# Thyroid Nodules and Thyroid Cancer

## Thyroid Nodules - Epidemiology and Evaluation

### Prevalence

Thyroid nodules are extraordinarily common, with their apparent prevalence varying dramatically depending on the method of detection. Palpable nodules are found in 4 to 7% of adults on physical examination. However, when thyroid ultrasound is used for detection, the prevalence rises to 20 to 76% depending on the population studied and the resolution of the imaging equipment. The increasing use of cross-sectional imaging, including CT, MRI, and PET scanning, has further expanded the detection of thyroid "incidentalomas." Despite this high prevalence, the malignancy risk in clinically detected nodules is only 5 to 15%, meaning the vast majority are benign entities including colloid nodules, follicular adenomas, cysts, and foci of thyroiditis.

### Initial Evaluation

#### History and Physical

The clinical evaluation of a thyroid nodule begins with a careful assessment of risk factors for malignancy. These include a history of childhood head and neck radiation, family history of thyroid cancer (particularly medullary thyroid carcinoma or MEN2 syndromes), rapid nodule growth, new hoarseness suggesting recurrent laryngeal nerve involvement, dysphagia, a firm or fixed nodule on examination, and the presence of cervical lymphadenopathy. Most thyroid nodules are asymptomatic, with large nodules potentially causing compressive symptoms. Age at the extremes, either below 14 or above 70 years, is associated with a higher malignancy rate, as is male sex, which carries a slightly higher malignancy rate per nodule.

#### Laboratory

TSH is the essential first-line laboratory test for all thyroid nodules. A low or suppressed TSH should prompt thyroid scintigraphy with I-123 or Tc-99m pertechnetate to determine whether the nodule is "hot" (autonomously functioning). Hot nodules carry a very low malignancy risk of less than 1 to 3%, and FNA is generally not indicated. When TSH is normal or elevated, the evaluation proceeds with ultrasound characterization, with the understanding that an elevated TSH is itself an independent risk factor for malignancy within a nodule.

The role of routine calcitonin measurement remains controversial. The ATA does not recommend universal calcitonin screening, while European guidelines suggest measuring it in the initial workup. An elevated calcitonin above 100 pg/mL is strongly suggestive of medullary thyroid carcinoma. Thyroglobulin measurement is not useful for initial nodule evaluation because it is nonspecific, being elevated in virtually any thyroid pathology.

### Thyroid Ultrasound - Risk Stratification

#### ATA Sonographic Pattern Classification (2015)

The 2015 ATA guidelines introduced a structured sonographic risk stratification system that categorizes thyroid nodules based on their ultrasound appearance and assigns corresponding malignancy risk estimates and FNA thresholds. High-suspicion nodules, carrying a malignancy risk exceeding 70 to 90%, are solid hypoechoic nodules with one or more of the following features: irregular margins, microcalcifications, taller-than-wide shape on transverse view, extrathyroidal extension, or associated suspicious lymph nodes. FNA is recommended for high-suspicion nodules 1 cm or larger.

Intermediate-suspicion nodules, with a malignancy risk of 10 to 20%, are solid hypoechoic without the additional high-suspicion features, and FNA is recommended at 1 cm or larger. Low-suspicion nodules (5 to 10% risk) include isoechoic or hyperechoic solid nodules and partially cystic nodules with an eccentric solid component, with FNA recommended at 1.5 cm or larger. Very-low-suspicion nodules (less than 3% risk) include spongiform nodules (predominantly cystic with more than 50% cystic content) and partially cystic nodules without suspicious solid components, with FNA considered at 2 cm or larger, or observation. Purely cystic nodules have a malignancy risk of less than 1% and do not require FNA.

#### ACR TI-RADS (Thyroid Imaging Reporting and Data System)

The ACR TI-RADS system provides a complementary points-based approach, scoring composition, echogenicity, shape, margin characteristics, and echogenic foci. Scores are summed and translated into risk categories from TR1 (benign, 0 points, no FNA needed) through TR5 (highly suspicious, 7 or more points, FNA at 1 cm or larger). This system offers a more granular assessment that can improve interobserver agreement.

