Familial and Hereditary Nonmedullary Thyroid Carcinoma

Familial clustering of thyroid follicular cell–derived carcinoma (papillary, follicular, oncocytic, and poorly differentiated), encompassing a heterogeneous nonsyndromic form with a polygenic basis and no validated screening gene, and syndromic forms in which the carcinoma is one component of a monogenic tumour predisposition (PHTS/PTEN, familial adenomatous polyposis/APC, DICER1, Carney complex/PRKAR1A, Werner syndrome/WRN, Pendred syndrome/SLC26A4).

Literature and regulatory information updated: 17 October 2026.

0. Reference Summary Table

The following table describes the principal entity, familial nonmedullary thyroid carcinoma (FNMTC), in its nonsyndromic form. The second table summarises the syndromic forms and associated genetic entities.

1. Introduction

1.1. Scope and definitions

Thyroid cancer accounts for about 1% of new malignancies; approximately 94% are differentiated carcinomas derived from follicular cells, 5% are medullary carcinomas derived from neuroendocrine C cells, and 1% are anaplastic carcinomas [42]. Nonmedullary thyroid carcinoma (NMTC) comprises tumours of follicular origin: papillary carcinoma (PTC), follicular carcinoma (FTC), oncocytic carcinoma (formerly called Hürthle cell carcinoma and classified as a subtype of follicular carcinoma), and poorly differentiated carcinoma (PDTC), with anaplastic carcinoma at the extreme of dedifferentiation [4]. Medullary carcinoma, whose hereditary form is caused by germline RET variants in multiple endocrine neoplasia type 2, has its own biology, markers, and management and is not developed here [42, 43].

In its 2025 guidelines for adult DTC, the ATA divides familial nonmedullary thyroid carcinoma (FNMTC) into two categories. Syndromic FNMTC is that in which thyroid carcinoma is one of several tumours of a defined predisposition syndrome, such as PTEN hamartoma tumour syndrome (PHTS, which includes Cowden syndrome), familial adenomatous polyposis (FAP), Carney complex, and Werner syndrome. Nonsyndromic FNMTC is that in which DTC is the only or predominant inherited neoplasm [4]. DICER1 tumour predisposition, Pendred syndrome, long telomere syndrome, and CHEK2 founder variants complete the group of identifiable monogenic causes [4, 10, 35, 40]. The histopathological classification of Guilmette and Nosé separates a first group of syndromes in which nonthyroidal tumours predominate (FAP, Cowden, Werner, Carney, Pendred) from a second group in which NMTC predominates, comprising familial pure PTC with or without oxyphilia, familial PTC with papillary renal carcinoma, and familial PTC with MNG [44].

Orphanet defines familial papillary or follicular thyroid carcinoma (ORPHA:319487) as a rare hereditary nonmedullary thyroid carcinoma characterised by DTC of follicular origin in two or more first-degree relatives, in the absence of other familial tumour syndromes or radiation exposure; it describes frequent capsular invasion and multicentric tumours with multiple adenomatous nodules, with or without oxyphilia, and follicular or papillary histology [1]. The group ORPHA:319494 covers the nonsyndromic forms in which thyroid carcinoma is the main feature in a familial context, and the subtype ORPHA:97290 corresponds to familial PTC with papillary renal neoplasia [2, 3]. The definition based on two affected individuals is disputed, because the probability that two sporadic cases cluster by chance is high (see 3.1 and 5.1); several research groups require three or more affected relatives [23, 45, 46].

1.2. Historical overview

The first cited description of familial clustering of NMTC is that of Robinson and Orr in 1955, on thyroid carcinoma and other thyroid diseases in identical twins [47]. In the 1990s and early 2000s, linkage analysis in multiplex families mapped susceptibility loci without identifying the responsible gene: MNG1 on 14q, in a large Canadian family with 18 cases of MNG, in 1997 [48]; TCO on 19p13.2, in a French family with oxyphilic cell thyroid tumours, in 1998 [27]; fPTC/PRN on 1q21, in a three-generation family with PTC and papillary renal neoplasia, in 2000 [49]; and NMTC1 on 2q21, from a Tasmanian family, in 2001 [50].

The syndromic forms were clarified during the same period. In 1994, Harach, Williams, and Williams described FAP-associated thyroid carcinoma as a distinct type of follicular cell neoplasm, with a cribriform pattern, solid areas with a spindle cell component, and multifocality, and proposed that multicentricity should raise suspicion of FAP [51]. In 1997, Liaw and colleagues identified germline PTEN variants in Cowden disease, then termed "inherited breast and thyroid cancer syndrome" [52]. That same year, Everett and colleagues identified the PDS (SLC26A4) gene in Pendred syndrome [53], and in 2000 Kirschner and colleagues described PRKAR1A variants in Carney complex [54].

The genomic era produced findings of a different kind. In 2009, the Icelandic genome-wide association study by Gudmundsson and colleagues identified common variants at 9q22.33 (near FOXE1) and 14q13.3 (NKX2-1) associated with PTC and FTC in European populations, each with a modest effect [6]; the same year, Ngan and colleagues described the germline NKX2-1 variant p.Ala339Val (historical numbering) in patients from Hong Kong with MNG and PTC [8]. These were followed by SRGAP1 in 2013, through linkage in 38 families [7], SRRM2 in 2015 [55], and SEC23B in 2015, in the setting of Cowden syndrome without a PTEN variant [56]. The HABP2 p.Gly534Glu variant, presented in 2015 as a cause of FNMTC [22], was refuted by case-control studies and by reanalysis of the original data, and became an example of the pitfalls of assigning familial cancer genes on the basis of exome sequencing [9, 57, 58]. In 2016, Rutter and colleagues documented the direct association between germline DICER1 variants and familial DTC without prior chemotherapy [59], and in 2024 DeBoy and colleagues characterised a syndromic subtype of PTC associated with germline telomere-lengthening variants [10].

1.3. Nomenclature and reclassification

The former "cribriform-morular variant of papillary carcinoma" is now termed cribriform-morular thyroid carcinoma (CMTC). The 2025 ATA guidelines emphasise that the tumour cells are positive for NKX2-1 (TTF-1) and negative for thyroglobulin, which has prompted hypotheses of a thymic or ultimobranchial pouch origin, and that aberrant β-catenin immunoreactivity is strong diagnostic evidence [4]. Oncocytic carcinoma is no longer a subtype of follicular carcinoma, and the ATA assumes that it shares the syndromic associations of FTC [4]. The term "Cowden syndrome" has been incorporated into PHTS, which also includes Bannayan-Riley-Ruvalcaba syndrome, allelic at the same gene, whose carriers should be followed under a similar protocol [15, 60]. In the literature, the same variant often appears with legacy or abbreviated protein numbering (e.g., "G534E" in HABP2 or "A339V" in NKX2-1); the Human Genome Variation Society (HGVS)–standardised names with versioned MANE (Matched Annotation from NCBI and EMBL-EBI) transcripts are given in 2.9 [61].

2. Molecular Genetics and Pathophysiology

2.1. Genetic architecture of nonsyndromic FNMTC

The heritability of NMTC is among the highest of all cancers [46]. In the Swedish Family-Cancer Database, covering 10.5 million individuals, the standardised incidence ratio (SIR) for PTC in offspring was 3.21 when a parent was affected and 6.24 when a sibling was affected, reaching 11.19 in sisters of affected women [62]. In the pooled cohort of five Nordic countries, twins had a risk of concordant PTC 23 times higher than expected [63]. Despite this clustering, only about 5% of FNMTC corresponds to syndromic forms with a known gene, and the vast majority of nonsyndromic families have no identified cause [5, 64]. Current reviews describe nonsyndromic FNMTC as a polygenic disease, with multiple low- to moderate-penetrance genes and incomplete penetrance [13]. The 2025 ATA guidelines summarise the state of the field: in the multigenerational families studied by linkage and sequencing, candidate gene variants are nonrecurrent and appear "private" to each family, and the data do not support their inclusion in clinical panels; most families show an autosomal dominant transmission pattern [4].

2.2. Linkage loci

The review by Yang and Ngeow places TCO at "19q13.2," but the original article by Canzian and colleagues maps the locus to 19p13.2, the designation adopted here [5, 27].

2.3. Genome-wide association loci (common variants)

The Icelandic genome-wide association study (192 cases and 37,196 controls, with replication in European populations) identified rs965513 at 9q22.33, near FOXE1 (thyroid transcription factor 2), with an odds ratio (OR) of 1.75 (P=1.7×10⁻²⁷), and rs944289 at 14q13.3, a region that includes NKX2-1 and the noncoding transcript PTCSC3, with an OR of 1.37 (P=2.0×10⁻⁹); about 3.7% of individuals homozygous for both risk alleles had a 5.7-fold higher risk, and the risk alleles were associated with lower thyroid-stimulating hormone (TSH) levels [6]. In Japan (507 sporadic PTCs and 2766 controls), rs965513 had an OR of 1.69 (95% confidence interval [95% CI], 1.29–2.21) and rs944289 an OR of 1.21 (1.04–1.39), and rs965513 was associated mainly with BRAF p.Val600Glu PTCs [65]. Orphanet lists FOXE1, NKX2-1, and DIRC3 as major susceptibility factors for DTC [14]. These variants modify population risk and do not constitute a mendelian disease; the established mendelian disorder of FOXE1 is Bamforth-Lazarus syndrome, which is autosomal recessive (congenital hypothyroidism due to dysgenesis, cleft palate, and spiky hair; ORPHA:1226) [12, 14]. The review by Hińcza and colleagues lists 14 genes associated with nonsyndromic FNMTC (DICER1, FOXE1, PTCSC2, MYH9, SRGAP1, HABP2, BRCA1, CHEK2, ATM, RASAL1, SRRM2, XRCC1, NKX2-1, PTCSC3) and the rs966423 polymorphism in DIRC3 as a possible prognostic factor [23].

2.4. Candidate genes in nonsyndromic FNMTC

HABP2 (10q25.3) encodes factor VII–activating protease (FSAP), a plasma serine protease secreted by hepatocytes. In 2015, Gara and colleagues described the p.Gly534Glu variant ("Marburg I" allele, rs7080536) in seven affected members of one family and in 4.7% of 423 patients with thyroid cancer, with increased tumour expression and a loss of tumour-suppressor effect attributed to the variant [22]. In the British Isles, the variant was present in 4.2% of NMTC cases and 4.6% of controls (OR 0.896; 95% CI 0.746–1.071; P=0.233), with no association in any histological subtype [9]. In Ohio, it was present in 6.1% of familial cases, 8.0% of sporadic cases, and 8.7% of controls, did not cosegregate in one large and six small families, and HABP2 is not expressed in the thyroid (highest expression in the liver) [57]. Negative replications followed in Middle Eastern, Hispanic, Italian, Brazilian, Spanish, and Polish series, along with critical letters [12, 66, 67]. Reanalysis of the original exome data revealed errors in the interpretation of allele frequencies across populations, of the cosegregation probability, of the functional predictions, and of the biological support [58]. The same variant is associated with venous thromboembolism (OR 3.5 in a German series of 213 patients) and is common in the population [12]. OMIM keeps entry 616535 as provisional; a germline p.Gly534Glu finding should not be used to estimate thyroid cancer risk [12].

SRGAP1 (12q14.2), a SLIT-ROBO GTPase-activating protein, was identified at the 12q14 linkage peak in 38 families with PTC, followed by association studies in two cohorts; the missense variants p.Gln149His and p.Ala275Thr, in the F-BAR/FCH domain, each cosegregated in one family, and p.Arg617Cys, in the RhoGAP domain, in another, and p.Gln149His and p.Arg617Cys impaired CDC42 inactivation [7]. The authors regarded it as a probable low-penetrance or modifier gene, and a heterozygous SRGAP1 variant in a genomic report should not be treated as an established risk [7, 12].

NKX2-1 (14q13.3; TTF-1, TITF-1) encodes the transcription factor that activates thyroglobulin, thyroid peroxidase, and the TSH receptor; its haploinsufficiency causes brain-lung-thyroid syndrome [21]. Ngan and colleagues found the missense variant c.1016C>T p.(Ala339Val) in blood from 4 of 20 patients (20%) from Hong Kong with MNG and PTC; it was absent in 284 PTCs without MNG and in 349 controls, showed dominant transmission in two families (MNG with PTC or colon cancer), was associated with more advanced tumours in carriers (P=0.022), and in vitro caused increased proliferation, TSH-independent growth, STAT3 activation, and reduced transcription of thyroglobulin, TSHR, and PAX8 [8]. The GeneReviews NKX2-1 chapter cites this study among those describing somatic variants in sporadic PTC, but the article's own title and methods (blood DNA and familial segregation) establish that the variant is germline, which is the interpretation adopted here [8, 21]. The historical designation uses the short isoform NM_003317.4; in the MANE transcript NM_001079668.3, the same variant corresponds to c.1106C>T p.(Ala369Val), a conversion obtained with VariantValidator [61].

FOXE1 (9q22.33), in addition to the common association locus, has a rare variant described in Portuguese families: at the Portuguese Institute of Oncology in Lisbon, sequencing of 60 probands from families with FNMTC identified nine polymorphisms and the novel variant c.743C>G p.(Ala248Gly), which cosegregated in one family, was found in one sporadic NMTC, and promoted proliferation and migration in vitro [68]. SRRM2, a splicing factor, had the variant c.1037C>T p.(Ser346Phe) cosegregating with PTC in a family with six affected members; it was absent in 138 other families and present in 7 of 1170 sporadic PTCs versus 0 of 1404 controls (p=0.004), with altered alternative splicing of 1642 exons [55]. NOP53 p.Asp31His (minor allele frequency 1.8%) was present in all affected members of three Spanish families with nonsyndromic FNMTC, showed oncogenic function in vitro, and was classified as a probable low-penetrance modifier locus [69]. SEC23B, a component of the COPII complex, had the heterozygous variant c.1781T>G p.(Val594Gly) cosegregating in a multigenerational family with Cowden syndrome and thyroid cancer, and heterozygous variants in 3 of 96 (3%) probands with Cowden syndrome without a PTEN variant and with thyroid cancer [56]; the gene-disease relationship is classified as limited, and a heterozygous carrier (e.g., the parent of a child with congenital dyserythropoietic anaemia type II) should not be told that they have an oncological risk [12]. MINPP1 is listed in Orphanet as a candidate gene tested in familial papillary or follicular carcinoma, without clinical curation for cancer [1, 12].

