Oncocytic thyroid carcinoma is a malignant epithelial neoplasm derived from thyroid follicular cells, composed predominantly of oncocytic cells, historically known as Hürthle cells, and defined by the presence of capsular invasion, vascular invasion, lymph node metastases or distant metastases. According to the 2022 World Health Organization (WHO) classification, it should not be regarded merely as a morphological variant of conventional follicular carcinoma, but as a distinct category of follicular cell-derived thyroid carcinoma. Its identity derives from the combination of an oncocytic phenotype, mitochondrial accumulation, a distinctive genomic profile, potentially reduced sensitivity to radioactive iodine and more variable clinical behaviour than minimally invasive follicular forms.
It is a rare tumour compared with papillary carcinoma and is also less common than conventional follicular carcinoma, but it has disproportionate clinical relevance because it may present with vascular invasion, distant metastases and late recurrences. It predominantly affects adults and older individuals, with a higher frequency among women, and may be discovered as a cytologically oncocytic thyroid nodule, a suspicious thyroid mass, a lymph node metastasis or a distant metastatic lesion. The diagnosis cannot be established with certainty on cytology alone when there is no evidence of invasion or metastasis, because oncocytic adenoma and oncocytic carcinoma may share the same cellular appearance. The essential clinical issue is therefore to distinguish an indeterminate oncocytic neoplasm from an invasive oncocytic carcinoma, and then to stratify risk according to the capsule, vessels, local extension, metastases and therapeutic response.
Oncocytic carcinoma is composed of transformed follicular cells that acquire abundant, granular and eosinophilic cytoplasm because of marked mitochondrial accumulation. The term “oncocytic” therefore describes a cellular phenotype and does not automatically indicate malignancy. Oncocytic cells may occur in chronic autoimmune thyroiditis, hyperplastic nodules, adenomas, papillary carcinomas with oncocytic features, follicular oncocytic carcinomas historically referred to as Hürthle cell carcinomas and more aggressive tumours. To define oncocytic carcinoma, an invasive oncocytic follicular neoplasm is required, usually composed of at least 75% oncocytic cells, lacking the diagnostic nuclear features of papillary carcinoma and not classifiable as high-grade or poorly differentiated thyroid carcinoma. Diagnosis therefore requires the combination of cellular morphology and evidence of invasiveness.
Differentiation from oncocytic papillary carcinoma is essential. A tumour may contain oncocytic cells but display optically clear nuclei, grooves, pseudoinclusions, nuclear overlapping and other features typical of papillary carcinoma; in that case, the diagnosis remains papillary carcinoma with oncocytic differentiation, rather than oncocytic carcinoma in the strict sense. Similarly, a thyroid carcinoma with tumour necrosis and high mitotic activity may fall within the follicular cell-derived high-grade carcinoma categories if it fulfils the relevant criteria, because its biological behaviour is different. True oncocytic carcinoma is therefore a precisely delimited diagnosis based on oncocytic cells, absence of diagnostic papillary-type nuclei and demonstration of malignancy.
The diagnosis of malignancy follows the same general principle as for encapsulated follicular neoplasms: fine-needle aspiration may suggest an oncocytic neoplasm, but it cannot independently demonstrate capsular or vascular invasion. A cytological sample rich in oncocytic cells, with granular cytoplasm, round nuclei, prominent nucleoli and scant colloid, may be compatible with an oncocytic neoplasm, but it cannot distinguish oncocytic adenoma from oncocytic carcinoma unless metastases or macroscopic invasion are present. Definitive diagnosis requires histological examination of the excised nodule, with careful sampling of the capsule and vessels. This limitation of oncocytic cytology must be explained clearly, because it prevents both false reassurance and a premature cancer diagnosis.
Oncocytic carcinoma may be minimally invasive, encapsulated angioinvasive or widely invasive. In minimally invasive forms, malignancy is demonstrated by focal penetration through the capsule or limited invasion, whereas in angioinvasive forms the tumour enters capsular or extracapsular vessels. In widely invasive forms, the capsular profile is lost or extensively exceeded and the neoplasm infiltrates the thyroid parenchyma or perithyroidal tissues. The number of foci of vascular invasion, the presence of macroscopic extrathyroidal extension, margin status, necrosis, mitotic activity and metastases substantially modify risk. The mere presence of oncocytosis is not sufficient: prognosis depends on the degree of capsulovascular invasion.
