From Bench-to-Bedside: How Artificial Intelligence is Changing Thyroid Nodule Diagnostics, a Systematic Review.

CONTEXT: Use of artificial intelligence (AI) to predict clinical outcomes in thyroid nodule diagnostics has grown exponentially over the past decade. The greatest challenge is in understanding the best model to apply to one's own patient population, and how to operationalize such a model in practice. EVIDENCE ACQUISITION: A literature search of PubMed and IEEE Xplore was conducted for English-language publications between January 1, 2015 and January 1, 2023, studying diagnostic tests on suspected thyroid nodules that used AI. We excluded articles without prospective or external validation, nonprimary literature, duplicates, focused on nonnodular thyroid conditions, not using AI, and those incidentally using AI in support of an experimental diagnostic outside standard clinical practice. Quality was graded by Oxford level of evidence. EVIDENCE SYNTHESIS: A total of 61 studies were identified; all performed external validation, 16 studies were prospective, and 33 compared a model to physician prediction of ground truth. Statistical validation was reported in 50 papers. A diagnostic pipeline was abstracted, yielding 5 high-level outcomes: (1) nodule localization, (2) ultrasound (US) risk score, (3) molecular status, (4) malignancy, and (5) long-term prognosis. Seven prospective studies validated a single commercial AI; strengths included automating nodule feature assessment from US and assisting the physician in predicting malignancy risk, while weaknesses included automated margin prediction and interobserver variability. CONCLUSION: Models predominantly used US images to predict malignancy. Of 4 Food and Drug Administration-approved products, only S-Detect was extensively validated. Implementing an AI model locally requires data sanitization and revalidation to ensure appropriate clinical performance.

Outcomes of cytologically indeterminate thyroid nodules managed with Genomic Sequencing Classifier.

CONTEXT: Molecular testing can refine the risk of malignancy in thyroid nodules with indeterminate cytology to decrease unnecessary diagnostic surgery. OBJECTIVE: This study was performed to evaluate the outcomes of cytologically indeterminate thyroid nodules managed with Afirma genomic sequencing classifier (GSC) testing. DESIGN, SETTING, PATIENTS, AND INTERVENTION: Adult patients who underwent a biopsy at three major academic centers between July 2017 and June 2021 with Bethesda III or IV cytology were included. All patients had surgery or minimum follow-up of 1 year ultrasound surveillance. MAIN OUTCOME MEASURES: The primary outcomes were the sensitivity, specificity, PPV, and NPV of GSC in Bethesda III and IV nodules. RESULTS: The median nodule size of the 834 indeterminate nodules was 2.1 cm and the median follow-up was 23 months. GSC's sensitivity, specificity, PPV, and NPV across all institutions were 95%, 81%, 50%, and 99% for Bethesda III nodules and 94%, 82%, 65%, and 98% for Bethesda IV nodules, respectively. The overall false negative rate was 2%. The NPV of GSC in thyroid nodules with oncocytic predominance was 100% in Bethesda III nodules and 98% in Bethesda IV nodules. However, the PPV of oncocytic nodules was low (17% in Bethesda III nodules and 45% in Bethesda IV nodules). Only 22% of thyroid nodules with benign GSC results grew during surveillance. CONCLUSIONS: GSC is a key tool for managing patients with indeterminate cytology, including the higher-risk Bethesda IV category. GSC benign thyroid nodules can be observed similarly to thyroid nodules with benign cytology.

Extent of Surgery for Medullary Thyroid Cancer and Prevalence of Occult Contralateral Foci.