Studies from 2025 and 2026 broadened the spectrum without establishing new screening genes. In a 94-gene panel applied to 48 Portuguese probands with nonsyndromic FNMTC, 15 families had likely pathogenic germline variants in DNA repair genes (ATM, CHEK2, ERCC2, BRCA2, ERCC4, FANCA, FANCD2, FANCF, PALB2) and in DICER1, FLCN, PTCH1, BUB1B, and RHBDF2, with MAPK pathway activation during tumour progression [70]. Genome sequencing of an Italian family prioritised variants in RET, GALNT10, UBN1, and PTGIR, with in silico evidence only [71]. In France, two families with three or more affected first-degree relatives across two generations carried P2RX5 p.Leu32Gln (frequency of 0.43% in European controls, consistent with low penetrance or no association) and MST1R c.345del, which encodes the RON receptor [72]. In 17 Ohio families with PTC, 56 of 121 rare noncoding variants cosegregated with all affected members and were absent in unaffected members, including candidates in five of six families without a coding variant [46].

2.5. Long telomere syndrome

Germline variants in shelterin complex genes that lengthen telomeres cause a multiple-cancer predisposition that can include PTC [4]. In 470 individuals, DeBoy and colleagues found pathogenic or likely pathogenic variants in POT1, TINF2, and ACD in 4.5% of familial cases and 1.5% of unselected cases; carriers had ultralong telomeres, 15 of 18 (83%) developed other cancers (melanoma, lymphoma, sarcoma), 22% of patients with PTC and melanoma had a deleterious variant, and successive generations showed longer telomeres and, in some cases, more and earlier cancers; all ten sequenced tumours had BRAF p.Val600Glu and no telomere maintenance mechanism, including the TERT promoter [10]. In a review of mendelian randomisation studies, PTC was one of 12 cancers associated with genetically long telomeres, and short telomeres were also associated with thyroid cancer, in a U-shaped relationship; the ATA considers further research necessary before making screening recommendations [4].

2.6. CHEK2

Pathogenic CHEK2 variants are frequently found on multigene panels and confer a moderate risk of breast cancer; the association with DTC is modest, with an approximately twofold risk of PTC for the common founder variants [4]. In three unselected US patient cohorts, c.1100del had frequencies of 0.92%, 1.65%, and 0.80%, with ORs of 1.71 (0.73–4.29), 2.64 (0.95–7.63), and 2.5 (0.63–8.46), and p.Ile157Thr had frequencies of 0.91%, 0.76%, and 0.80%, with ORs of 1.75 (0.74–4.39), 1.52 (0.42–4.96), and 2.31 (0.58–7.90); the Polish variant c.444+1G>A was found in a single case [11]. All intervals include 1, and the ATA notes that a carrier's DTC cannot always be attributed to the predisposition [4, 11].

2.7. Genes of the syndromic forms

In PHTS, SDHx genes and KLLN epimutation have been described in patients with a Cowden-like phenotype; in a prospective cohort of 2723 patients with Cowden or Cowden-like syndrome, the SIR for thyroid cancer was 72 (95% CI 51–99) with pathogenic PTEN variants, 63 (42–92) with SDHx variants, and 45 (26–73) with KLLN epimutation [73, 76]. The panel proposed by GeneReviews for PHTS phenotypes includes PTEN, AKT1, PIK3CA, SDHB, SDHC, SDHD, SEC23B, and WWP1 [15]. Among the polyposis syndromes, biallelic MUTYH variants (MUTYH-associated polyposis) are less frequently associated with thyroid, ovarian, and bladder cancer [77].

2.8. Somatic biology and distinguishing the germline component

Sporadic PTC harbours somatic driver alterations in BRAF, HRAS, KRAS, NRAS, TERT, and EIF1AX, and fusions involving RET (with CCDC6, NCOA4, TRIM24/27/33, ERC1, GOLGA5, PCM1, TFG, TPR), NTRK1, NTRK3, ALK, PAX8–PPARG, and ETV6 [14]. These alterations guide systemic therapy (7.5) and do not indicate hereditary predisposition. Conversely, a tumour with a biallelic somatic DICER1 variant or a somatic PRKAR1A variant does not establish the corresponding syndrome, because these alterations occur in sporadic tumours without a germline variant [12, 18]. Tumour sequencing assays are optimised for somatic detection and miss up to 8.1% of pathogenic germline variants, so confirmatory germline testing, with counselling, should precede family evaluation [4]. The ESMO Precision Medicine Working Group proposes targeted germline testing when variants in up to 40 genes are detected in the tumour, particularly in the seven most actionable (BRCA1, BRCA2, MLH1, MSH2, MSH6, PALB2, RET), based on a tumour-to-germline conversion rate above 5–10% for variant allele fractions above 30% in a cohort of 49,264 samples, of which about 2% were thyroid; the National Comprehensive Cancer Network (NCCN) specifies that a pathogenic or likely pathogenic PTEN variant detected on tumour profiling of any type should prompt evaluation of the personal and family history [4].

2.9. Standardisation of cited variants

The following designations were standardised using VariantValidator and Ensembl VEP (GRCh38) [61]. The codon numbering of the APC variants in the 2000 European series and the protein variants of SRGAP1 and SEC23B were not converted to coding DNA nomenclature, because the sources do not specify the nucleotide change or the transcript.

3. Clinical Spectrum and Manifestations by System

3.1. Nonsyndromic FNMTC

The thyroid phenotype of nonsyndromic FNMTC overlaps with that of sporadic DTC. In a multicentre European series of 20 families with two or more affected first-degree relatives (41 patients, 30 women and 11 men, mean age 40.6 years, range 18–73), 95% of tumours were PTC and two were FTC, Hashimoto thyroiditis coexisted in 24%, mean tumour size was 1.45 cm, and 29% were multifocal [26]. Orphanet describes multicentricity, multiple adenomatous nodules, and frequent capsular invasion [1], and some families have benign and malignant tumours with oxyphilia (TCO locus) [27].

The probability that a family represents chance clustering depends on the number of affected members. In Charkes's model, using data from the SEER (Surveillance, Epidemiology, and End Results) program and Mayo Clinic fine-needle aspirations, the probability of two sporadic DTC cases arising in a nine-member first-degree family was 1.25% of all families with DTC, equivalent to 39.4% of 306 published multiplex families; in families with two affected members, 62% to 69% of affected individuals would be sporadic cases, whereas in families with three or more affected members, fewer than 6% would have one sporadic case and fewer than 0.15% two; about 1 in 338 DTC cases would carry the familial trait [45]. The 2025 ATA guidelines restate these figures: a 62% probability of sporadic disease with two affected first-degree relatives and 6% or less with three or more [4].

Whether the disease is more aggressive remains an open question. Several studies have suggested earlier onset, more aggressive behaviour, and anticipation in successive generations, while others have not confirmed this relationship [4]. The meta-analysis by Yang and colleagues (21 studies; 3036 FNMTC and 10,497 sporadic cases) compared families with two affected members (FNMTC-2) and with three or more (FNMTC-3) with sporadic DTC: slightly younger age (mean difference −0.81 and −1.66 years), more multifocality (OR 1.48 and 1.91), bilaterality (1.53 and 1.48), lymph node metastases (1.16 and 1.74), distant metastases (2.43 and 3.93), and recurrence (1.29 and 1.49); compared with FNMTC-2, FNMTC-3 had smaller tumours and more multifocality (OR 1.26), bilaterality (1.40), and lymph node (1.48) and distant (1.97) metastases [25]. In the Korean Yonsei cohort (46,572 surgical patients, 3829 with FNMTC defined by family history), FNMTC showed more multifocality (39.0% vs 30.5%), bilaterality (23.5% vs 17.5%), and central lymph node metastases (41.5% vs 38.8%), slightly smaller tumours (0.9 ± 0.7 vs 1.0 ± 0.9 cm), similar recurrence (1.9% vs 2.3%; p=0.1), and overall survival of 99.6% vs 98.6%; family history affected recurrence-free survival only in the intermediate-to-high-risk group (hazard ratio [HR] 1.65) [24]. At West China Hospital (2015–2024; 224 patients from 98 families and 448 sporadic cases), FNMTC showed more multifocality (43.75% vs 34.82%), bilaterality (32.59% vs 23.88%), and lymph node metastases (52.68% vs 44.42%), with no difference in recurrence-free survival [78].

Age at diagnosis appears to distinguish families with different behaviour. In Siena, among 721 PTCs, the 95 families with two affected first-degree relatives both aged 45 years or younger had more multifocality, bilaterality, extrathyroidal extension, and worse outcomes than sporadic PTCs, whereas families with one affected member older than 45 did not differ [79]. In Sardinia, 31 FNMTC cases matched with 31 sporadic cases had a more advanced TNM stage, and FNMTC with onset at or before age 35 showed more BRAF expression, lymph node metastases, and intermediate-to-high ATA risk than late-onset disease, which in turn had more chronic lymphocytic thyroiditis [80]. Most of these studies are retrospective, come from surgical centres, use heterogeneous definitions of FNMTC, and measure treated disease; the meta-analysis shows heterogeneity, and ORs are not equivalent to absolute risks.

The PTC-RCC subtype (fPTC/PRN) was described in a three-generation family with PTC, nodular thyroid disease, and papillary renal neoplasia in two members; Orphanet records two families and onset at an advanced age [3, 49].

3.2. PHTS (Cowden syndrome)

Benign thyroid lesions (MNG, adenomatous nodules, follicular adenomas) occur in up to 75% of patients, and Hashimoto thyroiditis and lymphoid hyperplasia may reach about 40% [15, 76]. Estimates of thyroid carcinoma risk vary with method and era, and sources diverge: GeneReviews gives a lifetime risk of 33% to 35% and a median age at onset of 37 years [15]; the 2017 Oxford Desk Reference gives about 35% in the text and 20% to 50% in a table of cancer patterns [76, 81]; ERN GENTURIS 2026 lists older lifetime risks of 6% to 38%, affected by ascertainment bias, and cumulative estimates by age 70 of 17% (95% CI 8–32%) to 29% (15–53%), depending on whether index cases are excluded or included, with a median age of 42 years (interquartile range 30–48), and of 33% (24–40%) by age 80 in another cohort [32]; the 2025 international consensus gives 17% to 38% by ages 70 and 80 [82]; and the 2025 ATA syndrome table gives 3% to 10% [4]. The differences arise mainly from the denominator (index cases with severe phenotypes or carriers identified through cascade testing) and from the reference age; the most recent ascertainment-corrected cohort estimates lie between 17% and 29% by age 70. Cumulative risk is higher in women than in men [32].

Risk begins to rise at around age 10 [82]. Between 3% and 7% of carriers develop DTC before age 20, and 15% before age 30 [28]. The youngest reported age differs between sources and within GeneReviews itself: 7 years in the prospective cohort of Ngeow and colleagues, cited in the adult phenotype section of GeneReviews and in the AACR (American Association for Cancer Research) paediatric update, and 4 years in reports cited in the paediatric phenotype and counselling sections of GeneReviews and in the international consensus [15, 28, 73, 82]. In the Ngeow cohort, all six patients (16.7%) with cancer before age 18 had a pathogenic PTEN variant [73]. Thyroid nodules occur in 44% to 51% of children with PHTS; in one series, surveillance ultrasound detected DTC in 2 of 43 individuals before age 18, and, when the baseline ultrasound showed no nodules, more than 90% remained free of actionable nodules at 3 years and 85% at 6 years, with no cancer during follow-up [82].

Histology is also described inconsistently. GeneReviews, the Oxford Desk Reference, and the ERN GENTURIS 2026 risk table state that the cancer is usually follicular, rarely papillary, and never or almost never medullary [15, 32, 76]. The 2025 ATA guidelines, based on Ngeow 2011 and Laury 2011, report enrichment of FTC relative to the general population but PTC (including the invasive encapsulated follicular variant) as the most common histology, with 56% to 60% PTC and 25% to 45% FTC [4]. The two readings are compatible: the proportion of FTC is much higher than in the general population, where PTC predominates, without FTC being the majority. DTC in PHTS carries a low risk of invasive behaviour at all ages [28]. The other manifestations of the syndrome (breast, endometrium, kidney, colon, skin, macrocephaly, autism spectrum disorder) are not developed here.

3.3. Familial adenomatous polyposis and cribriform-morular thyroid carcinoma

The lifetime risk of thyroid cancer in FAP lies between 1.5% and 12% (2024 European guideline) [33], and GeneReviews gives 1% to 12% [16]. The meta-analysis by Chenbhanich and colleagues (12 studies, n=9821) found a pooled prevalence of 2.6% (95% CI 1.3–4.8), benign thyroid masses in 48.8% (33.8–64.0), and thyroid dysfunction in 6.9% (4.5–10.3); prevalence was higher in small studies, in those with ultrasound screening, and in those published after 2002 (5% vs 1%), the cancer preceded the diagnosis of FAP in 34%, mean ages at diagnosis of FAP and of cancer were 29 and 31 years, 95% were women, and histology was papillary in 83.3% [34]. Retrospective studies report 0.4% to 2.6% and prospective studies 2.6% to 11.8%, the female-to-male ratio reaches 80:1, and more than 80% of diagnoses occur between ages 18 and 35 [16]. Age at diagnosis ranges from 26 to 42 years depending on the population [33]. In the 2000 European cooperative series, all 15 patients were women, with a mean age at thyroidectomy of 24.9 years (19–39), and PTC affected about 1% to 2% of patients with FAP [74]. The recent rise in prevalence reflects longer life expectancy after prophylactic colectomy, greater attention to the thyroid, and the overall increase in thyroid cancer incidence [33].

CMTC accounts for about 0.16% of all PTCs but may represent up to 90% of PTCs in FAP in some series [33]. About 40% of patients with CMTC have FAP; no microscopic feature distinguishes the familial from the sporadic form, but multifocality is more common in the familial form, and the thyroid cancer may precede clinically detectable colonic changes in up to 40% of patients [4]. In a 2025 report, the authors cite an association with FAP in about 50% to 60% of CMTCs [83]. Harach and colleagues reported metastases or death from thyroid carcinoma in 6 of 63 cases in the literature and thyroid diagnosis preceding recognition of FAP in several [51]. The cytological features of CMTC may overlap with those of classic PTC, and the nodules are often small and do not meet ultrasound criteria for fine-needle aspiration [83].

3.4. DICER1 tumour predisposition

MNG is the most common manifestation, particularly in women [75]. In the National Cancer Institute (NCI) family cohort (145 carriers and 135 family controls from 48 families), the cumulative incidence of MNG or thyroidectomy by age 40 was 75% in women (95% CI 59–89) and 17% in men (7.3–35), versus 8% and 0% in controls, and the risk of thyroid cancer was 16 times that expected from SEER (95% CI 4.3–41), over 3937 person-years [36]. By age 20, 32% of women and 13% of men had MNG or thyroidectomy, and the risk of DTC is 16 to 24 times that of the population [17, 28]. DTC occurs in up to 10% of carriers [17]. In the international registry and the NCI study (713 carriers from 38 countries, median age at enrolment 9 years, median follow-up 35 months), there were 41 thyroid carcinomas with a median age of 13 years (6–45), three before age 8, 85% (34/40) in women, and 95% (39/41) DTC; 67% (12/18) presented with MNG, all 36 DTCs with available data were stage I under the AJCC (American Joint Committee on Cancer) classification, and there were two PDTCs, at ages 9 and 22 [35]. By 2021, about 30 patients had been described, more than 90% diagnosed before age 40, with a peak between ages 10 and 20 [75].