Oncocytic carcinoma must also be differentiated from oncocytic adenoma and hyperplastic oncocytic lesions. Oncocytic adenoma is a benign encapsulated neoplasm without capsular or vascular invasion; oncocytic lesions occurring in chronic thyroiditis may be multifocal, non-encapsulated and located within inflamed thyroid parenchyma. Misdiagnosis may have major consequences: labelling a non-invasive lesion as carcinoma causes overtreatment, whereas underestimating an angioinvasive carcinoma may delay completion surgery, selective radioactive iodine treatment, metastatic imaging and follow-up. The differential diagnosis therefore requires assessment of architecture, capsule, vessels, immunophenotype when necessary and clinical context.
The histological report should specify tumour size, percentage of oncocytic component, capsular status, vascular invasion and the number of foci, extrathyroidal extension, margins, necrosis, mitoses, any high-grade transformation, lymph nodes if assessed, metastases and the relationship with other thyroid lesions. The threshold of at least 75% oncocytic cells prevents follicular or papillary tumours with limited oncocytic metaplasia from being inappropriately reclassified. Histological assessment must be as accurate as possible because, in oncocytic carcinoma, pathological reporting is not merely descriptive, but determines the extent of treatment and intensity of surveillance.
Most oncocytic carcinomas have no single demonstrable aetiological cause. The tumour arises from thyroid follicular cells that acquire an oncocytic phenotype and invasive capacity through nuclear, mitochondrial and chromosomal alterations. Associated factors include middle to advanced age, female sex, thyroid nodularity, a history of oncocytic neoplasia, multinodular goitre and, in some cases, chronic autoimmune thyroiditis. The association with Hashimoto thyroiditis should not be interpreted as automatic causality, because oncocytosis may represent a metaplastic response to chronic inflammation. Oncocytic carcinoma develops when neoplastic transformation is combined with capsular or vascular invasion or dissemination.
The central biological feature is mitochondrial accumulation. Oncocytic cells have abundant granular cytoplasm because they contain numerous mitochondria, which are often functionally altered. Mitochondrial DNA mutations, particularly those involving genes encoding complex I subunits of the respiratory chain, may reduce the efficiency of oxidative phosphorylation and promote a compensatory increase in mitochondrial mass. Oncocytosis is therefore not merely a microscopic appearance, but reflects mitochondrial reprogramming that modifies energy metabolism, oxidative stress and cellular adaptation.
Genomic studies have shown that oncocytic carcinoma has a distinctive chromosomal landscape, different from that of classic papillary carcinoma and conventional follicular carcinoma. Widespread chromosomal losses, near-haploid states, extensive loss of heterozygosity and subsequent chromosomal duplication or uniparental disomy have been described in tumour subgroups. This pattern may contribute to genomic instability, clonal selection and aggressive progression. The combination of mitochondrial alterations and extensive chromosomal loss supports the concept that oncocytic carcinoma is a distinct molecular entity, rather than merely an aesthetic variant of follicular carcinoma.
In addition to mitochondrial and chromosomal alterations, mutations or rearrangements may occur in genes involved in growth pathways, cell-cycle control and epigenetic remodelling. Alterations involving rat sarcoma viral oncogene homolog (RAS), phosphatase and tensin homolog (PTEN), tumour protein p53 (TP53), telomerase reverse transcriptase (TERT), DAXX chromatin remodeler (DAXX), AT-rich interaction domain 1A (ARID1A), MEN1 and other genes have been described with variable frequencies. A TERT promoter mutation, when present, has an adverse significance because it promotes prolonged replicative capacity and may be associated with more aggressive behaviour, particularly in invasive or metastatic tumours.