<image>A comparison panel showing thyroid ultrasound patterns with corresponding risk categories. Five panels arranged horizontally: (1) High suspicion - solid hypoechoic nodule with irregular margins, microcalcifications shown as bright punctate echogenic foci, and taller-than-wide shape; (2) Intermediate suspicion - solid hypoechoic nodule with smooth margins, no calcifications; (3) Low suspicion - isoechoic solid nodule with smooth borders; (4) Very low suspicion - spongiform nodule with multiple small cystic spaces giving honeycomb appearance; (5) Benign - purely cystic anechoic nodule with posterior acoustic enhancement. Below each panel, show the ATA risk category, approximate malignancy risk percentage, and FNA size threshold. Use grayscale ultrasound appearance with labeled annotations.</image>

## Fine Needle Aspiration (FNA) Cytology

### Technique

Ultrasound-guided FNA is preferred over palpation-guided aspiration, as it improves diagnostic yield and reduces the non-diagnostic rate. A 25- to 27-gauge needle is used, with 2 to 4 passes typically providing adequate cellular material. On-site cytopathology assessment, known as rapid on-site evaluation (ROSE), further reduces non-diagnostic rates by allowing immediate assessment of specimen adequacy. For cystic nodules, the fluid should be aspirated and sent for cytology, with the solid component targeted separately for sampling.

### Bethesda System for Reporting Thyroid Cytopathology (2017 Revision)

| Bethesda Category | Diagnosis | Malignancy Risk | Recommended Management |
|---|---|---|---|
| I | Non-diagnostic / Unsatisfactory | 5-10% | Repeat FNA |
| II | Benign | 0-3% | Ultrasound surveillance |
| III | AUS / FLUS | 10-30% | Repeat FNA, molecular testing, or diagnostic lobectomy |
| IV | Follicular neoplasm / Suspicious for FN | 25-40% | Molecular testing or lobectomy |
| V | Suspicious for malignancy | 50-75% | Lobectomy or total thyroidectomy |
| VI | Malignant | 97-99% | Total thyroidectomy (or lobectomy for PTC <4 cm) |

The Bethesda system provides a standardized framework for reporting thyroid FNA results, assigning each specimen to one of six categories with associated malignancy risks and management recommendations. Category I (non-diagnostic or unsatisfactory) carries a 5 to 10% malignancy risk and warrants repeat FNA. Category II (benign) has a 0 to 3% malignancy risk and is followed with ultrasound surveillance. Category III (atypia of undetermined significance, AUS, or follicular lesion of undetermined significance, FLUS) carries a 10 to 30% malignancy risk and can be managed with repeat FNA, molecular testing, or diagnostic lobectomy. Category IV (follicular neoplasm or suspicious for follicular neoplasm) has a 25 to 40% malignancy risk and is addressed with molecular testing or lobectomy. Category V (suspicious for malignancy) carries a 50 to 75% risk and typically leads to lobectomy or total thyroidectomy. Category VI (malignant) has a 97 to 99% risk and is managed with total thyroidectomy or, for smaller papillary carcinomas under 4 cm, lobectomy.

### Molecular Testing for Indeterminate Nodules (Bethesda III/IV)

Molecular testing has transformed the management of indeterminate thyroid nodules by reducing the rate of diagnostic surgery from approximately 60% to 30%. The Afirma Gene Sequencing Classifier (GSC) functions primarily as a rule-out test, with a "benign" result carrying a negative predictive value exceeding 95%, thereby avoiding unnecessary surgery in the majority of patients with benign disease. ThyroSeq v3 serves as both a rule-in and rule-out test, employing a next-generation sequencing panel of 112 genes to achieve a benign NPV of approximately 97% and a malignant positive predictive value of approximately 66%. It identifies specific driver mutations including BRAF, RAS, RET/PTC, PAX8/PPARgamma, and TERT promoter mutations. The ThyGeNEXT/ThyraMIR system combines a mutation panel with a microRNA classifier.