DICER1-associated DTC is usually encapsulated, without lymphatic or vascular invasion, extrathyroidal extension, or lymph node metastases, and the prognosis is favourable; the exceptions are PDTC, anaplastic carcinoma with a DICER1 variant, and thyroblastoma, the last described with tumour DICER1 alterations but not yet with a germline variant [17, 35, 75]. A history of pleuropulmonary blastoma (PPB) shortens latency, with DTC possible as early as five years after treatment, attributable to chemotherapy and repeated imaging [17]. Thyroid disease may arise or worsen during pregnancy [75]. Among paediatric patients, a germline DICER1 variant was found in 1 of 6 with PDTC, 4 of 15 with FTC, and 0 of 20 with PTC [4]. In the family reported by Rutter and colleagues (DICER1 c.5441C>T p.(Ser1814Leu)), two sisters aged 12 and 14 with MNG had foci of PTC at total thyroidectomy, their mother had DTC and bilateral Sertoli-Leydig cell tumours, and three other siblings had MNG on surveillance [59].

3.5. Carney complex

Up to 75% of patients have multiple thyroid nodules, mostly nonfunctioning follicular adenomas; papillary or follicular carcinoma may develop, sometimes after a long history of multiple adenomas [18]. In a series of 12 families (53 affected individuals), there were two thyroid carcinomas (one papillary and one follicular; 3.8%) and one follicular adenoma in three families; in the 11 patients screened, thyroid function was normal, but ultrasound showed hypoechoic, cystic, solid, or mixed lesions in 60% of adults and 67% of children [37]. Carriers of a PRKAR1A variant present with thyroid tumours earlier than patients without an identified variant, and thyroid carcinoma is among the causes of death in a syndrome with an average life expectancy of about 50 years [18]. The ATA classifies the lifetime risk as unknown [4].

3.6. Werner syndrome

In Werner syndrome, neoplasms occur in 44% of patients, mainly sarcomas, melanoma, and thyroid carcinoma [19]. A systematic review of 189 patients with 248 neoplasms found the most frequent to be thyroid neoplasms, melanoma, meningioma, soft tissue sarcomas, leukaemia, and primary bone tumours; in Japan, the SIR for thyroid neoplasms was 8.9 (95% CI 4.9–15.0) relative to the Osaka registry [38]. In a Japanese survey of 845 patients, there were 23 histologically confirmed thyroid carcinomas among 150 cancers, with a mean age of 39 years (49 years in the national registry), a female-to-male ratio of 2.3:1 (6.6:1 in the population), and a distribution of 35% papillary, 48% follicular, and 13% anaplastic (78%, 14%, and 2% in the Japanese population); the four genotyped FTCs had C-terminal variants and the only genotyped PTC had an N-terminal variant, and there was no excess of thyroid carcinoma in White patients [39]. Diagnosis occurs at a mean age of 39 years, with first symptoms at around age 13, and the mean age at death is 54 years [19, 84]. Recent data show a trend towards later and more epithelial cancers (breast, thyroid, lung) [19].

3.7. Pendred syndrome

Deficient pendrin impairs iodide organification and causes hypothyroidism with or without goitre; euthyroid goitre has incomplete penetrance, usually appears after age 10 in iodine-poor diets, and is rare in countries with universal salt iodisation, and the probability of goitre increases 1.1-fold with each year of age [20]. About 80% of patients have cochlear malformations (enlarged vestibular aqueduct), and many, especially children, have no goitre [85]. GeneReviews does not mention thyroid carcinoma [20]. The association rests on case reports: a follicular-variant papillary carcinoma with a somatic TP53 p.Arg175His variant, whose authors cite a risk of about 1% without a population source, and two Polish siblings with large MNG, FTC in the brother, and the SLC26A4 c.1001+1G>A variant [40, 41]. In a Dutch series of paediatric NMTC, one patient had Pendred syndrome [29]. The authors' interpretation is that carcinoma in dyshormonogenetic goitre requires a genetic alteration in addition to TSH stimulation [40].

3.8. Other entities

In brain-lung-thyroid syndrome (NKX2-1), thyroid dysgenesis occurs in about two-thirds (40 of 46 patients with overt or subclinical hypothyroidism in one series), there is an increased risk of lung carcinoma in young adults, and GeneReviews recommends thyroid palpation to screen for thyroid cancer [21]. In long telomere syndrome, PTC is associated with melanoma, lymphoma, sarcoma, and clonal haematopoiesis, and families with PTC and melanoma have the highest proportion of carriers [4, 10]. For Li-Fraumeni syndrome and SDHx variants, the Oxford Desk Reference tables flag the thyroid association as uncertain [81].

3.9. Natural history and treated disease

The risks in 3.2 to 3.6 describe carriers who were mostly without dedicated surveillance or under heterogeneous surveillance, and the FNMTC series in 3.1 describe disease already treated at surgical centres. Ultrasound screening shifts diagnosis towards smaller tumours: in the prospective National Institutes of Health (NIH) programme, screen-detected FNMTCs measured 0.7 ± 0.5 cm versus 1.5 ± 1.1 cm in index cases, with fewer central metastases (17.6% vs 51.1%), less extensive surgery, and less radioactive iodine (23.5% vs 79%) [86]. This lead-time and overdiagnosis effect must be considered when interpreting any comparison of aggressiveness.

4. Clinical Recognition and Red Flags

4.1. When to suspect nonsyndromic FNMTC

Suspicion rests on the family history. A family with three or more related individuals (first or second degree) with NMTC meets the clinical criterion used by the ATA for offering ultrasound to first-degree relatives; with two affected members, the probability of chance clustering is high, but young ages at diagnosis, particularly both at 45 years or younger, or a small family structure increase suspicion [4, 45, 79]. The history should include radiation exposure, because head and neck radiation in childhood, total body irradiation for bone marrow transplantation, and exposure to radioactive fallout in childhood or adolescence are accepted risk factors and exclude the Orphanet definition; adult occupational exposure below 0.5 Gy is not associated with thyroid cancer [1, 4].

4.2. Red flags for a syndromic form

The Oxford Desk Reference recommends looking, in any patient with thyroid cancer, for signs of PHTS such as macrocephaly (head circumference at or above the 97th percentile) and benign tumours [76]. The absence of a family history does not exclude a syndromic form, because the carcinoma may be the first manifestation, as in FAP, where thyroid cancer preceded the diagnosis in 34% of cases [34, 83].

4.3. Emergency signs

A rapidly growing neck mass raises the possibility of thyroblastoma or PDTC in a DICER1 carrier [28]. In any carrier, persistent cervical lymphadenopathy, hoarseness, dysphagia, neck pain, or cough warrants prompt evaluation [17]. A progressively enlarging nodule under active surveillance, suspected extrathyroidal extension or posterior growth, and proven lymph node metastases are indications for surgery [4].

4.4. Newborn screening

There is no newborn screening for FNMTC or for the syndromic forms described.

5. Diagnosis: Criteria, Testing, and Algorithms

5.1. FNMTC criteria

5.2. Indications for germline testing

The 2025 ATA guidelines are based on a formal process: pretest counselling, test selection, and post-test disclosure, ideally by a certified genetic counsellor or a physician experienced in cancer genetics, using telemedicine when necessary. Pretest counselling includes a three- to four-generation pedigree, estimation of the risk of being a carrier and of absolute cancer risks, psychosocial assessment, information on the suspected syndrome (inheritance and penetrance), personal, family, and insurance implications, and informed consent; post-test counselling includes the results given verbally and in writing, their significance for risk, management options, family implications, and communication with the treating physicians [4].

The sources above are cited in [4]. In the NCCN criteria for PHTS, FTC (and presumably oncocytic carcinoma) is a major criterion, and PTC, including the follicular variant, and structural thyroid lesions are minor criteria, so many presentations meet criteria for PTEN testing [4]; the older International Cowden Consortium criteria, reproduced in the Oxford Desk Reference, included nonmedullary epithelial thyroid cancer (especially follicular) as a major criterion and other thyroid lesions (adenoma, MNG) as minor criteria [76]. In children, the AACR recommends a multigenerational family history and germline PTEN and/or DICER1 testing in children with MNG or FTC, and PTEN testing in children with macrocephaly greater than 2 standard deviations associated with autism spectrum disorder, developmental delay, skin lesions, vascular malformation, or gastrointestinal polyps [28]. The Dutch series of 97 patients with paediatric NMTC (1970–2020) found a pathogenic or likely pathogenic variant in a predisposition gene in 13 (13%) — APC (1), BRCA2 (2), CHEK2 (4), DICER1 (4), HOXB13 (1), and MITF (1) — plus one patient with Pendred syndrome, and the authors recommend genetic counselling for all patients with childhood NMTC [29].

The International DICER1 Symposium suggests considering germline DICER1 testing for MNG or thyroid cancer in two or more first-degree relatives; the ATA notes that, given how common this presentation is in adults and how rare the variants are, the yield may be low and the optimal strategy remains to be defined [4]. In FAP, the 2024 European guideline recommends considering the diagnosis (genetic testing or colonoscopy) in women younger than 35 with CMTC, particularly if multifocal (low-level evidence; 89% agreement) [33]. In Carney complex, the clinical diagnosis requires two major criteria, or one suggestive finding plus a pathogenic PRKAR1A variant; the major thyroid criterion is histologically confirmed thyroid carcinoma or multiple hypoechoic nodules on ultrasound in a child younger than 18, and supplementary findings include a single benign nodule before age 18, multiple nodules after age 18, and a family history of thyroid, pancreatic, and ovarian carcinoma [18]. In Werner syndrome, the revised Japanese criteria (2009–2011 national survey) adopted as cardinal signs those with an incidence above 90%, added WRN genetic testing and Achilles tendon calcification, and removed urinary hyaluronic acid [87].

5.3. Tests and order of testing

5.4. Test limitations

PTEN sequencing may miss deletions and duplications depending on the method, and older negative tests may have missed deletions, deep intronic variants, or mosaicism; the AACR recommends considering retesting in these situations and considering somatic mosaicism when the phenotype is suggestive and blood germline testing is negative [15, 28]. More than 1200 unique pathogenic PTEN variants have been described, 76% of them truncating [28]. In NKX2-1-related disease, sequencing identifies 82% of variants, deletion/duplication analysis 16%, and chromosomal microarray 2%; these are proportions of detected variants by method and are not equivalent to clinical sensitivity [12, 21]. WRN sequencing identifies biallelic variants in about 97% of affected individuals [19]. In DICER1, mosaicism is common in severe phenotypes, and biallelic somatic variants are common in sporadic tumours, so a tumour-only finding does not establish the syndrome [12, 75]. The ACR-TIRADS and EU-TIRADS ultrasound risk stratification systems have not been validated in DICER1 [75]. A negative panel result does not exclude hereditary predisposition: most families with FNMTC have no identified gene, and the known loci have no clinical test [4, 5].

5.5. Reporting and interpretation

The report should state the gene, the versioned reference transcript (MANE Select: PTEN NM_000314.8, APC NM_000038.6, DICER1 NM_177438.3, PRKAR1A NM_002734.5, WRN NM_000553.6, SLC26A4 NM_000441.2), the variant in HGVS coding DNA and protein nomenclature, the zygosity (heterozygosity in the dominant forms; homozygosity or compound heterozygosity, with confirmed phase, in Werner syndrome and Pendred syndrome), and the class according to ACMG/AMP (American College of Medical Genetics and Genomics/Association for Molecular Pathology) criteria [19, 61]. A PRKAR1A variant of uncertain significance neither establishes nor excludes a diagnosis of Carney complex, and the same principle applies to the other genes; family segregation and functional data may allow reclassification [18]. Low-penetrance variants or those with a refuted association (HABP2 p.Gly534Glu, SRGAP1, heterozygous SEC23B) should not be reported as a syndrome or trigger surveillance [12]. CHEK2 founder variants should be reported for their breast cancer risk, with the thyroid risk described as modest [4, 11].

5.6. Tumour molecular testing

The 2025 ATA guidelines do not recommend routine preoperative genomic testing of confirmed DTC or routine postoperative molecular profiling of the surgical specimen, although the latter may refine risk when available (both conditional, low certainty) [4]. Before systemic therapy, testing for actionable alterations is recommended (strong, moderate), preferably by next-generation sequencing of tissue [4, 89].

5.7. Algorithm

The algorithm's decisions follow [4, 28, 33, 45].

6. Differential Diagnosis

6.1. Comparison table

In the table, AD denotes autosomal dominant inheritance, AR autosomal recessive, CT computed tomography, and MRI magnetic resonance imaging.

6.2. Main differentials

The most important distinction, because it is the most frequent, is between true FNMTC and the coincidence of sporadic cases, because DTC is common and ultrasound screening increases the detection of microcarcinomas. With two affected members, most families represent chance clustering, whereas with three or more, the familial trait is likely [45]. Young age of both affected members and a small family structure modify this interpretation [4, 79].

Among the syndromic forms, histology is the first discriminator. CMTC, with a cribriform architecture, spindle cell morules, aberrant nuclear β-catenin, TTF-1 positivity and thyroglobulin negativity in the morules, multifocality, and a predominance in young women, points to APC [4, 51]. FTC, multiple adenomas, and adenomatous nodules, particularly with macrocephaly, trichilemmomas, or a history of breast or endometrial cancer, point to PTEN [4, 76]. Juvenile nodular disease, paediatric FTC, and juvenile PDTC point to DICER1 [4, 17]. Nonthyroidal features complete the differential: pigmentation and myxomas in Carney complex, premature ageing in Werner syndrome, hearing loss in Pendred syndrome, and melanoma, leukaemia, sarcoma, and clonal haematopoiesis in long telomere syndrome [4, 10].

Medullary carcinoma is a histological differential, not a form of FNMTC: it derives from C cells, raises calcitonin, and, in its hereditary form, is caused by germline RET variants; all patients with medullary carcinoma should receive genetic counselling and germline RET testing (ESMO, IV, A) [42, 43]. The prevalence of pathogenic RET variants in All of Us and the UK Biobank is much higher than previously estimated, with a predominance of RET p.Val804Met and p.Val804Leu (65% of MEN2 variants in All of Us), and no carrier of these two variants had thyroid cancer [31].