The pathogenesis of invasion follows a principle similar to that of other encapsulated follicular neoplasms: malignancy becomes clinically relevant when the tumour crosses the capsule or invades vessels. Capsular invasion enables local spread; vascular invasion provides access to the systemic circulation and explains pulmonary and bone metastases. In oncocytic carcinoma, this phenomenon may be particularly important because the disease has a greater propensity for distant metastases than many low-risk papillary carcinomas and may show reduced iodine uptake. Angioinvasion therefore represents the anatomical link between histological diagnosis and metastatic risk.
Oncocytic metabolism may also contribute to the reduced effectiveness of radioactive iodine in a proportion of patients. Reduced expression or function of the sodium iodide symporter (NIS), loss of thyroid differentiation, alteration of the follicular transcriptional programme and selection of more glucose-avid clones may make some lesions less capable of taking up radioactive iodine and more visible on fluorodeoxyglucose positron emission tomography (FDG-PET). This relationship is not absolute, because some oncocytic carcinomas take up iodine and respond to radiometabolic treatment, but the risk of radioiodine refractoriness should be considered earlier than in many classic differentiated forms.
Progression to high-grade, poorly differentiated or anaplastic thyroid carcinoma is rare but possible. This evolution is characterised by necrosis, increased mitotic activity, loss of differentiation markers, greater invasiveness, genomic instability and resistance to thyroid-selective treatments. The presence of tumour necrosis and increased mitotic activity changes the classification and clinical behaviour because it indicates more aggressive biology than well-differentiated oncocytic carcinoma. Disease control therefore depends on the ability to recognise dedifferentiation, structural progression and loss of iodine uptake at an early stage.
Oncocytic carcinoma may present as a solitary encapsulated nodule, a dominant thyroid mass within a multinodular goitre or a widely invasive infiltrative tumour. The cut surface is often reddish-brown or mahogany-coloured because of the cellular and mitochondrial abundance, but this macroscopic finding is not sufficient to establish malignancy. As in follicular neoplasms, a circumscribed appearance may conceal focal capsular or vascular invasion. A well-demarcated, apparently expansile lesion may be a carcinoma if it shows capsular penetration or vascular invasion.
The most characteristic route of dissemination is haematogenous. Vascular invasion allows neoplastic cells to reach the lungs, bones and other systemic sites. Compared with papillary carcinoma, cervical lymph node involvement is less frequent, but it is not absent; when present, it may indicate more advanced disease, aggressive behaviour or the need for careful reassessment of the histological diagnosis. Oncocytic carcinoma should not be interpreted according to the same model as papillary carcinoma, in which cervical lymph nodes dominate the natural history. In this disease, surveillance must account for the risk of haematogenous metastases.
Pulmonary metastases may be micronodular, multiple nodular or solitary, and may or may not take up iodine. Iodine-avid lesions may benefit from radioactive iodine, whereas FDG-avid, non-iodine-avid lesions often indicate reduced differentiation and a more complex prognosis. Bone metastases are clinically important because they may be osteolytic, painful, hypervascular and associated with pathological fractures, vertebral instability or spinal cord compression. The impact of bone disease depends not only on the number of lesions, but also on their anatomical location and mechanical or neurological risk.
Local invasion may involve perithyroidal soft tissues, strap muscles, the trachea, oesophagus, recurrent laryngeal nerve and vascular structures. Widely invasive forms may present with a hard mass, fixation, dysphonia, dysphagia, dyspnoea, cough or haemoptysis. Macroscopic invasion is more important than minimal microscopic extension because it modifies the surgical resection, the risk of residual disease and the need for additional local treatment. The description of extrathyroidal invasion must be anatomical rather than generic, specifying the structure involved and whether complete resection is possible.
Oncocytic carcinoma may recur locally, in lymph nodes or at distant sites even after a prolonged interval. Recurrences may be difficult to identify if they do not take up iodine and if thyroglobulin is interpreted without considering anti-thyroglobulin antibodies, residual tissue or analytical variability. Computed tomography (CT), magnetic resonance imaging (MRI) and FDG-PET have a more important role in high-risk, metastatic cases or when thyroglobulin findings are discordant with scintigraphy. The clinical course depends on structural growth, lesion location and sensitivity to available treatments.