Importance: Standard treatment for patients with medullary thyroid cancer (MTC) consists of total thyroidectomy with central neck dissection, but the rationale for bilateral surgery in patients with unilateral disease on ultrasonography remains unclear. Objective: To determine the presence of occult contralateral disease (lesions not seen on preoperative ultrasonography) in patients with MTC as a rationale for total thyroidectomy. Design, Setting, and Participants: This multi-institutional, retrospective cohort study was conducted from September 1998 to April 2022 in academic medical centers and included patients with MTC who underwent thyroidectomy with preoperative imaging. Main Outcomes and Measures: The primary end point was the prevalence of sonographically occult foci of MTC in the contralateral lobe among patients with sporadic MTC. Results: The cohort comprised 176 patients with a median age at diagnosis of 55 years (range, 2-87 years), 69 (57.6%) of whom were female. Genetic testing was performed in 109 patients (61.9%), 48 (27.5%) of whom carried germline RET variants. Initial surgical management consisted of total thyroidectomy (161 [91.0%]), lobectomy followed by completion thyroidectomy (7 [4.0%]), and lobectomy alone (8 [4.5%]). Central and lateral neck dissections were performed as part of initial therapy for 146 patients (83.1%). In the entire cohort of 176 patients, 46 (26.0%) had contralateral foci disease and 9 (5.1%) had occult contralateral foci that were not identified on preoperative ultrasonography. Among 109 patients who underwent genetic testing, 38 (34.9%) had contralateral disease, 8 (7.3%) of whom had occult contralateral disease not seen on preoperative ultrasonography. Patients with sporadic MTC experienced a 95.7% reduction in the odds of having a focus of MTC in the contralateral lobe compared with patients with a germline RET variant (odds ratio, 0.043; 95% CI, 0.013-0.123). When adjusting for age, sex, tumor size, and lymph node involvement, the odds ratio of having contralateral MTC in patients with sporadic disease was 0.034 (95% CI, 0.007-0.116). Among patients who underwent lobectomy alone with postoperative calcitonin levels, 5 of 12 (41.7%) achieved undetectable calcitonin levels (<2.0 pg/mL; to convert to pmol/L, multiply by 0.292). Conclusions and Relevance: The results of this cohort study suggest that a staged approach involving initial thyroid lobectomy could be considered in patients with sporadic MTC and no contralateral ultrasonography findings, with no further surgery if calcitonin levels became undetectable. Further work using prospective randomized clinical trials to evaluate lobectomy as a biochemical cure in patients presenting with unilateral disease is warranted.

Diagnostic performance of molecular testing in indeterminate (Bethesda III and IV) thyroid nodules with Hürthle cell cytology.

BACKGROUND: Indeterminate thyroid nodules with Hürthle cell cytology remain a diagnostic challenge. The low benign call rate and positive predictive value of first-generation molecular tests precluded their use to rule out malignancy. We examined the diagnostic performance of current tests. METHOD: This subset analysis of our prospective randomized trial compared the benign call rate and positive predictive value of Afirma Gene Sequencing Classifier and Thyroseq v3 in Bethesda III and IV nodules with Hürthle cell cytology. Molecular test samples were obtained at initial fine-needle aspiration (8/2017-7/2022) and reflexively sent for processing. RESULTS: Molecular testing was performed on 140 Hürthle cell nodules. Of 79 nodules tested with the Afirma Gene Sequencing Classifier, the benign call rate was 84% (66/79). Nine of 66 nodules with benign results were resected, with no malignancies. Twelve of 13 nodules with suspicious results were resected, revealing 3 malignancies-2 papillary thyroid carcinomas and one Hürthle cell carcinoma (positive predictive value 25%). Of 61 nodules tested with Thyroseq v3, the benign call rate was 56% (34/61; (P < .01 versus Afirma Gene Sequencing Classifier). Five of 34 nodules with negative results were resected, with no malignancies. Nineteen of 27 nodules with positive results were resected, revealing 3 malignancies-2 papillary thyroid carcinomas and 1 Hürthle cell carcinoma (positive predictive value 16%). CONCLUSION: The high benign call rate of current molecular tests in Hürthle cell nodules strengthens their value in enabling patients to avoid surgery.

Histopathology of telomerase reverse transcriptase promoter (TERT) mutated indeterminate thyroid nodules.

Objective: The objective of this study was to analyze the risk of malignancy and the histopathology of telomerase reverse transcriptase promoter (TERT) mutated cytologically indeterminate thyroid nodules (ITN). Methods: A PUBMED search of molecularly tested ITN was conducted and data on TERT mutated ITN with histopathology correlation were extracted. Results: Twenty-six manuscripts (published between 2014 and 2022) reported on 77 TERT mutated ITN. Sixty-five nodules were malignant (84 %), with 16 (25 %) described with high-risk histopathology, 5 (8 %) described as low-risk, and most without any description. Isolated TERT mutations were malignant in 26/30 ITNs (87 %) with 9 (35 %) described as high risk and none described as low risk. TERT + RAS mutated ITNs were malignant in 29/34 ITNs (85 %) with 3 (10 %) described as high risk and 4 (14 %) described as low risk. Finally, all 5 TERT + BRAFV600E mutated nodules were malignant and 3/5 (60 %) were described as high risk. Conclusion: TERT mutated ITNs have a high risk of malignancy (84 %), and the current data does not show a difference in malignancy rate between isolated TERT mutations and TERT + RAS co-mutated ITNs. When described, TERT + RAS co-mutated ITNs did not have a higher rate of high-risk histopathology as compared to isolated TERT mutated lesions. Most TERT mutated ITNs did not have a description of histopathology risk and the oncologic outcomes, including rate of recurrence, metastases, and disease specific survival, are unknown. Further data is needed to determine if TERT mutated ITNs should be subjected to aggressive initial treatment.