Familial MNG requires its own differential: in iodine-sufficient countries, the prevalence of MNG is 4% to 10% and increases with age, and it is associated with iodine deficiency and female sex; early-onset MNG without iodine deficiency suggests a familial form, usually dominant, and MNG associated with other tumours should prompt consideration of FNMTC, DICER1, and the other entities in the table [17].

7. Treatment and Risk Management Strategies

7.1. General principles

There is no specific treatment for FNMTC. The 2025 ATA guidelines manage FNMTC and the syndromic forms with the same algorithms as sporadic DTC, and family history enters only as a modifying factor in some decisions [4]. The NIH comparative study (65 FNMTC cases matched to 170 sporadic DTC cases by age, sex, ATA risk, surgery, and date, with median follow-up of 4.73 and 5.83 years) found no difference in progression-free survival (progression in 15.4% vs 18.2%; p=0.61) or in repeat neck operations (13.8% vs 17.7%), and associated progression with ATA risk (high vs low, HR 9.2; CI 2.67–31.85) and male sex (HR 2.5; 1.11–5.6); the authors conclude that management of FNMTC should be aligned with that of DTC to avoid overtreatment [30]. The GeneReviews chapters on FAP and DICER1 likewise state that nodules and DTC, including CMTC, are managed as in sporadic disease [16, 17]. Syndrome-specific considerations are given in 7.3.

7.2. Surgery for DTC

The ATA recommends referral to a high-volume surgeon (more than 25 to 50 thyroidectomies per year), especially when extensive surgery is anticipated (strong, moderate); preoperative neck ultrasound of the central and lateral compartments in all patients undergoing surgery for malignant cytology or a malignant molecular test (strong, moderate); ultrasound-guided aspiration of suspicious lymph nodes larger than 8 to 10 mm in the smallest diameter if it would change management (strong, moderate); and contrast-enhanced computed tomography or magnetic resonance imaging when advanced or invasive disease is suspected (strong, moderate); it recommends against routine preoperative fluorodeoxyglucose positron emission tomography (FDG-PET) (strong, moderate) and against preoperative measurement of thyroglobulin (Tg) and antithyroglobulin antibodies (TgAb) (conditional, low) [4].

The recommendations in the table are from [4]. During active surveillance, surgery is indicated for biopsy-proven lymph node metastasis, growth of the primary tumour of 3 mm or more, distant metastasis, extrathyroidal extension, posterior growth, anxiety, inability to undergo follow-up, or patient preference [4]. The ATA does not recommend total thyroidectomy merely because the patient belongs to a family with FNMTC; family history is only one of the factors in the decision on completion thyroidectomy [4]. ESMO 2019 proposes active ultrasound surveillance (every 6 to 12 months) for unifocal papillary microcarcinomas up to 10 mm without extension or nodal involvement (III, B), with age as the only known predictor of growth (estimated 10-year risks of 36% under age 30, 14% between ages 30 and 50, and 6% between 50 and 60), lobectomy for selected low-risk T1a–T2 N0 tumours (IV, B), and notes that prophylactic central neck dissection in low-risk tumours varies between centres (IV, C) and may improve regional control in T3–T4 tumours (IV, C); total thyroidectomy is associated with recurrent laryngeal nerve injury in 2.5% and with temporary or permanent hypoparathyroidism in 8.1%, about twice the rate with lobectomy [43].

Perioperatively, the ATA recommends voice assessment in all patients (strong) and laryngoscopy in the presence of dysphonia, prior neck or chest surgery, posterior extension, or extensive central metastases; it accepts neuromonitoring of the recurrent laryngeal nerve and the external branch of the superior laryngeal nerve (conditional), recommends parathyroid hormone–guided calcium and vitamin D supplementation after total thyroidectomy or central neck dissection (strong, moderate), and recommends against drains (conditional, high) [4]. Serum Tg should be measured 6 to 12 weeks after total thyroidectomy, on levothyroxine or after TSH stimulation (strong, low), and the first assessment of response and risk is made within about 3 months after surgery, using the AJCC 8th edition, postoperative imaging, and Tg/TgAb [4].

7.3. Considerations specific to the syndromic forms

In PHTS, GeneReviews recommends total rather than subtotal thyroidectomy when aspiration is suspicious, because subtotal thyroidectomy leads to regrowth of thyroid tissue with similar cancer risks [15]. The 2025 international consensus does not recommend routine risk-reducing thyroidectomy and refers individual decisions to a multidisciplinary team (100% agreement; strong evidence) [82]. In DICER1, ultrasound characterises nodules and guides the decision to perform aspiration according to American College of Radiology criteria and paediatric and adult guidelines; nodules at low risk on ultrasound (cystic or mixed, isoechoic or hyperechoic, with smooth margins, without punctate echogenic foci) may be observed without biopsy, and aspiration is indicated for nodules that are solid, hypoechoic, taller than wide, with irregular, lobulated, or invasive margins, with microcalcifications, or with abnormal lymph nodes [17]. Surgery is indicated for nodules that are symptomatic, show significant growth, or have abnormal cytology under the Bethesda system; hemithyroidectomy may be considered for a unifocal nodule if the family accepts possible completion thyroidectomy, and total thyroidectomy is preferred for indeterminate or malignant cytology and bilateral nodular disease [17]. Radioactive iodine in DICER1 is reserved for distant (lung) metastases or persistent unresectable regional disease and is not indicated for intrathyroidal carcinoma without lymphatic or vascular invasion [17]. In Carney complex, treatment of malignant adenomas is surgical [18]. In FAP, Harach and colleagues advocated total thyroidectomy in 1994 because of multicentricity [51]; current GeneReviews and ATA guidance apply the criteria for sporadic DTC, which, in the presence of multifocal or bilateral disease, lead to total thyroidectomy [4, 16]. Colectomy and gastrointestinal surveillance follow the recommendations for familial adenomatous polyposis.

7.4. Radioactive iodine (RAI, 131I)

The table reproduces Table 10 of the 2025 ATA guidelines, adapted from the Martinique guidelines [4]. In the HiLo trial (438 patients, 1.1 vs 3.7 GBq, median 6.5 years), cumulative recurrence was similar, and in a multi-institutional study of 1298 low-risk patients with a median follow-up of 10.3 years, radioactive iodine did not improve overall or disease-free survival [4]. ESMO 2019 recommends against radioactive iodine for intrathyroidal DTC up to 1 cm without locoregional metastases (II, E), notes less consensus for other low-risk tumours, for which it recommends 1.1 GBq after recombinant human TSH (rhTSH) if given (I, A), accepts 1.1 to 3.7 GBq for intermediate risk (IV, B), and recommends activities of 3.7 GBq or more for high risk (IV, A) [43]. ESMO points out that the European Association of Nuclear Medicine did not endorse the 2015 ATA recommendation against routine use in low-risk disease, owing to the lack of prospective noninferiority trials, and that the ESTIMABL2 and IoN trials were addressing this question [43]. The reference activities are those in the table (ATA 2025 and ESMO 2019); the sodium iodide (131I) labels serve for comparison: the FDA label specifies 1.1–3.7 GBq (30–100 mCi) orally for the treatment of thyroid carcinoma and 3.7–7.4 GBq (100–200 mCi) for subsequent ablation of metastases; the Portuguese SmPC (Iodeto de Sódio [131I] Cápsulas T) specifies 1.85–3.7 GBq to destroy remnant tissue after thyroidectomy and 3.7–11.1 GBq for the treatment of metastases, with careful consideration of the activity in renal impairment [90, 91]. The 1.1 GBq activity recommended by the ATA and ESMO for low-risk disease falls within the US label and below the ablation range of the Portuguese SmPC, and is thus, for information, outside the conditions of that label; the other activities in the table fall within the ranges of both labels [4, 43, 91].

Preparation is with rhTSH (thyrotropin alfa), preferred over levothyroxine withdrawal for ablation or adjuvant therapy (strong, high), and with either for distant metastases (conditional, low) [4]. ESMO describes levothyroxine withdrawal for 4 to 5 weeks, ideally until TSH is 30 µIU/mL or higher, as an alternative, and prefers withdrawal in the presence of distant metastases [43]. The dose of thyrotropin alfa (Thyrogen®, in the USA and the EU) is 0.9 mg intramuscularly in the buttock, followed by a second 0.9 mg injection 24 hours later, with radioiodine on the third day [43, 92]; thyrotropin alfa can cause transient hyperthyroidism (7 to 14 days) in patients with substantial thyroid tissue or functioning metastases, with deaths reported in patients who had not undergone thyroidectomy or who had distant metastases, as well as sudden, painful enlargement of metastases, so hospitalisation should be considered in at-risk patients, and glucocorticoids if tumour expansion threatens vital structures [92]. The European EPAR authorises ablation with 3.7 GBq of radioiodine after thyrotropin alfa in patients without distant metastases [93].

The ATA recommends a low-iodine diet (less than 50 µg/day) for about 1 to 2 weeks (good practice), avoiding intravenous iodinated contrast in the preceding 3 months and amiodarone, and not restricting non-iodised salt, because of the risk of hyponatraemia, especially in older adults undergoing withdrawal, patients with metastases, those taking thiazides, or those on the diet for more than one week [4]. Post-therapy whole-body scintigraphy should be performed (strong, moderate), with single-photon emission computed tomography combined with computed tomography (SPECT/CT) when available [4]. The patient should be informed of acute and chronic salivary morbidity, nasolacrimal duct stenosis, and the risk of a second malignancy; the absolute risk attributable to radioactive iodine is small and does not justify additional screening, although there is an increase in acute myeloid leukaemia; a baseline complete blood count and renal function should precede treatment [4]. In patients older than 70 or with renal impairment, the ATA recommends against empiric activities above 5.55 GBq (150 mCi) without dosimetry (strong, moderate); for pulmonary micrometastases and avid macronodular metastases, it recommends 3.7 to 7.4 GBq, or 3.7 to 5.55 GBq above age 70, or dosimetry (whole-body retention up to 2.96 GBq at 48 hours; 200 cGy to the bone marrow) [4]. ESMO recommends repeating treatment every 6 to 12 months as long as there is uptake, with an individual decision after a cumulative activity of 600 mCi [43].

7.5. Levothyroxine and TSH target

Levothyroxine replaces thyroid function after total thyroidectomy and, at suppressive doses, reduces TSH stimulation of residual tissue. The FDA label (Synthroid®) specifies, for adult hypothyroidism, a full replacement dose of 1.6 µg/kg/day, with a lower starting dose in older adults and in patients with heart disease or at risk of atrial fibrillation, titration every 6–8 weeks, and doses rarely above 200 µg/day; in children, the starting dose ranges from 10–15 µg/kg/day (0–3 months) to 2–3 µg/kg/day (older than 12 years with incomplete growth); for TSH suppression in DTC, the dose depends on the TSH target defined by stage and clinical status [94]. The Portuguese SmPC (Eutirox®) specifies 150–300 µg/day for suppressive therapy in thyroid cancer, 100–200 µg/day for adult maintenance replacement, and a low starting dose (e.g., 12.5 µg/day) in older adults and patients with coronary disease [95]. In both labels, the final dose is adjusted to the TSH target. The 2025 ATA guidelines recommend individualising suppression below the reference range, which is more likely to be beneficial in high-risk disease (conditional, low), do not suggest prolonged suppression in low- and intermediate-risk disease without evidence of recurrence (conditional, low), and set the target according to response (Table 9 in 11.2): TSH within the reference range for excellent and indeterminate response, and below the range for biochemical or structural incomplete response [4]. With a normal target, 70% to 80% of patients undergoing lobectomy do not need levothyroxine, versus 20% to 30% with a target of 0.5 to 2.0 mIU/L; in the NTCTCSG registry (3238 patients, median 6 years), moderate suppression was associated with better survival, with no additional benefit from undetectable TSH; the risks of suppression include angina, atrial fibrillation, stroke, cardiovascular mortality, and postmenopausal osteoporosis and fracture, and, in 771 low- or intermediate-risk patients, TSH of 0.4 mIU/L or less was associated with more atrial fibrillation and osteoporosis; only 29% of 1125 patients from 21 centres had TSH within the target [4]. ESMO 2019 proposes TSH between 0.5 and 2 µIU/mL for excellent response and for low-risk patients with biochemical incomplete or indeterminate response (IV, B), mild suppression (0.1–0.5 µIU/mL) for intermediate- to high-risk patients with biochemical incomplete or indeterminate response (IV, B), and TSH below 0.1 µIU/mL for persistent structural disease (III, B) [43]. ESMO thus sets numeric values that the 2025 ATA guidelines have replaced with references to the laboratory reference range.

7.6. Radioiodine-refractory disease and systemic therapy

Radioiodine-refractory (RAIR) disease can be diagnosed only after an ablative or therapeutic activity, and patients with RAIR disease should not receive further empiric radioactive iodine [4]. Asymptomatic, stable RAIR patients remain on suppressive TSH therapy with imaging every 3 to 12 months (conditional, low) [4]; ESMO adds that the decision to start a multikinase inhibitor (MKI) depends on symptoms, tumour burden, performance status, lesion location, and progression by RECIST 1.1 criteria (a 20% increase in the sum of target lesions or a new lesion), that an isolated rise in Tg without radiological progression should not be used to select patients for systemic therapy, and that the decision rests on patient preference after discussion of benefits and risks [43]. Before systemic therapy, the ATA recommends testing for actionable tumour biomarkers (strong, moderate), with tissue biopsy preferred over circulating tumour DNA, and ESMO recommends next-generation sequencing when available (III, C) [4, 89].

The 2025 ATA hierarchy is as follows: without an actionable biomarker, lenvatinib or sorafenib, with lenvatinib preferred (strong, high) and a starting dose of 24 mg once daily in most patients (strong, high); cabozantinib as second line (strong, high); an NTRK inhibitor as first line for NTRK fusions (strong, moderate); a RET inhibitor as first line for RET fusions (strong, moderate); an ALK inhibitor for ALK fusions (strong, low); BRAF-directed therapy for the BRAF p.Val600Glu variant as first line when the patient is a poor candidate for lenvatinib (conditional, moderate) and after at least one MKI (strong, moderate); approved BRAF inhibitors should not be used for alterations other than p.Val600 (strong, moderate); for other targets, a clinical trial or lenvatinib [4]. Immunotherapy may be considered for high tumour mutational burden or mismatch repair deficiency (conditional, low); redifferentiation through MAPK pathway blockade may be used in selected patients (conditional, low), but adjuvant redifferentiation without genetic selection is not recommended (strong, moderate); cytotoxic chemotherapy remains a last resort, preferably within a trial [4]. Lenvatinib, sorafenib, and the other agents have no specific indication for the hereditary forms; their use in FNMTC follows the indication for advanced DTC.