The most common presentation is a solitary or dominant thyroid nodule, often detected by ultrasonography, palpation or during investigations for multinodular goitre. Patients are generally euthyroid because oncocytic carcinoma rarely produces enough thyroid hormone to cause clinical hyperthyroidism. The medical history should assess nodule duration and growth, chronic thyroiditis, family history of thyroid neoplasms, previous cervical irradiation, dysphonia, dysphagia, dyspnoea, local pain, cough, bone pain, fractures, neurological symptoms and weight loss. Progressive growth of an oncocytic nodule warrants attention because it may indicate invasive transformation or increased surgical risk.
On physical examination, the thyroid may contain a firm, non-tender nodule that moves with swallowing and may occasionally be large. Fixation to deep planes, reduced mobility, dysphonia or tracheal deviation suggest local invasion or compression. Cervical lymph nodes should be assessed systematically, even though lymphatic spread is less typical than in papillary carcinoma. Examination should include assessment of the voice, respiratory signs, swallowing, elicitable skeletal pain and focal neurological findings. In oncocytic carcinoma, the absence of lymphadenopathy does not exclude systemic disease, particularly when histology demonstrates extensive angioinvasion.
A proportion of patients are diagnosed after indeterminate oncocytic cytology. In these cases, the clinical problem is not to prove carcinoma immediately, but to determine the operation required to obtain a histological diagnosis. The fine-needle aspiration report may describe a “Hürthle cell neoplasm”, “oncocytic neoplasm” or a Bethesda category compatible with an oncocytic follicular lesion. This result indicates a risk of neoplasia, but does not distinguish adenoma from carcinoma. Clinical communication must be precise: the suspicion concerns an oncocytic lesion requiring histological evaluation, not a preoperative certainty of invasive carcinoma.
Metastatic manifestations may precede or accompany the thyroid nodule. Bone metastases may cause vertebral pain, pelvic pain, bone swellings, pathological fractures, neurological deficits or spinal cord compression. Pulmonary metastases may be asymptomatic or cause cough and dyspnoea. A metastatic lesion may be biopsied before the thyroid diagnosis and show immunoreactivity for thyroglobulin, paired box gene 8 (PAX8) and thyroid transcription factor 1 (TTF-1), supporting a thyroid follicular origin. This presentation with an initial metastasis is less common, but is consistent with the haematogenous propensity of the disease.
Systemic symptoms are non-specific and occur mainly in advanced forms. Fatigue, weight loss, persistent pain, dyspnoea or reduced physical performance should be interpreted in the context of tumour burden, metastases and comorbidities. Hyperthyroidism is not a typical manifestation of oncocytic carcinoma, but a suppressed thyroid-stimulating hormone (TSH) level requires scintigraphy to define functional autonomy of the nodule or associated goitre. Thyroid function describes the endocrine context, whereas malignancy and prognosis depend on invasion, histology and spread.
The diagnostic pathway begins with medical history, physical examination, TSH measurement and thyroid and cervical ultrasonography. Ultrasonography assesses size, composition, echogenicity, margins, peripheral halo, vascularity, relationship with the thyroid capsule, possible extrathyroidal extension and lymph nodes. Oncocytic lesions may be hypoechoic, isoechoic, heterogeneous, solid or partially cystic, with variable vascularity; no ultrasonographic pattern can distinguish oncocytic adenoma from oncocytic carcinoma with certainty unless clear invasion or metastases are present. Ultrasonography therefore defines risk, location, size and the indication for fine-needle aspiration, but does not replace the histological diagnosis.
If TSH is suppressed, thyroid scintigraphy is useful for distinguishing hyperfunctioning from non-functioning nodules. Oncocytic lesions may coexist with multinodular goitre or functional autonomy, but most differentiated malignant tumours are hypofunctioning or non-functioning. If TSH is normal or elevated, ultrasound-guided fine-needle aspiration is indicated according to size and ultrasonographic characteristics. Cytology is interpreted according to The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC), with possible classification as atypia, follicular neoplasm, oncocytic neoplasm or suspicion of malignancy. The crucial point is that cytology may identify the oncocytic phenotype, but cannot demonstrate capsular or vascular invasion.