Rural-Urban Disparities in the Continuum of Thyroid Cancer Care: Analysis of 92,794 Cases.

Objective: Rurality is associated with higher incidence and higher disease-specific mortality for most cancers. Outcomes for rural and ultrarural ("frontier") patients with thyroid cancer are poorly understood. This study aimed to identify actionable deficits in thyroid cancer outcomes for rural patients. Methods: We queried linked California Cancer Registry and California Office of Statewide Health Planning and Development databases for patients diagnosed with thyroid cancer (1999-2017). We analyzed time from disease stage at diagnosis, time from diagnosis to surgery, receipt of appropriate radioactive iodine ablation, surveillance status, and overall and disease-specific mortality for urban, rural, and frontier patients. Cox and logistic regression models controlled for clinical and demographic covariates a stepwise manner. All incidence figures are expressed as a proportion of newly diagnosed cases. Results: Our cohort comprised 92,794 subjects: (65,475 women [70.6%]; mean age 50.0 years). Compared to urban patients, rural and frontier patients were more likely to be American Indian, White, uninsured, and from lower quintiles of socioeconomic status (p < 0.01). Distant disease at diagnosis was more common in rural (56.0 vs. 50.4 cases per 1000 new cases, p < 0.01) and frontier patients (80.9 vs. 50.4 per 1000, p < 0.01) compared to urban patients. The incidence of medullary thyroid cancer was greater in rural patients (17.9 vs. 13.6 cases per 1000, p < 0.01) and frontier patients (31.0 vs. 13.6 per 1000, p < 0.01) compared to urban patients. The incidence of anaplastic thyroid cancer was higher in frontier versus urban patients (15.5 vs. 7.1 per 1000, p < 0.01). When compared to urban patients, rural and frontier patients were more often lost to follow-up (odds ratio [OR] 1.69 [confidence interval, CI 1.54-1.85], and OR 3.03 [CI 1.89-5.26], respectively) and had higher disease-specific mortality (OR 1.18 [CI 1.07-1.30], and OR 1.92 [CI 1.22-2.77], respectively). Rural and frontier residence was independently associated with being lost to follow-up, suggesting that it is a key driver of disparities. Conclusion: Compared to their urban counterparts, rural and frontier patients with thyroid cancer present with later-stage disease and experience higher disease-specific mortality. They also are more often lost to follow-up, which presents an opportunity for targeted outreach to reduce the observed disparities in outcomes.

Disparities in Initial Thyroid Cancer Care by Hospital Treatment Volume: Analysis of 52,599 Cases in California.