The ATA warns that MKIs can cause fetal harm and that contraception should be advised and breastfeeding avoided; in children, lenvatinib had pharmacokinetics within the adult range at 24 mg, but it has no paediatric indication in the USA, and its use in minors is off-label [96, 97]. ESMO adds that TSH often rises during MKI therapy and should be checked monthly to maintain suppression [43].

7.7. Comparative regulatory status

The dates and indications in the table come from the labels and from the Drugs@FDA, DailyMed, and EMA records accessed on 17 Oct 2026 [89, 92, 93, 96–116]. The 27 Sep 2024, Drugs@FDA date for selpercatinib is not linked, in the record consulted, to the paediatric extension from age 2. The sources consulted do not include data on access and reimbursement in Portugal or Brazil.

ESMO 2022 assigned scores on the Magnitude of Clinical Benefit Scale (ESMO-MCBS 1.1): 3 to lenvatinib and 2 to sorafenib in 2019, as standard first line (I, A); 2 to cabozantinib after MKI (I, A); 3 to selpercatinib, pralsetinib, larotrectinib, and entrectinib (V, B), with ESCAT actionability levels of I-B for RET and I-C for NTRK [43, 89].

7.8. Pivotal trials

SELECT, DECISION, and COSMIC-311 compared each drug with placebo and not with one another, so indirect comparison of median PFS or response rates is not valid; the entry criteria (progression in the preceding 12 months in SELECT, prior anti-VEGFR therapy in COSMIC-311) differ. HRs express relative reductions in the risk of progression and should be read alongside the absolute medians. The absence of a significant overall survival benefit reflects crossover to the active drug. ESMO stresses that lenvatinib and sorafenib have never been compared head to head and that the practical choice depends on regulatory and cost-effectiveness heterogeneity [89]. None of these trials analysed patients with FNMTC or a hereditary syndrome separately. Data for RET and NTRK inhibitors come from single-arm studies, with no randomised trial results in DTC in the sources consulted, and ESMO 2019 mentioned a lenvatinib starting-dose trial (24 vs 18 mg; NCT02657369) whose result is not included in the sources consulted; the 2025 ATA guidelines retain 24 mg as the starting dose (strong, high) [4, 43].

7.9. Recurrence and localised metastatic disease

Percutaneous ethanol injection and radiofrequency ablation are alternatives in patients at high surgical risk; additional radioactive iodine can treat isolated lymph node metastases after local therapy; avid bone metastases are associated with better survival with radioactive iodine (strong, low; 3.7–7.4 GBq); with stimulated Tg below 10 ng/mL after withdrawal or below 5 ng/mL after rhTSH and no structural disease, levothyroxine alone is continued [4]. Adjuvant external beam radiotherapy is considered in selected cases at high locoregional risk, and intensity-modulated or stereotactic radiotherapy for unresectable recurrences [4]. For oligometastases (2 to 5 lesions), focal ablation is used; for local symptoms, surgery, radiotherapy, or thermal ablation; for multiple or symptomatic bone metastases, a bone-modifying agent (strong, low), whose administration every 3 months may reduce osteonecrosis of the jaw (comparison with the label in the next paragraph); and for brain metastases, resection and/or stereotactic radiosurgery [4]. ESMO accepts bisphosphonates and denosumab for bone metastases (V, B) and pulmonary metastasectomy for oligometastatic disease with good performance status (V, C), and recommends always excluding invasion of the upper aerodigestive tract in locoregional disease [43]. The ATA does not specify the unit dose, which is taken from the labels for the analogous indication: zoledronic acid (Zometa® and generics) is given as 4 mg intravenously, infused over at least 15 minutes, every 3–4 weeks to prevent skeletal-related events in advanced malignancies involving bone, with creatinine clearance above 60 mL/min and dose reduction below this value, and with daily oral supplementation of 500 mg of calcium and 400 IU of vitamin D in the European SmPC; denosumab (Xgeva®) is given as 120 mg subcutaneously every 4 weeks, with calcium and vitamin D [117–120]. The reference interval is the ATA's 3 months; the labels specify, by comparison, 3–4 weeks (zoledronic acid) and 4 weeks (denosumab), so the guideline interval is, for information, outside the conditions of the label [4].

7.10. Developments in the past 12 months (October 2025 to October 2026)

Still recruiting are the NIDDK study on natural history, screening, and genes in FNMTC (NCT01109420, since 2010, which includes unaffected individuals from age 7 with two affected first-degree relatives), the International PPB/DICER1 Registry (NCT03382158), the NCI DICER1 natural history study (NCT01247597), and the Istituto Besta paediatric PHTS natural history study (NCT06805734, expected completion July 2029); the only therapeutic trial found in PHTS was a phase 2 study of sirolimus (NCT00971789, n=18, completed in 2012), and there is no therapeutic trial targeting FNMTC on ClinicalTrials.gov. The search for new developments was limited to bibliographic databases, ClinicalTrials.gov, and FDA and EMA records, and a negative search does not prove absence.

7.11. Pregnancy, lactation, agents to avoid, and transition

In pregnancy, surgery for DTC can be deferred until after delivery in most patients; if needed, it is performed in the second trimester, with neck ultrasound at least at the start of the second trimester, magnetic resonance imaging if needed, and no ionising radiation; with incomplete response, surveillance is by ultrasound and Tg, and with excellent or indeterminate response, as outside pregnancy [4]. A negative pregnancy test is mandatory before radioactive iodine, pregnancy should be avoided for at least 6 months, and radioactive iodine is not given to breastfeeding women, being deferred for at least 3 months after breastfeeding or milk expression ends; diagnostic 123I may reveal breast uptake [4]. MKIs and targeted inhibitors require effective contraception and discontinuation of breastfeeding [97, 103, 109]. In DICER1, thyroid disease may arise or worsen during pregnancy, and SIOPE recommends monitoring thyroid function in pregnant women who have undergone thyroidectomy or have MNG [75].

The following should be avoided: intravenous iodinated contrast in the preceding 3 months and amiodarone before radioactive iodine, restriction of non-iodised salt, empiric radioactive iodine in RAIR patients, empiric activities above 5.55 GBq above age 70 or in renal impairment, and BRAF inhibitors outside BRAF p.Val600 [4]. Sorafenib is withheld at least 10 days and cabozantinib at least 3 weeks before elective surgery [97, 98]. In NKX2-1-related disease, avoidance of smoking is recommended because of the risk of lung carcinoma [21]. The AACR recommends that children with PHTS and nodules be followed at a paediatric thyroid centre of excellence and that, in adulthood, ultrasound continue every 3 years if normal, which implies an organised transition to adult care [28].

7.12. Guidelines and divergences

8. Reproductive Options and Family Counselling

8.1. Recurrence risks by scenario

The high relative risk in relatives of patients with NMTC contrasts with low absolute risks, because the baseline incidence is low; counselling should present both figures. The earlier the diagnosis of PTC in the family, the higher the SIR in younger relatives, with a tendency towards concordant ages [63].

8.2. Cascade testing and minors

When a pathogenic variant is identified, targeted testing of at-risk relatives separates carriers, who enter surveillance, from noncarriers, who return to population risk. In PHTS, testing for the familial variant in early childhood is often appropriate, because early manifestations, such as thyroid disease and cancer, guide management [15]. In DICER1, carrier siblings are at risk of PPB, lung cysts, thyroid neoplasia, ovarian tumours, and cystic nephroma, which justifies early testing [17]. In long telomere syndrome and with CHEK2 variants, cascade testing mainly informs nonthyroidal risks (melanoma, sarcoma, lymphoma; breast cancer) [10, 11]. In the nonsyndromic form, there is no predictive test; information for relatives is based on clinical history and neck examination (11.4). Heterozygous HABP2 p.Gly534Glu, SRGAP1, and SEC23B variants should not be cascade tested to estimate thyroid risk [12]. For syndromes detected in children with NMTC, the Dutch series recommends presymptomatic testing of relatives and surveillance for second cancers [29].

8.3. Mosaicism and penetrance

In DICER1, mosaicism for RNase IIIb domain missense variants is associated with more severe phenotypes, and a mosaic parent may have offspring with a constitutional variant [75]. In PHTS, somatic mosaicism should be considered when the phenotype is suggestive and blood testing is negative [28]. The population prevalence of PTEN and APC variants is much higher than estimated from clinical cases, and most carriers identified in population databases do not have thyroid cancer, indicating lower penetrance than in clinical series [31].

8.4. Prenatal and preimplantation diagnosis

In the syndromic forms with a known familial variant, prenatal diagnosis and preimplantation genetic testing are technically feasible; the sources consulted contain no specific recommendations for these syndromes in the thyroid context, and the decision involves incomplete penetrance and the severity of the other manifestations (e.g., PPB in DICER1 or colorectal cancer in FAP). In nonsyndromic FNMTC, with no identified gene, neither prenatal nor preimplantation diagnosis is possible.

8.5. Pregnancy

Radioactive iodine requires a negative pregnancy test and contraception for at least 6 months, and it is not given during breastfeeding; MKIs and targeted inhibitors are contraindicated in pregnancy because of fetal risk; surgery for DTC diagnosed during pregnancy can, in most cases, wait until delivery (7.11) [4, 96]. In DICER1 carriers, thyroid function should be monitored during pregnancy [75].

9. Special Considerations: Population Diversity and Epidemiology

9.1. What each figure measures

The annual incidence of DTC (all causes) estimated by Orphanet is 5.25 per 100,000 worldwide and 2.0 per 100,000 in Europe (RARECARE data); these figures do not estimate the frequency of FNMTC or the risk to carriers of any variant [12, 14]. The proportion of NMTC occurring in a familial setting is reported as 3% to 9% [5, 22], 5% [26, 68], 5% to 10% [4], 5% to 15% with a definition of three or more first-degree relatives [23, 64], 10% to 15% for clustering with three or more affected in PTC [10], and nearly 10% [13]. These estimates come from hospital series and reviews, depend on the definition, on how family history is elicited, and on the era, and include a fraction of chance clustering that Charkes's model estimates at 62% to 69% of families with two affected members [45]. In the largest published series, from Seoul (46,572 patients operated on for NMTC), 8.2% had a family history, with an increasing proportion over time and a peak between ages 35 and 59 [24]. Orphanet gives no prevalence for ORPHA:319487 [1].

9.2. Population prevalence of the syndromes

The table summarises White and colleagues, who counted ClinVar pathogenic or likely pathogenic variants [31]. MEN2 and PHTS are about 10 to 20 times more prevalent than estimated from clinical cases, and most carriers do not have thyroid cancer [31]. ERN GENTURIS gives a prevalence for PHTS previously estimated at 1:200,000 to 1:250,000 and now at 1:9000 to 1:15,000 [32]. The prevalence of pathogenic DICER1 variants is 1:6099 to 1:8416 in two genomic cohorts [17, 28]. Werner syndrome has a prevalence of 1:20,000 to 1:40,000 in Japan and about 1:50,000 in Sardinia, owing to a founder effect, with a global estimate of 1:380,000 to 1:1,000,000 and more than 300 patients described [19]; c.1105C>T p.(Arg369Ter) accounts for 20% to 25% of pathogenic alleles in Europeans and Japanese [19]. Pendred syndrome has a reported incidence of 7.5 to 10 per 100,000 [40].

9.3. Geography and founder populations

In the USA, population-based data come from All of Us and SEER, the NIH prospective screening series and the Ohio series underpin candidate gene research [31, 46, 57, 86], and the CHEK2 variants c.1100del and p.Ile157Thr account for 61.1% and 63.8% of pathogenic CHEK2 variants in two large laboratory databases [11]. In Europe, the Swedish Family-Cancer Database and the five-country Nordic cohort provide population-based familial risks [62, 63]; CHEK2 c.1100del is a Northern and Eastern European founder variant and c.444+1G>A a Polish one [11]; the United Kingdom provided the case-control study that refuted HABP2 [9]; Italy contributed the Siena and Sardinia series and gene prioritisation by genome sequencing [71, 79, 80]; Spain, NOP53 [69]; France, MST1R and P2RX5 [72]; the Netherlands, the paediatric series of 97 patients [29]; and Germany, the multicentre European survey of 20 families [26]. In Portugal, the Portuguese Institute of Oncology in Lisbon studied 60 families with FNMTC for FOXE1, identifying c.743C>G p.(Ala248Gly) in one family [68], and 48 families with a 94-gene panel, with likely pathogenic variants in 15 [70]. For Brazil, the sources consulted mention only a series that did not replicate the HABP2 association [12]; no Brazilian data on the frequency of FNMTC were found in the sources consulted, which does not prove that none exist. In Asia, data come from Korea (Seoul), China (West China Hospital), Japan (Werner syndrome, association loci), and Hong Kong (NKX2-1) [8, 24, 39, 65, 78]. In Japan, thyroid carcinoma in Werner syndrome is more often follicular, and there was no excess of thyroid carcinoma in White patients with Werner syndrome, suggesting a population difference [39].

9.4. Special populations

Female sex predominates in almost every series: the female-to-male ratio reaches 80:1 in FAP-associated thyroid cancer, 85% of DICER1-associated carcinomas occur in women, and MNG or thyroidectomy by age 40 affects 75% of female and 17% of male DICER1 carriers [16, 35, 36]. In the general population, females have a threefold higher risk, and thyroid cancer is the most common cancer in adults aged 18 to 33 [88]. In children, 13% of NMTCs have a pathogenic variant in a predisposition gene [29]. Patients treated for PPB with chemotherapy or repeated imaging have a shorter latency to DTC [17]. Children with PHTS and autism spectrum disorder or developmental delay may need light sedation or child life specialists for ultrasound [28]. In older adults, overtreatment of screen-detected FNMTC is an explicit concern [86].

9.5. Reference networks and patient organisations

ERN GENTURIS (https://www.genturis.eu) covers genetic tumour risk syndromes (PHTS, FAP, DICER1), authored the 2026 PHTS guideline, and maintains the GENTURIS registry [32]. Endo-ERN (https://endo-ern.eu) covers rare endocrine conditions and has a thyroid thematic group, and EURACAN (https://euracan.eu) covers rare adult solid cancers, including endocrine tumours. Patient organisations include ThyCa: Thyroid Cancer Survivors' Association (USA; https://www.thyca.org), the PTEN Foundation (https://ptenfoundation.org), and the International PPB/DICER1 Registry (https://www.ppbregistry.org), and, for Brazil, Instituto Oncoguia (https://www.oncoguia.org.br). The ATA points to the National Society of Genetic Counselors "Find a Genetic Counselor" tool (www.nsgc.org) for genetic counselling, including via telemedicine [4].