A definitive diagnosis of oncocytic carcinoma requires histological examination of the surgical specimen. Sampling must include the tumour capsule, suspicious points, capsular and extracapsular vessels, margins and areas with a more solid, necrotic or infiltrative appearance. Demonstration of capsular invasion, vascular invasion, lymph node metastases or distant metastases establishes the diagnosis of oncocytic carcinoma in an oncocytic neoplasm. In the absence of invasion, an encapsulated oncocytic neoplasm remains an oncocytic adenoma or non-malignant lesion, even if cytology shows abundant Hürthle cells. Evidence of invasiveness is therefore the diagnostic cornerstone.
According to the WHO classification and international guideline framework, a diagnosis of oncocytic carcinoma requires demonstration of an invasive oncocytic follicular neoplasm composed predominantly of oncocytic cells, lacking the diagnostic nuclear features of papillary carcinoma and distinguished from high-grade thyroid carcinomas when specific high-grade criteria are present. Diagnostic reasoning must proceed sequentially:
Diagnosis of oncocytic carcinoma
Molecular tests may assist in the assessment of indeterminate nodules, but their performance in oncocytic lesions must be interpreted cautiously. Some molecular classifiers may increase or reduce the probability of malignancy, whereas the presence of mitochondrial mutations, chromosomal alterations or nuclear mutations does not automatically demonstrate histological invasion. In advanced, metastatic or radioiodine-refractory forms, molecular profiling becomes more relevant because it may identify prognostic alterations, TERT promoter mutations, TP53 alterations, rare fusions or therapeutic targets. The clinical value of the molecular profile depends on which decision it can modify.
The differential diagnosis includes oncocytic adenoma, chronic thyroiditis with oncocytic metaplasia, oncocytic hyperplastic nodule, oncocytic papillary carcinoma, conventional follicular carcinoma with oncocytic areas, high-grade thyroid carcinoma, poorly differentiated carcinoma, medullary carcinoma with eosinophilic cytoplasm, oxyphilic parathyroid tumours and intrathyroidal metastases. Immunohistochemistry may support a thyroid follicular origin through thyroglobulin, PAX8 and TTF-1, whereas calcitonin supports medullary carcinoma and parathyroid hormone supports a parathyroid origin. However, the distinction between oncocytic adenoma and oncocytic carcinoma remains based on the capsule and vessels.
Second-level investigations are indicated when the tumour is large, invasive, angioinvasive or metastatic, or when postoperative markers are suspicious. CT of the neck and chest assesses local extension, the mediastinum and pulmonary metastases; MRI is useful for infiltration of soft tissues, the spine, spinal cord and brain; radioactive iodine scintigraphy evaluates iodine uptake; FDG-PET is particularly useful when thyroglobulin is elevated, scintigraphy is negative or dedifferentiated disease is suspected. Imaging should be selected according to a precise clinical question concerning tumour localisation, extent or progression.
In oncocytic thyroid carcinoma, anatomical staging uses the eighth edition of the American Joint Committee on Cancer (AJCC) system applied to differentiated follicular cell-derived thyroid carcinomas. However, AJCC staging alone does not fully describe clinical risk, because behaviour also depends on vascular invasion, extracapsular extension, margins, radioactive iodine uptake, metastases and possible high-grade transformation. Three levels must therefore be distinguished: the TNM category, the AJCC stage and the risk stratification for persistence or recurrence. In oncocytic carcinoma, the importance of angioinvasion is particularly substantial.