Background: Racially minoritized patients with thyroid cancer are less likely to receive high-quality and guideline-concordant care. Inaccessibility of high-volume centers may contribute to inequalities in thyroid cancer outcomes. This study sought to understand the extent to which access to higher volume thyroid cancer centers is associated with patient outcomes. Methods: We queried linked California Cancer Registry and California Office of Statewide Health Planning and Development databases for thyroid cancer patients who received thyroid surgery between 1999 and 2017. Hospitals were stratified by their median annual volume of thyroid cancer operations: ultra-low volume (0-5 cases/year), low-volume (6-25 cases/year), mid-volume (26-50 cases/year), and high-volume (>50 cases/year). We analyzed the rates of complications, rates of reoperation for cancer recurrence, use of radioactive iodine (131I), and mortality by median hospital volume of thyroid surgery. A multivariable regression controlled for high-risk tumor features. Differences in access by center volume were assessed based on patient demographics. Results: We studied 52,599 thyroid cancer patients who underwent thyroidectomy. Patients who underwent thyroidectomy at ultra-low volume centers were more likely to undergo reoperations for recurrent/persistent disease compared with patients at low- (odds ratio [OR] 1.17 [CI 1.02-1.35]), mid- (OR 1.25 [CI 1.06-1.46]), and high-volume centers (OR 1.26 [CI 1.03-1.56]). Patients who received thyroid operations at ultra-low volume centers were also less likely to receive guideline-concordant 131I ablation compared with patients at higher volume centers (OR 0.77 [CI 0.72-0.82]). A pair-wise comparison between all volume categories for all outcomes revealed no statistically significant differences in outcomes between low-, mid-, or high-volume centers. Only ultra-low volume centers had significantly higher rates of adverse outcomes. Ultra-low volume centers were disproportionately accessed by women (p < 0.05), Hispanic, Asian/Pacific Islander, and American Indian people (p < 0.01), those from the lowest three quintiles of socio-economic status (p < 0.01), and the uninsured and those on Medicaid or Medicare (p < 0.01) when compared with higher volume centers. Conclusions: Patients receiving thyroid cancer surgery at centers performing ≤5 such operations per year were more likely to require reoperation for recurrent/persistent disease and less likely to receive appropriate 131I ablation. Ultra-low volume centers served higher proportions of socially and economically marginalized communities.

Prognostic Value of Preoperative Molecular Testing and Implications for Initial Surgical Management in Thyroid Nodules Harboring Suspected (Bethesda V) or Known (Bethesda VI) Papillary Thyroid Cancer.

Importance: Molecular testing is commonly used in the diagnosis of thyroid nodules with indeterminate cytology. The role of molecular testing in prognosticating oncologic outcomes in thyroid nodules with suspicious or malignant cytology is unclear. Objective: To determine whether molecular profiling of Bethesda V (suspicious for thyroid cancer) and VI (thyroid cancer) nodules is associated with improved prognostication and whether it may inform initial treatment. Design, Setting, and Participants: This retrospective cohort study included consecutive patients with Bethesda V or VI nodules who underwent surgery, with histopathology showing differentiated thyroid cancer, between May 1, 2016, and July 31, 2019 in the University of California, Los Angeles health system. Data were analyzed between April 2, 2021, and January 18, 2023. Exposures: Masked ThyroSeq, version 3 molecular analysis after completion of initial treatment and acquisition of follow-up data. Main Outcomes and Measures: Structural disease persistence or recurrence, distant metastasis, and recurrence-free survival were assessed using ThyroSeq Cancer Risk Classifier (CRC) molecular risk groups (low, RAS-like; intermediate, BRAF-like; high, combination of BRAF/RAS plus TERT or other high-risk alterations) using Cox proportional hazards regression models. Results: In 105 patients with papillary thyroid cancer (median [IQR] follow-up, 3.8 [3.0-4.7] years), ThyroSeq identified genomic alterations in 100 (95%) samples (6 [6%] low risk, 88 [88%] intermediate risk, and 6 [6%] high risk; median [IQR] age, 44 [34-56] years; 68 [68%] female and 32 [32%] male). No patients with low-risk or negative results experienced recurrence. Of the 88 patients with intermediate risk, 6 (7%) experienced local recurrence, with 1 of them also developing distant metastasis. The 6 patients with high risk (all with BRAF V600E plus TERT mutation) underwent total thyroidectomy followed by radioactive iodine (RAI) ablation. Four patients with high risk (67%) experienced local recurrence, with 3 of them also developing distant metastasis. Thus, patients with high-risk alterations were more likely to experience persistence or recurrence and distant metastasis than patients with intermediate risk. In a multivariable analysis incorporating patient age, sex, cancer size, ThyroSeq molecular risk group, extrathyroidal extension, lymph node positivity, American Thyroid Association risk, and RAI ablation, only cancer size (hazard ratio, 1.36; 95% CI, 1.02-1.80) and ThyroSeq CRC molecular risk group (high vs intermediate and low: hazard ratio, 6.22; 95% CI, 1.04-37.36) were associated with structural recurrence. Conclusions and Relevance: Among the 6% of patients with high-risk ThyroSeq CRC alterations in this cohort study, the majority experienced recurrence or distant metastasis despite initial treatment with total thyroidectomy and RAI ablation. In contrast, patients with low- and intermediate-risk alterations had a low recurrence rate. Preoperative knowledge of molecular alteration status at diagnosis may allow for deescalation of initial surgery and refining of the intensity of postoperative surveillance in patients presenting with Bethesda V and VI thyroid nodules.