| Molecular Marker | Associated Cancer | PPV for Malignancy | Clinical Significance |
|---|---|---|---|
| BRAF V600E | Papillary thyroid carcinoma | >99% | Virtually diagnostic; associated with aggressive features |
| RAS (NRAS, HRAS, KRAS) | Follicular or papillary carcinoma (and adenoma) | 30-40% | Found in both benign and malignant lesions |
| RET/PTC rearrangements | Papillary (radiation-related) | High | Common in children and radiation-exposed patients |
| PAX8/PPARgamma fusions | Follicular carcinoma | Moderate-high | Characteristic of FTC |
| TERT promoter mutations | Aggressive DTC | High | Synergistic with BRAF V600E for poor prognosis |

Among the key mutations detected, BRAF V600E is highly specific for papillary thyroid carcinoma with a positive predictive value exceeding 99% and is associated with aggressive histological features, making it virtually diagnostic of malignancy. RAS mutations (NRAS, HRAS, KRAS) are found in both follicular adenomas and follicular or papillary carcinomas, carrying a 30 to 40% malignancy risk. RET/PTC rearrangements are associated with radiation-exposure-related papillary carcinoma and are common in children. PAX8/PPARgamma fusions characterize follicular carcinoma. TERT promoter mutations, particularly when co-occurring with BRAF V600E, are associated with aggressive behavior, distant metastases, and reduced survival.

## Thyroid Cancer Classification

### Differentiated Thyroid Cancer (DTC) - 95% of thyroid cancers

#### Papillary Thyroid Carcinoma (PTC) - 85%

Papillary thyroid carcinoma is the most common thyroid malignancy and carries an excellent prognosis, with greater than 98% 10-year survival for localized disease. Histologically, PTC is characterized by papillary architecture and distinctive nuclear features including the optically clear "Orphan Annie eye" nuclei, nuclear grooves, and intranuclear pseudoinclusions, along with psammoma bodies (laminated calcified structures). PTC spreads preferentially via lymphatics to cervical lymph nodes, with nodal metastases present in 30 to 80% of cases at diagnosis. Distant metastasis is uncommon, occurring in only 2 to 5%, with the lung being the most common site.

Multiple histological variants exist with varying prognostic implications. Classic PTC has the most common and favorable prognosis. The follicular variant of PTC (FVPTC) is the second most common, and importantly, the encapsulated subtype with RAS mutations has been reclassified as NIFTP (Non-Invasive Follicular Thyroid Neoplasm with Papillary-Like Nuclear Features), an essentially benign entity curable by lobectomy alone without RAI or TSH suppression. The tall cell variant, defined by more than 50% of cells with height at least three times their width, almost universally harbors BRAF V600E and behaves more aggressively with higher recurrence rates. The hobnail and micropapillary variants carry elevated risks of distant metastasis. The diffuse sclerosing variant occurs in younger patients with bilateral, diffuse involvement and extensive lymph node involvement. The columnar cell variant is aggressive with a poor prognosis.

Papillary thyroid microcarcinoma (PTMC), defined as PTC measuring 1 cm or less, has very low risk and may be managed with active surveillance rather than surgery in select cases. Landmark data from Kuma Hospital in Japan, following 1,235 patients for more than 10 years, demonstrated that only 8% of observed PTMCs grew by more than 3 mm and only 3.8% developed lymph node metastasis, supporting the safety of observation.