10. Illustrative Clinical Case

10.1. Cribriform-morular thyroid carcinoma as the first manifestation of familial adenomatous polyposis

In 2025, Qu and colleagues described a 27-year-old woman with no medical history who noticed an asymptomatic left neck mass; her mother had Hashimoto thyroiditis, and there was no family history of thyroid or gastrointestinal disease [83]. TSH was 3.57 mIU/L. Ultrasound showed a 1.0 × 0.8 × 0.9 cm right upper nodule, solid and hypoechoic (TI-RADS 4), and a 3.4 × 2.0 × 2.3 cm left mid nodule, mixed (TI-RADS 2), neither meeting American College of Radiology criteria for aspiration. Reassessment was delayed 2.5 years by the COVID-19 pandemic; by then the left nodule had possible new macrocalcifications (TI-RADS 3, larger than 2.5 cm), and aspiration was classified as Bethesda VI, with nuclear pseudoinclusions and grooves [83]. Figure 1 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC12150397/figure/luaf122-F1/) shows, in panels A and B, the initial left nodule, oval, well circumscribed, and of mixed echogenicity with anechoic areas, and, in panels C and D, the same nodule 2.5 years later, predominantly solid, with white arrows marking punctate hyperechoic foci; Figure 2 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12150397/figure/luaf122-F2/) shows Papanicolaou-stained cytology with sheets of follicular cells with enlarged nuclei, a nuclear pseudoinclusion, irregular contours with grooves, and chromatin clearing and margination, without a visible cribriform pattern or morules, illustrating the difficulty of distinguishing CMTC from classic PTC on cytology [83].

Because of the bilateral nodules, the patient underwent total thyroidectomy. Histology revealed multifocal CMTC, staged by the authors as T2N0a, positive for TTF-1, β-catenin, and PAX8 and negative for thyroglobulin; Figure 3 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12150397/figure/luaf122-F3/) shows diffuse nuclear staining for TTF-1 and strong cytoplasmic and nuclear staining for β-catenin in cribriform and anastomosing papillary structures [83]. She started levothyroxine 112 µg/day with a TSH target below 2.0 mIU/L and received 115 mCi of 131I after rhTSH, with stimulated Tg below 0.1 ng/mL, negative TgAb, and uptake only in the central neck. Despite the absence of a family history, she was referred for genetic counselling and gastroenterology: colonoscopy showed multiple tubular adenomas, a pathogenic APC variant was confirmed (the HGVS designation is not given in the article), and she underwent total proctocolectomy with end ileostomy; family screening is unknown [83]. The authors conclude that, given the increased risk of CMTC in FAP, ultrasound criteria for aspiration may be insufficient in this population, and they propose continued ultrasound surveillance of low-suspicion nodules [83].

The case illustrates three points: CMTC may be the first manifestation of FAP; immunohistochemistry (aberrant β-catenin, negative thyroglobulin) is decisive; and the thyroid diagnosis should lead to colonoscopy and APC testing (ATA 2C; EHTG-ESCP OEM.4) [4, 33]. Some aspects of management in the report do not match current recommendations and should not be read as routine: the 2025 ATA guidelines do not recommend routine radioactive iodine in low-risk disease and propose considering it in intermediate-risk disease at 1.1 to 3.7 GBq (30 to 100 mCi), so the 115 mCi activity exceeds the table's range for intermediate risk; the TSH target below 2.0 mIU/L corresponds to the ESMO numeric criterion and not to the reference range adopted by the ATA [4, 43]. The choice of colonic surgery follows the recommendations for familial adenomatous polyposis.

10.2. Complementary case: three siblings with nonsyndromic FNMTC

Rashid and colleagues described three siblings from Karachi (Pakistan) [121]. The first, aged 58, had a 6 × 4 cm left nodule in an MNG; frozen section showed a 6 × 6 cm follicular carcinoma, and after total thyroidectomy a 2 × 1.5 × 1 cm PTC was also identified in the right lobe, with no lymph node involvement; he received 5550 MBq (150 mCi) of 131I, and stimulated Tg rose from 19.10 to 56.56 at 6 months with negative ultrasound and scintigraphy. The second, aged 55, asymptomatic and evaluated by ultrasound screening, had a 5.0 cm classic PTC with minimal extrathyroidal extension, a contralateral Hürthle cell microcarcinoma and follicular adenoma, and a level VI micrometastasis; he received 150 mCi and was disease-free at 3 years. The third, aged 52, with chronic urticaria and antiperoxidase antibodies of 695, underwent ultrasound because of the family history, which revealed MNG; the tumour was a 7 × 5.5 × 3 cm classic PTC with capsular invasion and bilateral lymph node metastases (levels II to IV and VI), treated with 6660 MBq (180 mCi), with persistent paratracheal and level II disease (PET with an SUV of 5.2) and reoperation in 2015 [121]. The case shows histological heterogeneity within a family (follicular, papillary, oncocytic) and coexisting nodular disease; the 131I activities reflect the practice of the time and not the 2025 ATA guidelines [4, 121].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

The prognosis of treated FNMTC is close to that of sporadic DTC. In the NIH series, progression-free survival did not differ from that of matched sporadic DTC and depended on ATA risk and sex [30]; in the Korean cohort, recurrence was 1.9% vs 2.3% and overall survival 99.6% vs 98.6% [24]; in the European series of 20 families, with a mean follow-up of 7.2 years, 95% were disease-free, there was one local and pulmonary recurrence (10 and 25 years later) and one death from metastatic disease at 2 years [26]. The subgroups with worse outcomes are families with two affected members both aged 45 or younger and families with three or more affected members, in whom the meta-analysis shows more metastases and recurrence, and family history affected recurrence-free survival only in intermediate-to-high-risk disease [24, 25, 79]. Among the syndromic forms, DICER1-associated DTC presents at stage I in all cases with available data, except PDTC and thyroblastoma [17, 35]; PHTS-associated DTC carries a low risk of invasive behaviour [28]; in Carney complex, thyroid carcinoma is among the causes of death [18].

The 2025 ATA guidelines stratify the risk of structural recurrence as low (less than 10%), intermediate-low (10% to 15%), intermediate-high (16% to 30%), and high (greater than 30%), using the AJCC 8th edition, postoperative imaging, and Tg/TgAb, about 3 months after surgery (strong, moderate), and reassess risk throughout follow-up according to response to treatment [4].

11.2. Response to treatment and follow-up of the treated patient

The table reproduces Table 9 of the 2025 ATA guidelines [4]. Ultrasound of the thyroid bed and nodal compartments is the preferred imaging method (strong, moderate), lesions smaller than 8–10 mm can be followed without aspiration, Tg on levothyroxine is monitored after total thyroidectomy (strong, moderate) but not routinely after lobectomy, and, with positive TgAb, the trend is followed using the same assay and imaging is favoured [4]. In low-risk patients with an excellent response sustained for 5 to 8 years, routine ultrasound may be discontinued and follow-up continued with biochemical markers every 1 to 2 years (conditional, low); diagnostic scintigraphy is not routine with an excellent response in low- and intermediate-low-risk disease, and FDG-PET is reserved for high-risk patients with elevated Tg and negative scintigraphy [4]. ESMO accepts highly sensitive basal Tg (<0.2 ng/mL) instead of stimulated Tg to confirm an excellent response (II, B) [43]. In stable RAIR disease, imaging is repeated every 3 to 12 months, and during systemic therapy ESMO proposes imaging reassessment every 3 to 12 weeks [4, 43].

11.3. Surveillance of carriers of the syndromic forms

Sources diverge on when to start surveillance in PHTS: NCCN proposed annual ultrasound from age 7, ERN GENTURIS 2020 annual ultrasound from age 18 (earlier with a family history of early-onset thyroid cancer), the international consensus and AACR 2025 propose a baseline ultrasound at age 12 with stratified intervals, and ERN GENTURIS 2026 proposes starting between ages 10 and 18, never before age 10 in asymptomatic individuals, and lengthening the interval after two normal ultrasounds [28, 32, 82]. The AACR justifies starting at age 12 because this is the age at which DTC incidence exceeds that of the population and approaches 5%, and the 5-year delay (from age 7 previously) as carrying no anticipated risk [28]. The 2017 Oxford Desk Reference proposed ultrasound and thyroid function testing from age 16, earlier depending on family history [76]. In FAP, the European guideline stresses the lack of evidence for a dedicated programme and favours patient education and regular physical examination; young people with a negative ultrasound rarely develop cancer in the following 4 to 5 years [33]. In DICER1, ultrasound should be performed by an operator with specific experience, because multiple heterogeneous nodules increase false positives, and TIRADS systems have not been validated in this syndrome [28, 75].

11.4. Surveillance of relatives in nonsyndromic FNMTC

In the NIH prospective programme (109 individuals from 25 families, 12 with two affected members and 13 with three or more), screening detected cancer in 4.6% (2/43) of at-risk relatives from families with two affected members and in 22.7% (15/66) of those from families with three or more (p=0.01); the earliest nodule appeared at age 7 and the earliest cancer at age 18 [4, 86]. Screen-detected cancers were smaller, with fewer central metastases and less extensive treatment, and the authors warn of overtreatment, especially in older adults [86]. The USPSTF (United States Preventive Services Task Force) recommends against screening asymptomatic adults in the general population, there are no data on the effect of screening on FNMTC outcomes, and some guidelines do not recommend screening relatives [4, 13]. In families with PTC and melanoma, lymphoma, or sarcoma, or with breast cancer, thyroid screening should be preceded by evaluation for long telomere syndrome or CHEK2, for which thyroid surveillance recommendations have not been established [4, 10].

12. Summary and Concluding Remarks

Familial nonmedullary thyroid carcinoma brings together two entities with distinct clinical consequences. The syndromic forms are monogenic, with established clinically testable genes (PTEN, APC, DICER1, PRKAR1A, WRN, SLC26A4), and thyroid carcinoma is one of several components; identifying them changes surveillance of the patient and relatives to include other organs, and rests on concrete clues: cribriform-morular carcinoma in FAP, follicular carcinoma with macrocephaly in PHTS, juvenile nodular disease, paediatric follicular carcinoma, and poorly differentiated carcinoma in DICER1, pigmentation and myxomas in Carney complex, the progeroid phenotype in Werner syndrome, and hearing loss with goitre in Pendred syndrome. About 13% of children with nonmedullary carcinoma have a germline variant in a predisposition gene, which justifies genetic counselling for all paediatric patients.

The nonsyndromic form, which accounts for most families, has a polygenic basis and no validated screening gene. The linkage loci described since 1997 have not yielded causal genes, candidate genes carry variants that are private to each family or of low penetrance, and the HABP2 p.Gly534Glu variant has been refuted. The common loci at 9q22.33 and 14q13.3, long telomere syndrome, and CHEK2 variants explain small fractions of the risk. For this reason, the 2025 ATA guidelines do not recommend germline testing in nonsyndromic familial carcinoma unless nonthyroidal tumours in the family suggest it.

The number of affected individuals and the age at diagnosis define the risk. With two affected members, chance clustering explains most families; with three or more, the familial trait is likely, ultrasound screening of first-degree relatives detects cancer in about one-fifth, and the ATA allows ultrasound to be offered, with no defined starting age or interval and with the risk of overtreatment made explicit. Familial carcinomas show more multifocality and bilaterality and, in families with three or more affected members or with two young affected members, more metastases and recurrence, but survival remains high.

Treatment does not differ from that of sporadic differentiated carcinoma: risk-adapted extent of surgery and radioactive iodine, a response-adapted TSH target, and systemic therapy guided by the molecular profile in refractory disease. The syndrome-specific considerations are the preference for total thyroidectomy when surgery is indicated in PHTS, restraint with radioactive iodine for intrathyroidal DICER1-associated carcinoma, and colonoscopy in the presence of cribriform-morular carcinoma. Thyroid surveillance recommendations for the syndromes diverge between societies, particularly in PHTS (starting between ages 7 and 18, depending on the source), and the trend in the 2024 to 2026 guidelines is to start later and lengthen intervals after normal ultrasounds.