Patient Preference in Physician Decision-Making for Patients With Low- to Intermediate-Risk Differentiated Thyroid Cancer.

This survey study describes the association between patient preference and physician decision-making in thyroid cancer.

Bethesda III and IV Thyroid Nodules Managed Nonoperatively After Molecular Testing With Afirma GSC or Thyroseq v3.

CONTEXT: Molecular testing has improved risk stratification and increased nonoperative management for patients with indeterminate thyroid nodules, but data on the long-term outcomes of current molecular tests Afirma Gene Sequencing Classifier (GSC) and Thyroseq v3 are limited. OBJECTIVE: To determine the rate of delayed operation and the false negative rate of the Afirma GSC and Thyroseq v3 in Bethesda III and IV thyroid nodules. METHODS: Prospective follow-up of a single center, randomized, clinical trial comparing the performance of Afirma GSC and Thyroseq v3 in the diagnosis of indeterminate thyroid nodules at the University of California, Los Angeles (UCLA). Consecutive participants who underwent thyroid biopsy in the UCLA health system with Bethesda III and IV cytology from August 2017 to November 2019. The main outcome measure was false negative rate of molecular testing. RESULTS: Of 176 indeterminate nodules with negative or benign molecular test results, 14 (8%) nodules underwent immediate resection, with no malignancies found on surgical pathology. Nonoperative management with active surveillance was pursued for 162 (92%) nodules with benign or negative test results. The median surveillance was 34 months (range 12-60 months), and 44 patients were lost to follow-up. Of 15 nodules resected during surveillance, 1 malignancy was found (overall false negative rate of 0.6%). This was a 2.7 cm minimally invasive Hurthle cell carcinoma that initially tested negative with Thyroseq v3 and underwent delayed resection due to sonographic growth during surveillance. CONCLUSIONS: The majority of Bethesda III/IV thyroid nodules with negative or benign molecular test results are stable over 3 years of follow-up. These findings support the high sensitivity of current molecular tests and their role in ruling out malignancy in indeterminate thyroid nodules.

Racial/ethnic disparities in thyroid cancer in California, 1999-2017.

BACKGROUND: Recent studies examining treatment disparities in thyroid cancer care found that appropriate use of surgery and radioactive iodine may be improving over time. METHODS: California Cancer Registry and California Office of Statewide Health Planning and Development data was evaluated for the effect of race on overall and disease-specific survival for thyroid cancer in California (1999-2017). Reoperation data was also examined. We hypothesized treatment and outcome disparities would persist between Black and white patients. RESULTS: Black patients with thyroid cancer had worse overall survival than white patients (p < 0.01). No difference was found in disease-specific survival between Black and white patients after adjusting for SES and health insurance status. Black patients underwent reoperation less frequently (1.4%) (HR = 0.70 [CI, 0.50-0.99], p < 0.05) than white patients (2.0%). CONCLUSIONS: SES and insurance type are drivers of disparities in thyroid cancer survival in Black patients. Addressing social determinants of health or healthcare access are paramount to addressing disparities in thyroid cancer between Black and white patients.

Surgical Aspects of Primary Hyperparathyroidism.

Parathyroidectomy (PTX) is the treatment of choice for symptomatic primary hyperparathyroidism (PHPT). It is also the treatment of choice in asymptomatic PHPT with evidence for target organ involvement. This review updates surgical aspects of PHPT and proposes the following definitions based on international expert consensus: selective PTX (and reasons for conversion to an extended procedure), bilateral neck exploration for non-localized or multigland disease, subtotal PTX, total PTX with immediate or delayed autotransplantation, and transcervical thymectomy and extended en bloc PTX for parathyroid carcinoma. The systematic literature reviews discussed covered (i) the use of intraoperative PTH (ioPTH) for localized single-gland disease and (ii) the management of low BMD after PTX. Updates based on prospective observational studies are presented concerning PTX for multigland disease and hereditary PHPT syndromes, histopathology, intraoperative adjuncts, localization techniques, perioperative management, "reoperative" surgery and volume/outcome data. Postoperative complications are few and uncommon (<3%) in centers performing over 40 PTXs per year. This review is the first global consensus about surgery in PHPT and reflects the current practice in leading endocrine surgery units worldwide. © 2022 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Preoperative Identification of Medullary Thyroid Carcinoma (MTC): Clinical Validation of the Afirma MTC RNA-Sequencing Classifier.

