to a median volume reduction ratio of 50% and 52% after 6 and 12 months from the procedure and was more successful than first treatment in terms of nodule volume reduction, especially if the nodules were less than 30 mL in volume and RFA was used, as opposed to LA. Overall, the authors noted the effect of thermal ablation retreatment appears somewhat limited. The study by Fugazzola et al. 17 explores the novel application of RFA for treating Bethesda III thyroid nodules with no detectable genetic alterations, a category that carries a 10% malignancy risk. This bi-institutional prospective study treated 33 nodules, achieving a mean volume reduction rate of 60–62% over one year. As the first report of RFA as a single-session alternative to surgery or active surveillance for indeterminate, oncogene-negative nodules, the study provides promising preliminary insights.
Evaluating the use of thermal ablation for malignancies, the single-center study by Li et al. 18 evaluates long-term outcomes of RFA for T1N0M0 papillary thyroid carcinoma (PTC) in a large cohort of 1613 patients with a median follow-up of 58.5 months. Local tumor progression occurred in 4.3% of patients, with multifocal tumors and subcapsular location (<2 mm from the capsule or trachea) identified as independent risk factors. This study is the first to report on long-term outcomes of RFA for stage T1b PTC, though it only made up 12% of the population. T1b PTC was an independent factor for tumor persistence, which tracks with larger tumors posing a greater challenge for RFA due to the limited size of the ablation zone obtained.
The same group also was the first to examine the pathology of the post-ablation zone after RFA in PTC with a five-year follow-up, in efforts to define criteria for successful RFA treatment of thyroid cancer, which heretofore has not been well delineated. 19 Primary endpoints included persistent disease, defined as persistently detected PTC in the ablated zone as confirmed by biopsy. In a subset of 382 biopsied patients from the above study, the incidence of persistent disease was 3.9%, which is higher than the range of 0–1% reported in previous studies. The persistent disease rates in the T1a and T1b groups were 2.9% and 12.2%, respectively.
Quality of life
A prospective cohort study by Sawka et al. examined decision regret and quality of life (QOL) among Canadian patients with small, low-risk PTC who chose surgery or active surveillance. 20 At one year, both groups reported low decision regret (scores below 25), with no significant differences in fear of disease progression, anxiety, depression, or body image after adjustments. Low levels of regret observed in both groups may reflect both cohorts avoiding the consequence they feared the most whether avoiding disease progression (surveillance) or treatment-related consequences (surgery).
Bach et al.’s systematic review compared health-related QOL (HRQOL) in low-risk DTC survivors after total thyroidectomy versus hemithyroidectomy. 21 Data from 16 studies involving 4957 patients suggest that hemithyroidectomy generally offers better HRQOL, particularly in the early postoperative period, with advantages noted in physical, psychological, and voice-related domains. However, most studies measuring outcomes beyond one year found no significant differences, highlighting a need for high-quality randomized trials to confirm these findings.
Pediatrics
Castellano et al.’s ambispective single-institution study was one of the first to evaluate outcomes in 93 pediatric patients treated with surgery alone, without adjuvant RAI, over a median follow-up of 5.5 years. 14 The study found that 91% of patients remained disease-free, including those with American Thyroid Association (ATA) intermediate- and high-risk disease who had favorable dynamic risk stratification (DRS) after the initial surgery. The findings suggest that DRS may guide RAI decisions in the pediatric population.
Intraoperative nerve monitoring
A study by Ramesh et al. 22 explored outcomes of planned total thyroidectomy in cases of first-side loss of intraoperative nerve monitoring signal, addressing the critical decision-making dilemma of whether to proceed with contralateral surgery or delay it to minimize the risk of bilateral vocal fold paralysis. The retrospective review analyzed 400 thyroidectomy cases, with 12.8% experiencing first-side signal loss. While outcomes varied based on whether the contralateral surgery was completed or aborted, most cases showed eventual recovery of vocal mobility. While findings suggest that immediate total thyroidectomy can be performed safely in select cases with low morbidity, the authors emphasize this approach should be reserved for highly selective situations. Decisions should weigh factors such as surgical complexity, symptom severity, and cancer invasiveness, favoring staged surgery for less pressing conditions.
Parathyroid
A case report from the University of Toronto describes one of the first successful deceased donor fresh tissue parathyroid allotransplants with immunosuppression in a transplant-naïve recipient. 23 The 40-year-old patient, who suffered severe long-term morbidity from hypoparathyroidism following thyroidectomy, underwent transplantation of four parathyroid glands into the brachioradialis muscle. Postoperatively, she achieved physiological normocalcemia, was weaned off calcium homeostatic medications by day 35, and experienced significant symptom resolution and mental health improvement. While the patient’s recovery highlights the potential of this approach, concerns about long-term immunosuppression risks remain and the authors suggest future efforts should focus on refining transplantation protocols, optimizing immunosuppressive regimens, and broadening the donor pool.
Concluding remarks
The authors thank the ATA for the honor of presenting the 2024 Year in Review lectures and hope that this summary article captures some of the notable advances that are emerging in the thyroid research field. The authors also acknowledge the outstanding contributions of many other investigators whose work has advanced the field of thyroid research but could not be included in this brief, selective review. Further exciting advances are anticipated in the coming year and will be highlighted at the upcoming International Thyroid Congress in Rio de Janeiro, Brazil (18–22 June) and the Annual Meeting of the American Thyroid Association in Scottsdale, Arizona, USA (10–14 September) in 2025.
Footnotes
Acknowledgments
The authors thank numerous colleagues for the discussion of current thyroid research and thank especially Joanna Klubo-Gwiezdzinska, Carmelo Nucera, Samuel Refetoff, Graham Williams, Nadia Schoenmakers, Anna Sawka, Kristien Boelaert, Megan Haymart, and Jennifer Kuo for additional, thoughtful comments.
Authors’ Contributions
D.F., J.A.S., and E.G.G. contributed equally to the creation of this article.
Author Disclosure Statement
D.F. and J.A.S. have no competing financial interests to disclose. E.G.G. oversees institutional research funding received by MD Anderson from Exelixis and Eli Lilly & Co.
Funding Information
This work was supported in part by the intramural research program at NIDDK at the National Institutes of Health (D.F.). J.A.S. and E.G.G. did not receive funding for this work.