13. References

1. Orphanet. Familial papillary or follicular thyroid carcinoma (ORPHA:319487). Available from: https://www.orpha.net/en/disease/detail/319487
2. Orphanet. Familial nonmedullary thyroid carcinoma (ORPHA:319494). Available from: https://www.orpha.net/en/disease/detail/319494
3. Orphanet. Familial papillary thyroid carcinoma with renal papillary neoplasia (ORPHA:97290). Available from: https://www.orpha.net/en/disease/detail/97290
4. Ringel MD, Sosa JA, Baloch Z, Bischoff L, Bloom G, Brent GA, et al. 2025 American Thyroid Association Management Guidelines for Adult Patients with Differentiated Thyroid Cancer. Thyroid. 2025;35(8):841-985. doi:10.1177/10507256251363120 PMID:40844370. PMC13090833.
5. Peiling Yang S, Ngeow J. Familial non-medullary thyroid cancer: unraveling the genetic maze. Endocr Relat Cancer. 2016;23(12):R577-R595. doi:10.1530/ERC-16-0067 PMID:27807061.
6. Gudmundsson J, Sulem P, Gudbjartsson DF, Jonasson JG, Sigurdsson A, Bergthorsson JT, et al. Common variants on 9q22.33 and 14q13.3 predispose to thyroid cancer in European populations. Nat Genet. 2009;41(4):460-4. doi:10.1038/ng.339 PMID:19198613. PMC3664837.
7. He H, Bronisz A, Liyanarachchi S, Nagy R, Li W, Huang Y, et al. SRGAP1 is a candidate gene for papillary thyroid carcinoma susceptibility. J Clin Endocrinol Metab. 2013;98(5):E973-80. doi:10.1210/jc.2012-3823 PMID:23539728. PMC3644596.
8. Ngan ES, Lang BH, Liu T, Shum CK, So MT, Lau DK, et al. A germline mutation (A339V) in thyroid transcription factor-1 (TITF-1/NKX2.1) in patients with multinodular goiter and papillary thyroid carcinoma. J Natl Cancer Inst. 2009;101(3):162-75. doi:10.1093/jnci/djn471 PMID:19176457.
9. Sahasrabudhe R, Stultz J, Williamson J, Lott P, Estrada A, Bohorquez M, et al. The HABP2 G534E variant is an unlikely cause of familial non-medullary thyroid cancer. J Clin Endocrinol Metab. 2016;10(3):1098-1103. doi:10.1210/jc.2015-3928 PMID:26691890. PMC4803181.
10. DeBoy EA, Nicosia AM, Liyanarachchi S, Iyer SS, Shah MH, Ringel MD, et al. Telomere-lengthening germline variants predispose to a syndromic papillary thyroid cancer subtype. Am J Hum Genet. 2024;111(6):1114-1124. doi:10.1016/j.ajhg.2024.04.006 PMID:38688277. PMC11179366.
11. Brock P, Liynarachchi S, Nieminen TT, Chan C, Kohlmann W, Stout LA, et al. CHEK2 Founder Variants and Thyroid Cancer Risk. Thyroid. 2024;34(4):477-483. doi:10.1089/thy.2023.0529 PMID:38279823. PMC10998703.
12. GeneGuidance (user database). Gene records for HABP2, SRGAP1, FOXE1, NKX2-1, SEC23B, DICER1, and MINPP1: identifiers, clinical curation, and linked conditions. Extracted with ggdump.py; accessed 17 Oct 2026.
13. Capezzone M, Robenshtok E, Cantara S, Castagna MG. Familial non-medullary thyroid cancer: a critical review. J Endocrinol Invest. 2021;44(5):943-950. doi:10.1007/s40618-020-01435-x PMID:33025555. PMC8049908.
14. Orphanet. Differentiated thyroid carcinoma (ORPHA:146). Available from: https://www.orpha.net/en/disease/detail/146
15. Tramontana T, Yehia L. PTEN Hamartoma Tumor Syndrome. 2001 Nov 29 [Updated 2026 Jul 28]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1488/
16. Yen T, Stanich PP, Axell L, Patel SG. APC-Associated Polyposis Conditions. 1998 Dec 18 [Updated 2022 May 12]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1345/
17. Schultz KAP, Nelson AT, Kamihara J, Bauer AJ, Hatton J, Merideth MA, Stratton P, Harris AK, Field A, Dehner LP, Stewart DR, Hill DA, Messinger YH. DICER1-Related Tumor Predisposition. 2014 Apr 24 [Updated 2026 Mar 25]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK196157/
18. Stratakis CA. Carney Complex. 2003 Feb 5 [Updated 2023 Sep 21]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1286/
19. Oshima J, Hisama FM. Werner Syndrome. 2002 Dec 2 [Updated 2026 Mar 25]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1514/
20. Smith RJH, Azaiez H, Odell AM. SLC26A4-Related Sensorineural Hearing Loss. 1998 Sep 28 [Updated 2025 Jan 9]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1467/
21. Patel NJ, Jankovic J. NKX2-1-Related Disorders. 2014 Feb 20 [Updated 2023 Jun 29]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK185066/
22. Gara SK, Jia L, Merino MJ, Agarwal SK, Zhang L, Cam M, et al. Germline HABP2 Mutation Causing Familial Nonmedullary Thyroid Cancer. N Engl J Med. 2015;373(5):448-55. doi:10.1056/NEJMoa1502449 PMID:26222560. PMC4562406.
23. Hińcza K, Kowalik A, Kowalska A. Current Knowledge of Germline Genetic Risk Factors for the Development of Non-Medullary Thyroid Cancer. Genes (Basel). 2019;10(7). doi:10.3390/genes10070482 PMID:31247975. PMC6678600.
24. Lee CR, Kim JK, Ho J, Kang SW, Lee J, Jeong JJ, et al. Familial Non-Medullary Thyroid Carcinoma: Distinct Clinicopathological Features and Prognostic Implications in a Large Cohort of 46,572 Patients. Cancers (Basel). 2025;17(20). doi:10.3390/cancers17203381 PMID:41154436. PMC12562981.
25. Yang G, Tian Y, Zhang Y, Ran Y, Fan Y, Li P, et al. Relationship between affected family members and clinical aggressiveness in Familial non-medullary thyroid carcinoma: a systematic review and meta-analysis. Endocrine. 2025;90(3):1149-1167. doi:10.1007/s12020-025-04413-6 PMID:40908394.
26. Hillenbrand A, Varhaug JE, Brauckhoff M, Pandev R, Haufe S, Dotzenrath C, et al. Familial nonmedullary thyroid carcinoma-clinical relevance and prognosis. A European multicenter study. ESES Vienna presentation. Langenbecks Arch Surg. 2010;395(7):851-8. doi:10.1007/s00423-010-0696-0 PMID:20683623.
27. Canzian F, Amati P, Harach HR, Kraimps JL, Lesueur F, Barbier J, et al. A gene predisposing to familial thyroid tumors with cell oxyphilia maps to chromosome 19p13.2. Am J Hum Genet. 1998;63(6):1743-8. doi:10.1086/302164 PMID:9837827. PMC1377646.
28. Schultz KAP, MacFarland SP, Perrino MR, Mitchell SG, Kamihara J, Nelson AT, et al. Update on Pediatric Surveillance Recommendations for PTEN Hamartoma Tumor Syndrome, DICER1-Related Tumor Predisposition, and Tuberous Sclerosis Complex. Clin Cancer Res. 2025;31(2):234-244. doi:10.1158/1078-0432.CCR-24-1947 PMID:39540884. PMC11747828.
29. van der Tuin K, Ruano D, Knijnenburg J, van der Luijt RB, Morreau H, Links TP, et al. Clinically Relevant Germline Variants in Children With Nonmedullary Thyroid Cancer. J Clin Endocrinol Metab. 2024;109(12):e2214-e2221. doi:10.1210/clinem/dgae107 PMID:38415346. PMC11570363.
30. Chuki E, Behairy N, Auh S, Makarewicz A, Uttarkar Vikram CN, Kumari S, et al. Comparison of Progression-Free Survival in Familial Non-Medullary Thyroid Cancer and Sporadic Differentiated Thyroid Cancer Patients. Thyroid. 2025;35(6):642-651. doi:10.1089/thy.2024.0588 PMID:40392709. PMC12223381.
31. White SL, Jamil T, Bell C, Fishbein L, Haugen BR, Gignoux CR, et al. Population Prevalence of the Major Thyroid Cancer-Associated Syndromes. J Clin Endocrinol Metab. 2025;110(12):e4049-e4054. doi:10.1210/clinem/dgaf236 PMID:40231587. PMC12623018.
32. Hoogerbrugge N, Blatnik A, Lautrup CK, Hüneburg R, Turchetti D, van Altena A, et al. ERN GENTURIS cancer surveillance guideline for individuals with PTEN hamartoma tumour syndrome (PHTS). Eur J Hum Genet. 2026. doi:10.1038/s41431-026-02181-z PMID:42463809.
33. Zaffaroni G, Mannucci A, Koskenvuo L, de Lacy B, Maffioli A, Bisseling T, et al. Updated European guidelines for clinical management of familial adenomatous polyposis (FAP), MUTYH-associated polyposis (MAP), gastric adenocarcinoma, proximal polyposis of the stomach (GAPPS) and other rare adenomatous polyposis syndromes: a joint EHTG-ESCP revision. Br J Surg. 2024;111(5). doi:10.1093/bjs/znae070 PMID:38722804. PMC11081080.
34. Chenbhanich J, Atsawarungruangkit A, Korpaisarn S, Phupitakphol T, Osataphan S, Phowthongkum P. Prevalence of thyroid diseases in familial adenomatous polyposis: a systematic review and meta-analysis. Fam Cancer. 2019;18(1):53-62. doi:10.1007/s10689-018-0085-3 PMID:29663106.
35. Schultz KAP, Nelson AT, Mallinger PHR, Harris AK, Kamihara J, Baldinger S, et al. DICER1-Related Tumor Predisposition: Identification of At-risk Individuals and Recommended Surveillance Strategies. Clin Cancer Res. 2024;30(24):5681-5692. doi:10.1158/1078-0432.CCR-24-1532 PMID:39400264. PMC11649450.
36. Khan NE, Bauer AJ, Schultz KAP, Doros L, Decastro RM, Ling A, et al. Quantification of Thyroid Cancer and Multinodular Goiter Risk in the DICER1 Syndrome: A Family-Based Cohort Study. J Clin Endocrinol Metab. 2017;102(5):1614-1622. doi:10.1210/jc.2016-2954 PMID:28323992. PMC5443331.
37. Stratakis CA, Courcoutsakis NA, Abati A, Filie A, Doppman JL, Carney JA, et al. Thyroid gland abnormalities in patients with the syndrome of spotty skin pigmentation, myxomas, endocrine overactivity, and schwannomas (Carney complex). J Clin Endocrinol Metab. 1997;82(7):2037-43. doi:10.1210/jcem.82.7.4079 PMID:9215269.
38. Lauper JM, Krause A, Vaughan TL, Monnat RJ Jr. Spectrum and risk of neoplasia in Werner syndrome: a systematic review. PLoS One. 2013;8(4):e59709. doi:10.1371/journal.pone.0059709 PMID:23573208. PMC3613408.
39. Ishikawa Y, Sugano H, Matsumoto T, Furuichi Y, Miller RW, Goto M. Unusual features of thyroid carcinomas in Japanese patients with Werner syndrome and possible genotype-phenotype relations to cell type and race. Cancer. 1999;85(6):1345-52. PMID:10189141.
40. Tong GX, Chang Q, Hamele-Bena D, Carew J, Hoffman RS, Nikiforova MN, et al. Targeted Next-Generation Sequencing Analysis of a Pendred Syndrome-Associated Thyroid Carcinoma. Endocr Pathol. 2016;27(1):70-5. doi:10.1007/s12022-015-9413-4 PMID:26744121.
41. Lacka K, Maciejewski A, Stawny B, Lacki JK. Follicular thyroid cancer in a patient with Pendred syndrome. Ann Endocrinol (Paris). 2021;82(6):622-624. doi:10.1016/j.ando.2021.05.001 PMID:34118212.
42. Firth HV, Hurst JA. Multiple endocrine neoplasia (MEN). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 592–5.
43. Filetti S, Durante C, Hartl D, et al.; ESMO Guidelines Committee. Thyroid cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2019;30(12):1856-83. doi:10.1093/annonc/mdz400
44. Guilmette J, Nosé V. Hereditary and familial thyroid tumours. Histopathology. 2018;72(1):70-81. doi:10.1111/his.13373 PMID:29239041.
45. Charkes ND. On the prevalence of familial nonmedullary thyroid cancer in multiply affected kindreds. Thyroid. 2006;16(2):181-6. doi:10.1089/thy.2006.16.181 PMID:16613533.
46. Comiskey DF Jr, Liyanarachchi S, Wu J, Sheikh MS, Hendrickson IV, Brock PL, et al. Identification of Rare Noncoding Variants in Familial Nonmedullary Thyroid Carcinoma. Thyroid. 2026;36(2):169-176. doi:10.1177/10507256251408827 PMID:41589591. PMC13535597.
47. ROBINSON DW, ORR TG. Carcinoma of the thyroid and other diseases of the thyroid in identical twins. AMA Arch Surg. 1955;70(6):923-8. doi:10.1001/archsurg.1955.01270120131015 PMID:14375516.
48. Bignell GR, Canzian F, Shayeghi M, Stark M, Shugart YY, Biggs P, et al. Familial nontoxic multinodular thyroid goiter locus maps to chromosome 14q but does not account for familial nonmedullary thyroid cancer. Am J Hum Genet. 1997;61(5):1123-30. doi:10.1086/301610 PMID:9345104. PMC1716029.
49. Malchoff CD, Sarfarazi M, Tendler B, Forouhar F, Whalen G, Joshi V, et al. Papillary thyroid carcinoma associated with papillary renal neoplasia: genetic linkage analysis of a distinct heritable tumor syndrome. J Clin Endocrinol Metab. 2000;85(5):1758-64. doi:10.1210/jcem.85.5.6557 PMID:10843148.
50. McKay JD, Lesueur F, Jonard L, Pastore A, Williamson J, Hoffman L, et al. Localization of a susceptibility gene for familial nonmedullary thyroid carcinoma to chromosome 2q21. Am J Hum Genet. 2001;69(2):440-6. doi:10.1086/321979 PMID:11438887. PMC1235316.
51. Harach HR, Williams GT, Williams ED. Familial adenomatous polyposis associated thyroid carcinoma: a distinct type of follicular cell neoplasm. Histopathology. 1994;25(6):549-61. doi:10.1111/j.1365-2559.1994.tb01374.x PMID:7698732.
52. Liaw D, Marsh DJ, Li J, Dahia PL, Wang SI, Zheng Z, et al. Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet. 1997;16(1):64-7. doi:10.1038/ng0597-64 PMID:9140396.
53. Everett LA, Glaser B, Beck JC, Idol JR, Buchs A, Heyman M, et al. Pendred syndrome is caused by mutations in a putative sulphate transporter gene (PDS). Nat Genet. 1997;17(4):411-22. doi:10.1038/ng1297-411 PMID:9398842.
54. Kirschner LS, Carney JA, Pack SD, Taymans SE, Giatzakis C, Cho YS, et al. Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex. Nat Genet. 2000;26(1):89-92. doi:10.1038/79238 PMID:10973256.
55. Tomsic J, He H, Akagi K, Liyanarachchi S, Pan Q, Bertani B, et al. A germline mutation in SRRM2, a splicing factor gene, is implicated in papillary thyroid carcinoma predisposition. Sci Rep. 2015;5:10566. doi:10.1038/srep10566 PMID:26135620. PMC4488885.
56. Yehia L, Niazi F, Ni Y, Ngeow J, Sankunny M, Liu Z, et al. Germline Heterozygous Variants in SEC23B Are Associated with Cowden Syndrome and Enriched in Apparently Sporadic Thyroid Cancer. Am J Hum Genet. 2015;97(5):661-76. doi:10.1016/j.ajhg.2015.10.001 PMID:26522472. PMC4667132.
57. Tomsic J, Fultz R, Liyanarachchi S, He H, Senter L, de la Chapelle A. HABP2 G534E Variant in Papillary Thyroid Carcinoma. PLoS One. 2016;11(1):e0146315. doi:10.1371/journal.pone.0146315 PMID:26745718. PMC4706330.
58. Gerhard GS, Bann DV, Broach J, Goldenberg D. Pitfalls of exome sequencing: a case study of the attribution of HABP2 rs7080536 in familial non-medullary thyroid cancer. NPJ Genom Med. 2017;2. doi:10.1038/s41525-017-0011-x PMID:28884020. PMC5584869.