The T category describes the size and local extension of the primary tumour. Tumour size is important, but does not replace histological assessment of the capsule and vessels. A small tumour with vascular invasion may have a different risk from a larger tumour that is well confined and completely resected. The T category must be reported explicitly:
T category in oncocytic thyroid carcinoma
The N category describes regional lymph node involvement. In oncocytic carcinoma, lymph node metastases are less typical than in papillary carcinoma, but they may occur and must be documented when present. The presence of metastatic lymph nodes increases the risk of locoregional persistence and may indicate more aggressive disease. The N category is defined as follows:
N category in oncocytic thyroid carcinoma
The M category is central because oncocytic carcinoma may spread haematogenously to the lungs and bones. The presence of distant metastases modifies the stage and guides imaging, radioactive iodine treatment, radiotherapy, local surgery, FDG-PET and systemic therapy. The M category must always be specified:
M category in oncocytic thyroid carcinoma
After T, N and M have been defined, the AJCC stage is assigned according to age, using a threshold of 55 years. In patients younger than 55 years, AJCC staging primarily distinguishes the absence or presence of distant metastases. This does not mean that local invasion, angioinvasion or lymph node involvement are irrelevant to treatment and follow-up, but only that the AJCC system retains a simplified structure for predicting disease-specific mortality. AJCC staging in patients younger than 55 years is as follows:
AJCC staging for patients younger than 55 years
In patients aged 55 years or older, AJCC staging is more detailed because tumour size, gross invasion, lymph node involvement and distant metastases have greater prognostic importance. In oncocytic carcinoma, this framework must be integrated with pathological assessment of vascular invasion and response to radioactive iodine. For patients aged at least 55 years, AJCC staging is as follows:
AJCC staging for patients aged at least 55 years
Recurrence-risk stratification adds biological and therapeutic variables that are not fully captured by the AJCC stage. In oncocytic carcinoma, vascular invasion, the extent of invasion, completeness of resection, distant metastases, iodine uptake, necrosis, mitotic activity and high-grade transformation are particularly important. Risk stratification may be summarised as follows:
Clinical risk in oncocytic thyroid carcinoma
Initial risk stratification must be updated during follow-up. After surgery, any radioactive iodine treatment and the first assessments, the response may be excellent, indeterminate, biochemically incomplete or structurally incomplete. In oncocytic carcinoma, a biochemically incomplete response requires attention because the disease may have poor iodine uptake and require CT, MRI or FDG-PET. Dynamic risk stratification allows treatment to be reduced in patients who are genuinely disease-free and surveillance or therapy to be intensified in those with persistent or progressive disease.
Treatment often begins with diagnostic and therapeutic surgery for an indeterminate oncocytic neoplasm. Lobectomy may be the appropriate initial procedure for oncocytic nodules without preoperative evidence of invasion, metastases or bilateral disease, because it enables definitive histological diagnosis and may be sufficient for low-risk minimally invasive tumours. After the definitive report, the need for completion thyroidectomy depends on size, vascular invasion, gross invasion, margins, metastases, bilateral disease and the need for radioactive iodine or more sensitive biochemical follow-up. Surgery should be guided by histological risk, not by the term “oncocytic” alone.
Total thyroidectomy is generally indicated or strongly considered for widely invasive forms, large tumours, extensive angioinvasion, locally advanced disease, distant metastases, positive margins or when radioactive iodine treatment is planned. Its advantages are removal of all thyroid tissue, easier interpretation of thyroglobulin and the possibility of radiometabolic treatment if the disease takes up iodine. Its disadvantages are an increased risk of hypoparathyroidism, recurrent laryngeal nerve injury and dependence on replacement therapy. Total thyroidectomy is therefore a tool for cancer control, not an automatic step for every oncocytic neoplasm.
Lymph node dissection should be therapeutic. Oncocytic carcinoma may metastasise to lymph nodes, but lymphatic spread is not its dominant pattern. Clinically or ultrasonographically suspicious lymph nodes should be investigated and, if metastatic, treated with compartment-oriented dissection of the involved level. Routine prophylactic dissection in the absence of suspicious lymph nodes is not justified in low-risk cases because it causes morbidity without a clear benefit. Lymph node surgery should address documented disease, rather than a generic fear of recurrence.
Radioactive iodine treatment with iodine-131 has a selective role. Some oncocytic carcinomas take up iodine and may benefit from radiometabolic treatment, particularly in the presence of residual disease, intermediate to high risk or iodine-avid metastases. However, compared with other differentiated thyroid carcinomas, oncocytic carcinoma more often shows reduced iodine uptake or radioiodine refractoriness. The indication must therefore be individualised by integrating histological risk, postoperative thyroglobulin, scintigraphy, anatomical imaging and the probability of benefit. Radiometabolic treatment is rational when iodine-avid disease is present or when the risk is sufficient to justify ablation or adjuvant therapy.