#### Follicular Thyroid Carcinoma (FTC) - 10%

Follicular thyroid carcinoma cannot be diagnosed by FNA alone, as the distinction between follicular adenoma and carcinoma requires histological demonstration of capsular or vascular invasion. Unlike PTC, FTC spreads primarily via the hematogenous route to bone and lung, with lymph node metastasis being less common. Minimally invasive FTC, showing only capsular invasion, has an excellent prognosis with 98% 10-year survival. Widely invasive FTC, with extensive capsular and vascular invasion, is considerably more aggressive with 50 to 70% 10-year survival. Hurthle cell carcinoma, now classified as a separate entity termed oncocytic thyroid carcinoma, tends to be less RAI-avid and more aggressive than conventional FTC.

### Medullary Thyroid Carcinoma (MTC) - 3-5%

Medullary thyroid carcinoma arises from the parafollicular C cells and produces calcitonin and CEA as tumor markers. Approximately 75% of cases are sporadic, while 25% are hereditary, associated with MEN2A, MEN2B, or familial MTC, all driven by germline mutations in the RET proto-oncogene. Calcitonin is a sensitive tumor marker, with its doubling time serving as an important prognostic indicator. A progressive rise in CEA with a declining calcitonin-to-CEA ratio suggests dedifferentiation and poorer prognosis.

MTC does not respond to RAI because C cells do not concentrate iodine. Surgical management requires total thyroidectomy with at minimum a central neck dissection, and lateral neck dissection when clinically involved nodes are present. For advanced or metastatic disease, targeted therapies have markedly improved outcomes, including the multi-kinase inhibitors vandetanib and cabozantinib, and the highly selective RET inhibitors selpercatinib and pralsetinib. RET germline testing is mandatory for all MTC patients, as a positive result necessitates family screening and may require prophylactic thyroidectomy in mutation carriers.

### Anaplastic Thyroid Carcinoma (ATC) - 1-2%

Anaplastic thyroid carcinoma is the most aggressive thyroid malignancy, with a median survival of only 3 to 6 months and a nearly uniformly fatal course. It often arises from pre-existing differentiated thyroid cancer through dedifferentiation, with BRAF V600E and TERT promoter mutations frequently co-occurring. The clinical presentation is a rapidly enlarging, fixed neck mass causing compressive symptoms. Treatment is multimodal, combining surgery (when resectable), radiation, and chemotherapy.

A major therapeutic advance has been the identification that BRAF V600E-mutated ATC responds to the combination of dabrafenib (BRAF inhibitor) and trametinib (MEK inhibitor), with a response rate of approximately 69% in the ROAR basket trial. This combination received FDA approval and has significantly improved outcomes in this subset of patients. Immune checkpoint inhibitors represent an emerging therapeutic avenue, as some ATC tumors demonstrate high tumor mutational burden and PD-L1 expression.

<image>A comprehensive classification diagram of thyroid cancer types arranged as a hierarchical tree. Main trunk divides into: Differentiated (DTC), Medullary (MTC), and Anaplastic (ATC). DTC branches into Papillary (85%) and Follicular (10%), with sub-branches showing key variants and their distinguishing features. For each cancer type, include boxes showing: cell of origin, key molecular alterations, typical spread pattern (lymphatic vs hematogenous), RAI sensitivity, prognosis (5-year survival), and key tumor markers. Include Hurthle cell carcinoma as a separate branch. Use color coding: green for excellent prognosis, yellow for intermediate, red for poor. Include NIFTP as a reclassified benign entity connected to FVPTC with a special notation.</image>

## Management of Differentiated Thyroid Cancer

### Surgery

The surgical approach is determined by the tumor's risk profile. Lobectomy is appropriate for low-risk patients with unifocal tumors between 1 and 4 cm without extrathyroidal extension, clinically negative nodes, no vascular invasion, no aggressive histological variant, and a favorable molecular profile. Total thyroidectomy is indicated for tumors larger than 4 cm, bilateral disease, extrathyroidal extension, clinically positive nodes, distant metastases, planned RAI therapy, prior head and neck radiation, and strong family history. Central neck dissection is performed therapeutically for clinically involved nodes, while prophylactic central dissection remains controversial. Lateral neck dissection is performed for biopsy-proven metastatic lateral cervical nodes.