59. Rutter MM, Jha P, Schultz KA, Sheil A, Harris AK, Bauer AJ, et al. DICER1 Mutations and Differentiated Thyroid Carcinoma: Evidence of a Direct Association. J Clin Endocrinol Metab. 2016;101(1):1-5. doi:10.1210/jc.2015-2169 PMID:26555935. PMC4701837.
60. Jones KL, Jones MC, del Campo M. Bannayan-Riley-Ruvalcaba syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 742–3.
61. Freeman PJ, Wagstaff JF, Fokkema IFAC, Cutting GR, Rehm HL, Davies AC, et al. Standardizing variant naming in literature with VariantValidator to increase diagnostic rates. Nat Genet. 2024;56(11):2284-2286. doi:10.1038/s41588-024-01938-w PMID:39358598.
62. Hemminki K, Eng C, Chen B. Familial risks for nonmedullary thyroid cancer. J Clin Endocrinol Metab. 2005;90(10):5747-53. doi:10.1210/jc.2005-0935 PMID:16030170.
63. Fallah M, Pukkala E, Tryggvadottir L, Olsen JH, Tretli S, Sundquist K, et al. Risk of thyroid cancer in first-degree relatives of patients with non-medullary thyroid cancer by histology type and age at diagnosis: a joint study from five Nordic countries. J Med Genet. 2013;50(6):373-82. doi:10.1136/jmedgenet-2012-101412 PMID:23585692.
64. Kamani T, Charkhchi P, Zahedi A, Akbari MR. Genetic susceptibility to hereditary non-medullary thyroid cancer. Hered Cancer Clin Pract. 2022;20(1):9. doi:10.1186/s13053-022-00215-3 PMID:35255942. PMC8900298.
65. Matsuse M, Takahashi M, Mitsutake N, Nishihara E, Hirokawa M, Kawaguchi T, et al. The FOXE1 and NKX2-1 loci are associated with susceptibility to papillary thyroid carcinoma in the Japanese population. J Med Genet. 2011;48(9):645-8. doi:10.1136/jmedgenet-2011-100063 PMID:21730105.
66. Zhao X, Li X, Zhang X. HABP2 Mutation and Nonmedullary Thyroid Cancer. N Engl J Med. 2015;373(21):2084. doi:10.1056/NEJMc1511631 PMID:26581002.
67. Carvajal-Carmona LG, Tomlinson I, Sahasrabudhe R. RE: HABP2 G534E Mutation in Familial Nonmedullary Thyroid Cancer. J Natl Cancer Inst. 2016;108(8). doi:10.1093/jnci/djw108 PMID:27154966. PMC5017950.
68. Pereira JS, da Silva JG, Tomaz RA, Pinto AE, Bugalho MJ, Leite V, et al. Identification of a novel germline FOXE1 variant in patients with familial non-medullary thyroid carcinoma (FNMTC). Endocrine. 2015;49(1):204-14. doi:10.1007/s12020-014-0470-0 PMID:25381600.
69. Orois A, Gara SK, Mora M, Halperin I, Martínez S, Alfayate R, et al. NOP53 as A Candidate Modifier Locus for Familial Non-Medullary Thyroid Cancer. Genes (Basel). 2019;10(11). doi:10.3390/genes10110899 PMID:31703244. PMC6896177.
70. Pires C, Marques IJ, Saramago A, Moura MM, Pojo M, Cabrera R, et al. Identification of novel candidate predisposing genes in familial nonmedullary thyroid carcinoma implicating DNA damage repair pathways. Int J Cancer. 2025;156(1):130-144. doi:10.1002/ijc.35159 PMID:39251783.
71. Srivastava A, Skopelitou D, Miao B, Giagiobbe S, Paramasivam N, Kumar A, et al. Prioritization of predisposition genes for familial non-medullary thyroid cancer by whole-genome sequencing. Eur J Endocrinol. 2025;192(4):398-407. doi:10.1093/ejendo/lvaf045 PMID:40177881.
72. Do Cao C, Desailloud R, Topolinski H, Crépin M, Flament C, Leteurtre E, et al. Analysis of genetic predisposition to familial non-medullary thyroid cancer by whole genome sequencing. Ann Endocrinol (Paris). 2025;86(5):102456. doi:10.1016/j.ando.2025.102456 PMID:40939960.
73. Ngeow J, Mester J, Rybicki LA, Ni Y, Milas M, Eng C. Incidence and clinical characteristics of thyroid cancer in prospective series of individuals with Cowden and Cowden-like syndrome characterized by germline PTEN, SDH, or KLLN alterations. J Clin Endocrinol Metab. 2011;96(12):E2063-71. doi:10.1210/jc.2011-1616 PMID:21956414. PMC3232626.
74. Cetta F, Montalto G, Gori M, Curia MC, Cama A, Olschwang S. Germline mutations of the APC gene in patients with familial adenomatous polyposis-associated thyroid carcinoma: results from a European cooperative study. J Clin Endocrinol Metab. 2000;85(1):286-92. doi:10.1210/jcem.85.1.6254 PMID:10634400.
75. Bakhuizen JJ, Hanson H, van der Tuin K, Lalloo F, Tischkowitz M, Wadt K, et al. Surveillance recommendations for DICER1 pathogenic variant carriers: a report from the SIOPE Host Genome Working Group and CanGene-CanVar Clinical Guideline Working Group. Fam Cancer. 2021;20(4):337-348. doi:10.1007/s10689-021-00264-y PMID:34170462. PMC8484187.
76. Firth HV, Hurst JA. Cowden syndrome (PTEN hamartoma tumour syndrome (PHTS)). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 564–7.
77. Harrold E, Stadler ZK. Cancer predisposition — familial adenomatous polyposis. In: ESMO Handbook — Cancer Genomics. Lugano: ESMO Press. p. 27–8.
78. Yang G, Zhang Y, Xu T, Fan Y, Ran Y, Li P, et al. Clinical characteristics and prognosis of familial non-medullary thyroid carcinoma: a retrospective study of 98 families. World J Surg Oncol. 2026;24(1). doi:10.1186/s12957-026-04357-9 PMID:41987180. PMC13196012.
79. Capezzone M, Sagnella A, Pilli T, Maino F, Forleo R, Cantara S, et al. Role of Age at Diagnosis in Defining Potential Familial Nonmedullary Thyroid Cancer in Kindreds With Two Affected Members. J Clin Endocrinol Metab. 2021;106(2):e855-e865. doi:10.1210/clinem/dgaa798 PMID:33175120.
80. Mura C, Corrias S, Lecca M, Lai ML, Canu GL, Medas F, et al. Early familial non-medullary thyroid cancer: differences with late-onset and sporadic forms. Eur J Endocrinol. 2026;194(4):510-518. doi:10.1093/ejendo/lvag059 PMID:41988787.
81. Firth HV, Hurst JA. Patterns of cancer (Table A14). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 846–51.
82. Dhawan A, Baitamouni S, Liu D, Yehia L, Anthony K, McCarther A, et al. Cancer and Overgrowth Manifestations of PTEN Hamartoma Tumor Syndrome: Management Recommendations from the International PHTS Consensus Guidelines Working Group. Clin Cancer Res. 2025;31(9):1754-1765. doi:10.1158/1078-0432.CCR-24-3819 PMID:39937242. PMC12010961.
83. Qu J, Roth MY, Endo M, O'Hara RA, Tylee TS. Challenges in Diagnosing Cribriform Morular Thyroid Carcinoma. JCEM Case Rep. 2025;3(7):luaf122. doi:10.1210/jcemcr/luaf122 PMID:40496470. PMC12150397.
84. Firth HV, Hurst JA. DNA repair disorders. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 398–401.
85. Firth HV, Hurst JA. Deafness in early childhood. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 122–5.
86. Klubo-Gwiezdzinska J, Yang L, Merkel R, Patel D, Nilubol N, Merino MJ, et al. Results of Screening in Familial Non-Medullary Thyroid Cancer. Thyroid. 2017;27(8):1017-1024. doi:10.1089/thy.2016.0668 PMID:28657510. PMC5564020.
87. Takemoto M, Mori S, Kuzuya M, Yoshimoto S, Shimamoto A, Igarashi M, et al. Diagnostic criteria for Werner syndrome based on Japanese nationwide epidemiological survey. Geriatr Gerontol Int. 2013;13(2):475-81. doi:10.1111/j.1447-0594.2012.00913.x PMID:22817610.
88. National Comprehensive Cancer Network. NCCN Guidelines for Patients: Thyroid Cancer, 2026 (based on NCCN Guidelines Thyroid Carcinoma, Version 2.2026). Plymouth Meeting (PA): NCCN; 2026. Available from: https://www.nccn.org/patients/guidelines/content/PDF/thyroid-patient.pdf. Accessed 17 Oct 2026.
89. Filetti S, Durante C, Hartl DM, et al.; ESMO Guidelines Committee. ESMO Clinical Practice Guideline update on the use of systemic therapy in advanced thyroid cancer. Ann Oncol. 2022;33(7):674-84. doi:10.1016/j.annonc.2022.04.009
90. Radnostix. Sodium Iodide I-131 capsules and solution, therapeutic. Prescribing information; label version of 23 Jan 2026 (set ID 380e8e26-0625-4233-b36e-afb9f66e8a77); identical text in the HICON label (Jubilant DraxImage, 21 Jan 2025) [Internet]. openFDA. Accessed 26 Oct 2026.
91. Curium Netherlands. Iodeto de Sódio [I 131] Cápsulas T, 37–7400 MBq (iodeto (131I) de sódio) [Sodium Iodide [I 131] Capsules T, 37–7400 MBq (sodium iodide (131I))]. Summary of product characteristics; approved by INFARMED on 9 Nov 2023 [Internet]. INFARMED, Infomed. SmPC. Accessed 26 Oct 2026.
92. Genzyme Corporation. THYROGEN (thyrotropin alfa for injection). Prescribing information. DailyMed setid b52dfa36-f90b-4e19-9b5e-26db9d04df2b. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b52dfa36-f90b-4e19-9b5e-26db9d04df2b
93. European Medicines Agency. Thyrogen (thyrotropin alfa): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/thyrogen
94. AbbVie. SYNTHROID (levothyroxine sodium) tablets. Prescribing information; label version of 20 Feb 2024 (set ID 1e11ad30-1041-4520-10b0-8f9d30d30fcc) [Internet]. openFDA. Accessed 26 Oct 2026.
95. Merck. Eutirox 0,05 mg comprimidos (levotiroxina sódica) [Eutirox 0.05 mg tablets (levothyroxine sodium)]. Summary of product characteristics [Internet]. INFARMED, Infomed. SmPC. Accessed 26 Oct 2026.
96. Eisai Inc. LENVIMA (lenvatinib) capsules, for oral use. Prescribing information. DailyMed setid f4bedd21-efde-44c6-9d9c-b48b78d7ed1e; revised 09/2026. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=f4bedd21-efde-44c6-9d9c-b48b78d7ed1e
97. Bayer HealthCare Pharmaceuticals Inc. NEXAVAR (sorafenib) tablets. Prescribing information. DailyMed setid b50667e4-5ebc-4968-a646-d605058dbef0. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b50667e4-5ebc-4968-a646-d605058dbef0
98. Exelixis Inc. CABOMETYX (cabozantinib) tablets. Prescribing information. DailyMed setid 3850cce2-6137-42e5-a792-d318c4a4b3b5. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=3850cce2-6137-42e5-a792-d318c4a4b3b5
99. European Medicines Agency. Cabometyx (cabozantinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/cabometyx
100. Eli Lilly and Company. RETEVMO (selpercatinib) capsules/tablets. Prescribing information. DailyMed setid 7fa848ba-a59c-4144-9f52-64d090f4d828. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7fa848ba-a59c-4144-9f52-64d090f4d828
101. Rigel Pharmaceuticals Inc. GAVRETO (pralsetinib) capsules. Prescribing information. DailyMed setid 16edd46e-28b8-f06d-e063-6394a90ae31b. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=16edd46e-28b8-f06d-e063-6394a90ae31b
102. European Medicines Agency. Gavreto (pralsetinib): EPAR — withdrawn. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/gavreto
103. Bayer HealthCare Pharmaceuticals Inc. VITRAKVI (larotrectinib). Prescribing information. DailyMed setid 0c8ca614-58b2-4aa4-83d3-0387a8f782fd. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0c8ca614-58b2-4aa4-83d3-0387a8f782fd
104. European Medicines Agency. Vitrakvi (larotrectinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/vitrakvi
105. Genentech Inc. ROZLYTREK (entrectinib). Prescribing information. DailyMed setid c7c71b0c-2549-4495-86b6-c2807fa54908. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=c7c71b0c-2549-4495-86b6-c2807fa54908
106. European Medicines Agency. Rozlytrek (entrectinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/rozlytrek
107. E.R. Squibb & Sons LLC. AUGTYRO (repotrectinib) capsules. Prescribing information. DailyMed setid fb526827-40ba-4462-94cf-179ae3b0cb8a. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=fb526827-40ba-4462-94cf-179ae3b0cb8a
108. European Medicines Agency. Augtyro (repotrectinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/augtyro
109. Novartis Pharmaceuticals Corporation. TAFINLAR (dabrafenib). Prescribing information. DailyMed setid fee1e6b1-e1a5-4254-9f2e-a70e0f8dbdea. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=fee1e6b1-e1a5-4254-9f2e-a70e0f8dbdea
110. Novartis Pharmaceuticals Corporation. MEKINIST (trametinib). Prescribing information. DailyMed setid 0002ad27-779d-42ab-83b5-bc65453412a1. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0002ad27-779d-42ab-83b5-bc65453412a1
111. European Medicines Agency. Lenvima (lenvatinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/lenvima
112. European Medicines Agency. Nexavar (sorafenib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/nexavar
113. European Medicines Agency. Retsevmo (selpercatinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/retsevmo
114. OncLive. FDA awards traditional approval to selpercatinib for RET+ advanced solid tumors. 2026 Jul 14. Available from: https://www.onclive.com/view/fda-awards-traditional-approval-to-selpercatinib-for-ret-advanced-solid-tumors
115. European Medicines Agency. Tafinlar (dabrafenib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/tafinlar
116. European Medicines Agency. Mekinist (trametinib): EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/mekinist
117. European Medicines Agency. Zometa (zoledronic acid): EPAR — product information (summary of product characteristics, section 4.2) [Internet]. PDF. Accessed 26 Oct 2026.
118. Fresenius Kabi USA. Zoledronic acid injection (generic of Zometa®). Prescribing information; label version of 21 Sep 2023 (set ID 16fcf8c9-ef8a-4e93-9ee6-acfadd35a861) [Internet]. openFDA. Accessed 26 Oct 2026.
119. European Medicines Agency. Xgeva (denosumab): EPAR — product information (summary of product characteristics, sections 4.1 and 4.2) [Internet]. PDF. Accessed 26 Oct 2026.
120. Amgen. XGEVA (denosumab) injection, for subcutaneous use. Prescribing information; label version of 23 Jun 2026 (set ID 628f0998-1206-4001-aeee-18133aa9f3bf) [Internet]. openFDA. Accessed 26 Oct 2026.
121. Rashid MO, Haq N, Farooq S, Kiran Z, Siddique S, Pervez S, et al. Three siblings with familial non-medullary thyroid carcinoma: a case series. J Med Case Rep. 2016;10:213. doi:10.1186/s13256-016-0995-3 PMID:27484350. PMC4970263.