Levothyroxine treatment is used to replace thyroid function after total thyroidectomy and to modulate TSH in patients at risk. The degree of suppression should be proportionate: more marked in persistent or high-risk disease, intermediate in patients at intermediate risk and less aggressive in low-risk patients with an excellent response. Unnecessary chronic suppression increases the risk of atrial fibrillation, tachycardia, bone loss and cardiovascular deterioration. In oncocytic carcinoma, the TSH target must balance cancer risk, age, bone health, cardiovascular status and response to therapy.
Metastatic disease requires a multimodal approach. Iodine-avid pulmonary metastases may be treated with radioactive iodine, whereas non-iodine-avid or progressive lesions require radiological surveillance, FDG-PET, local treatments or systemic therapy. Bone metastases may require orthopaedic or neurosurgical procedures, external-beam radiotherapy, stabilisation, vertebroplasty, thermal ablation, embolisation and pain control. The goal in bone metastasis is not only to reduce tumour burden, but also to prevent fracture, spinal cord compression and functional loss.
Systemic therapy may be considered for progressive radioiodine-refractory disease. Antiangiogenic multikinase inhibitors, such as lenvatinib and sorafenib, are validated options for locally advanced or metastatic differentiated thyroid carcinomas that are progressive and refractory to radioactive iodine. When actionable molecular alterations are present, such as neurotrophic tyrosine receptor kinase (NTRK) fusions or rearranged during transfection (RET) alterations, selective inhibitors may be considered when indicated. Systemic therapy should not be started for stable and asymptomatic lesions, but for documented progression, symptoms, anatomical threat or clinically relevant tumour burden.
Prognosis is highly variable. Minimally invasive forms that are completely resected, without significant angioinvasion or metastases, may have a favourable course. Widely invasive, extensively angioinvasive, metastatic, non-iodine-avid, FDG-avid forms or tumours with adverse mutations have a higher risk of recurrence, progression and disease-specific mortality. Older age, large tumour size, positive margins, bone metastases, high-grade transformation and radioiodine refractoriness worsen the outcome. Oncocytic carcinoma should therefore be presented as a disease with a risk-stratified prognosis, rather than merely as a follicular carcinoma “with large cells”.
Follow-up depends on the initial operation and histological risk. After total thyroidectomy, thyroglobulin is the main biochemical marker, but it must be interpreted together with anti-thyroglobulin antibodies, TSH, the laboratory method, any residual thyroid tissue and imaging. After lobectomy, thyroglobulin is less specific because the remaining lobe physiologically produces thyroglobulin. In oncocytic carcinoma, a progressive increase in thyroglobulin requires investigation for cervical, pulmonary, bone or non-iodine-avid disease.
Cervical ultrasonography assesses the thyroid bed, remaining lobe and central and lateral lymph nodes. It is useful for identifying local or lymph node recurrence, but is not sufficient in patients with a high risk of haematogenous spread. In tumours with extensive angioinvasion, metastases, rising thyroglobulin or negative scintigraphy, chest CT, targeted MRI, bone imaging and FDG-PET should be considered. Surveillance must be adapted to the route of spread of oncocytic carcinoma, which may be systemic even in the absence of cervical lymphadenopathy.
Radioactive iodine scintigraphy remains useful when the disease takes up iodine or when radiometabolic treatment is being considered. However, FDG-PET is particularly valuable in oncocytic carcinomas because many lesions are metabolically active and less capable of taking up iodine. Elevated thyroglobulin, negative radioactive iodine scintigraphy and positive FDG-PET suggest dedifferentiated or radioiodine-refractory disease. The so-called metabolic flip-flop, namely reduced iodine uptake and increased glycolytic avidity, has prognostic and therapeutic implications.
Response to therapy is classified as an excellent response, indeterminate response, biochemically incomplete response or structurally incomplete response. An excellent response allows the intensity of follow-up and TSH suppression to be reduced. A biochemically incomplete response requires serial trend assessment and appropriate imaging. A structurally incomplete response must be evaluated according to location, growth, symptoms, iodine uptake and FDG avidity. In oncocytic carcinoma, structural findings are decisive because an unstable bone lesion, a progressive pulmonary metastasis or an infiltrative cervical recurrence require different decisions. The structural response guides treatment more than an isolated marker value.