Background: Cytopathological evaluation of thyroid fine-needle aspiration biopsy (FNAB) specimens can fail to raise preoperative suspicion of medullary thyroid carcinoma (MTC). The Afirma RNA-sequencing MTC classifier identifies MTC among FNA samples that are cytologically indeterminate, suspicious, or malignant (Bethesda categories III-VI). In this study we report the development and clinical performance of this MTC classifier. Methods: Algorithm training was performed with a set of 483 FNAB specimens (21 MTC and 462 non-MTC). A support vector machine classifier was developed using 108 differentially expressed genes, which includes the 5 genes in the prior Afirma microarray-based MTC cassette. Results: The final MTC classifier was blindly tested on 211 preoperative FNAB specimens with subsequent surgical pathology, including 21 MTC and 190 non-MTC specimens from benign and malignant thyroid nodules independent from those used in training. The classifier had 100% sensitivity (21/21 MTC FNAB specimens correctly called positive; 95% confidence interval [CI] = 83.9-100%) and 100% specificity (190/190 non-MTC FNAs correctly called negative; CI = 98.1-100%). All positive samples had pathological confirmation of MTC, while all negative samples were negative for MTC on surgical pathology. Conclusions: The RNA-sequencing MTC classifier accurately identified MTC from preoperative thyroid nodule FNAB specimens in an independent validation cohort. This identification may facilitate an MTC-specific preoperative evaluation and resulting treatment.

Comparison of Preoperative Alpha-blockade for Resection of Paraganglioma and Pheochromocytoma.

OBJECTIVE: Phenoxybenzamine (nonselective, noncompetitive alpha-blocker) is the preferred drug for preoperative treatment of pheochromocytoma, but doxazosin (selective, competitive alpha-blocker) may be equally effective. We compared the efficacy of doxazosin vs phenoxybenzamine. METHODS: We conducted a prospective study of patients undergoing pheochromocytoma or paraganglioma resection by randomizing pretreatment with phenoxybenzamine or doxazosin at a single tertiary referral center. The high cost of phenoxybenzamine led to high crossover to doxazosin. Randomization was halted, and a consecutive historical cohort of phenoxybenzamine patients was included for a case-control study design. The efficacy of alpha-blockade was assessed with preinduction infusion of incremental doses of phenylephrine. The primary outcomes were mortality, cardiovascular complications, and intensive care unit admission. The secondary outcomes were hemodynamic instability index (proportion of operation outside of hemodynamic goals), adequacy of blockade by the phenylephrine titration test, and drug costs. RESULTS: Twenty-four patients were prospectively enrolled (doxazosin, n = 20; phenoxybenzamine, n = 4), and 15 historical patients treated with phenoxybenzamine were added (total phenoxybenzamine, n = 19). No major cardiovascular complications occurred in either group. The phenylephrine dose-response curves showed less blood pressure rise in the phenoxybenzamine than in the doxazosin group (linear regression coefficient = 0.008 vs 0.018, P = .01), suggesting better alpha-blockade in the phenoxybenzamine group. The median hemodynamic instability index was 14% vs 13% in the phenoxybenzamine and doxazosin groups, respectively (P = .56). The median highest daily cost of phenoxybenzamine was $442.20 compared to $5.06 for doxazosin. CONCLUSION: Phenoxybenzamine may blunt intraoperative hypertension better than doxazosin, but this difference did not translate to fewer cardiovascular complications and is offset by a considerably increased cost.

Disparities in Thyroid Care.

Thyroid disease affects an estimated 20 million Americans, with 1 in 8 women developing a thyroid disorder during her lifetime. Although most patients with thyroid cancer have a good prognosis and effective treatments for benign thyroid disease are available, disparities exist in thyroid care and result in worse outcomes for racial and ethnic minorities. Inequities in the diagnosis and treatment of thyroid disease are due to the complex interplay of systems-, physician-, and patient-level factors. Thus, innovative strategies that take an ecological approach to addressing racial disparities are needed to achieve equitable care for all patients with thyroid disease.