### ATA Risk Stratification (Initial Risk of Recurrence)

| ATA Risk Category | Recurrence Risk | Features | TSH Target |
|---|---|---|---|
| Low | <5% | Intrathyroidal PTC/FVPTC; ≤5 small LN mets (<0.2 cm); no vascular invasion; N0 | 0.5-2.0 mIU/L |
| Intermediate | 5-20% | Aggressive histology; minor ETE; vascular invasion; >5 LNs (0.2-3 cm); BRAF V600E | 0.1-0.5 mIU/L |
| High | >20% | Gross ETE; incomplete resection; distant mets; LN >3 cm; FTC with extensive vascular invasion (>4 foci) | <0.1 mIU/L |

The ATA risk stratification system categorizes patients into three groups based on initial risk of recurrence. Low-risk patients (less than 5% recurrence) have intrathyroidal PTC or FVPTC with 5 or fewer small (less than 0.2 cm) lymph node metastases, no vascular invasion, and clinically N0 disease. Intermediate-risk patients (5 to 20% recurrence) demonstrate aggressive histology, minor extrathyroidal extension, vascular invasion, more than 5 lymph nodes measuring 0.2 to 3 cm, RAI-avid disease in the neck outside the thyroid bed, or BRAF V600E mutation. High-risk patients (more than 20% recurrence) have gross extrathyroidal extension, incomplete resection, distant metastases, postoperative thyroglobulin suggestive of distant metastases, lymph nodes larger than 3 cm, or FTC with extensive vascular invasion (more than 4 foci).

### Radioactive Iodine (RAI) Therapy

The use of RAI has been significantly refined in recent years, with a notable shift away from routine administration in low-risk patients. For low-risk DTC, RAI is not routinely recommended per the 2015 ATA guidelines. For intermediate-risk patients, RAI adjuvant therapy at 30 to 150 mCi is considered on an individualized basis. For high-risk patients, RAI at 100 to 200 mCi is recommended.

RAI preparation requires TSH stimulation above 30 mIU/L, achieved either by thyroid hormone withdrawal for 4 to 6 weeks or by administration of recombinant TSH (Thyrogen, 0.9 mg intramuscularly for 2 days). A low-iodine diet providing less than 50 mcg iodine daily for 1 to 2 weeks before RAI maximizes iodine uptake. The HiLo trial demonstrated non-inferiority of low-dose RAI (30 mCi) compared to high-dose (100 mCi) for low-risk DTC with rhTSH preparation, and the ESTIMABL1 trial confirmed similar findings. A post-therapy whole-body scan obtained 5 to 7 days after RAI assesses disease extent.

Complications of RAI include sialadenitis (managed with sour candy or lemon drops to promote salivary flow), transient neck pain from radiation thyroiditis, nausea, bone marrow suppression at high cumulative doses, secondary malignancy risk (leukemia in approximately 1% with cumulative doses exceeding 600 mCi), and pulmonary fibrosis in patients with diffuse lung metastases receiving high cumulative doses.

### TSH Suppression Therapy

TSH suppression is rationale-based, as TSH functions as a growth factor for DTC cells. For high-risk patients, TSH should be suppressed below 0.1 mIU/L indefinitely until risk is reclassified. For intermediate-risk patients with incomplete response, the target is 0.1 to 0.5 mIU/L. For low-risk patients or those with an excellent treatment response, a relaxed target of 0.5 to 2.0 mIU/L is appropriate, avoiding the complications of chronic TSH suppression including atrial fibrillation (relative risk 3.1 in patients over 60), osteoporosis in postmenopausal women, anxiety, and insomnia.