Follow-up should be prolonged in patients with vascular invasion, widely invasive tumours, metastases or radioiodine-refractory disease. Recurrences may occur after a long interval and sometimes at non-cervical sites. In patients with a completely resected minimally invasive tumour, surveillance may instead be less intensive, avoiding repeated investigations without a genuine therapeutic impact. High-quality follow-up consists of adapting surveillance and imaging to the residual risk, while avoiding both diagnostic delays and overdiagnosis.
The most characteristic oncological complication is distant metastasis. Vascular invasion permits haematogenous spread to the lungs, bones and, less commonly, the brain, liver or other sites. Pulmonary metastases may be asymptomatic or cause cough, dyspnoea and respiratory deterioration. Bone metastases may produce pain, pathological fractures, spinal cord compression, neurological deficits and hypercalcaemia in advanced cases. Haematogenous metastasis is the complication that most strongly affects prognosis and treatment planning.
Locoregional recurrence may involve the thyroid bed, cervical soft tissues or lymph nodes. In widely invasive tumours, recurrence may involve the trachea, oesophagus, recurrent laryngeal nerve or major vessels, making repeat surgery and local treatment complex. Dysphonia may result from tumour invasion or surgical injury; dysphagia and dyspnoea may indicate a compressive mass or aerodigestive infiltration. Loss of cervical control may impair quality of life even when systemic disease is not immediately fatal.
Radioiodine refractoriness is a particularly important biological complication. When tumour cells reduce iodine uptake or organification, treatment with iodine-131 loses effectiveness. The disease may then become visible on FDG-PET, progress despite radiometabolic treatment or require systemic agents. Refractoriness does not always require immediate treatment because some lesions remain stable, but it becomes clinically problematic when combined with progression, symptoms or critical locations. The actual complication is the combination of loss of iodine uptake and tumour growth.
Surgical complications include transient or permanent hypoparathyroidism, hypocalcaemia, recurrent laryngeal nerve injury, superior laryngeal nerve injury, compressive cervical haematoma, infection, seroma, pathological scarring and lymphatic leakage if lateral neck dissection has been performed. Chronic hypoparathyroidism may cause paraesthesia, cramps, tetany, nephrocalcinosis or the need for permanent treatment with calcium and active vitamin D. Recurrent laryngeal nerve injury may cause persistent dysphonia, vocal fatigue and risk of aspiration. Surgical morbidity requires the extent of the operation to be calibrated to the actual risk.
Radioactive iodine may cause sialadenitis, xerostomia, taste alterations, nausea, cervical pain, lacrimal dysfunction, bone marrow toxicity with high cumulative doses, transient reduction in fertility and a small increase in the risk of second neoplasms in selected contexts. In oncocytic carcinoma, in which the probability of reduced iodine uptake may be higher, it is particularly important to avoid repeated treatment without demonstrable benefit. Radioactive iodine should be used when there is an expected benefit, rather than automatically after every diagnosis of oncocytic carcinoma.
Chronic TSH suppression with levothyroxine may cause iatrogenic subclinical thyrotoxicosis, palpitations, atrial fibrillation, loss of bone mass, fractures and worsening of pre-existing heart disease. In patients with persistent or high-risk disease, more intense suppression may be justified; in disease-free or low-risk patients, the same intensity may become harmful. The complication arises when endocrine therapy is not recalibrated according to the dynamic response.
A frequent management-related complication is overtreatment of indeterminate oncocytic neoplasms. Many oncocytic nodules are not carcinomas, and some minimally invasive carcinomas have a favourable course after adequate surgery. Conversely, underestimating an angioinvasive or widely invasive oncocytic carcinoma may result in unrecognised metastases, insufficient follow-up and delayed treatment. Appropriate management requires a balance between oncological caution and reduction of iatrogenic harm, using complete histological assessment, targeted imaging and dynamic reassessment as essential steps.