### Dynamic Risk Stratification (Ongoing Response Assessment)

Rather than relying solely on the initial risk assessment, dynamic risk stratification allows ongoing reclassification based on treatment response. An excellent response, defined by negative imaging with suppressed thyroglobulin below 0.2 ng/mL or stimulated thyroglobulin below 1 ng/mL, carries only a 1 to 4% recurrence risk and permits de-escalation of surveillance and TSH suppression. A biochemically incomplete response shows abnormal thyroglobulin or rising TgAb without structural disease. A structurally incomplete response indicates known structural disease on imaging, requiring consideration of surgery, RAI, external beam radiation, or systemic therapy. An indeterminate response features nonspecific imaging findings or detectable but stable thyroglobulin, warranting continued observation.

### Surveillance

Thyroglobulin and TgAb should be measured every 6 to 12 months as the primary tumor markers after thyroidectomy and RAI. Neck ultrasound is performed at 6 to 12 months postoperatively and then periodically based on risk. Diagnostic whole-body scanning is less frequently used for low-risk patients with excellent response and is reserved for intermediate or high-risk surveillance. Cross-sectional imaging with CT chest, MRI, or PET-CT is employed for structurally incomplete response or rising thyroglobulin without identifiable disease on ultrasound.

### Advanced/Refractory DTC

RAI-refractory disease, defined by loss of RAI avidity, progression despite RAI, or cumulative dose exceeding 600 mCi without benefit, represents a significant clinical challenge. First-line systemic therapy consists of multi-kinase inhibitors: lenvatinib (SELECT trial: median PFS 18.3 versus 3.6 months, objective response rate 65%) and sorafenib (DECISION trial: median PFS 10.8 versus 5.8 months, objective response rate 12%). Selective RET inhibitors (selpercatinib, pralsetinib) are available for RET-fusion positive DTC. NTRK inhibitors (larotrectinib, entrectinib) provide highly effective treatment for the rare NTRK-fusion positive thyroid cancers. BRAF/MEK inhibitor combinations (dabrafenib plus trametinib) can be used for BRAF V600E-positive refractory PTC. An exciting emerging approach is RAI resensitization using the MEK inhibitor selumetinib, which has been shown to restore RAI uptake in some RAI-refractory patients, though this is not yet standard of care.

## Key Clinical Pearls

- The ATA 2015 guidelines represent a significant shift toward less aggressive management of low-risk DTC: lobectomy instead of total thyroidectomy, omission of RAI, and relaxed TSH targets are now appropriate for many patients
- Active surveillance (no surgery) is a valid option for papillary microcarcinoma ≤1 cm without high-risk features; growing evidence from Japan (Kuma Hospital) and memorial cohorts support this approach
- BRAF V600E mutation is the most common driver in PTC (~60%); its prognostic significance is amplified when co-occurring with TERT promoter mutations (synergistic effect on recurrence and mortality)
- Molecular testing (Afirma GSC, ThyroSeq v3) has reduced unnecessary surgery for indeterminate thyroid nodules by approximately 50%; understand the strengths and limitations of each platform
- All patients with MTC should undergo RET germline testing; a positive result mandates family screening and may change surgical approach (prophylactic thyroidectomy in MEN2)
- Dabrafenib + trametinib for BRAF-mutated ATC is one of the most impactful advances in thyroid oncology; always test for BRAF in ATC

## References

1. Haugen BR, et al. "2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer." Thyroid. 2016;26(1):1-133.
2. Ali SZ, Cibas ES. "The Bethesda System for Reporting Thyroid Cytopathology." 2nd edition. Springer, 2018.
3. Filetti S, et al. "Thyroid Cancer: ESMO Clinical Practice Guidelines." Ann Oncol. 2019;30(12):1856-1883.
4. Ito Y, et al. "Active Surveillance for Papillary Thyroid Microcarcinoma." Thyroid. 2018;28(10):1365-1372.
5. Subbiah V, et al. "Dabrafenib plus Trametinib in BRAF V600E-Mutant Anaplastic Thyroid Cancer (ROAR)." J Clin Oncol. 2018;36(1):7-13.