PTEN Hamartoma Tumour Syndrome (Cowden)

Literature and regulatory information updated: 4 October 2026.

A single molecular spectrum encompassing Cowden syndrome, Bannayan-Riley-Ruvalcaba syndrome, PTEN-related Proteus-like syndrome, adult-onset Lhermitte-Duclos disease and autism spectrum disorder with macrocephaly associated with germline PTEN variants

0. Initial Reference Table

1. Introduction

PTEN hamartoma tumour syndrome is a hereditary cancer predisposition and tissue overgrowth syndrome with autosomal dominant inheritance, caused by germline loss-of-function variants in the tumour suppressor gene PTEN. It is characterised by multiple hamartomas — benign proliferations composed of mature cells native to the tissue of origin but arranged in a disorganised manner — in the skin, mucosae, gastrointestinal tract, thyroid, breast and central nervous system, by macrocephaly, by neurodevelopmental manifestations and by elevated lifetime risks of breast, thyroid, endometrial, renal and colon cancer, as well as melanoma [1, 4]. Because external manifestations are often subtle and variable, many carriers remain undiagnosed; the classical prevalence of 1 in 200,000 has always been considered an underestimate [4], and recent population-based studies suggest figures 10 to 20 times higher [10].

The history of the syndrome began in 1963, when Kenneth Lloyd and Macey Dennis described, in the Annals of Internal Medicine, a young woman with oral papillomatosis, goitre, bilateral fibrocystic breast disease with carcinoma, skin changes and skeletal abnormalities; the patient's surname, Rachel Cowden, gave its name to the new “symptom complex involving multiple systems” [26]. In parallel, the paediatric literature accumulated descriptions of children with macrocephaly, lipomas and haemangiomas: Riley and Smith (1960) described macrocephaly, pseudopapilloedema and multiple haemangiomas; Bannayan (1971) reported lipomatosis, angiomatosis and macrocephaly; and Ruvalcaba, Myhre and colleagues (1980) associated macrencephaly, intestinal polyposis and pigmented penile lesions [3]. In 1986, Saul and Stevenson proposed that Bannayan and Ruvalcaba-Myhre syndromes were a single entity; Dvir and colleagues added Riley-Smith syndrome, and Cohen established the name Bannayan-Riley-Ruvalcaba syndrome [3].

Molecular unification occurred within a few years. In 1996, Nelen and colleagues mapped the Cowden disease gene to 10q22-23 through linkage analysis in 12 families, with a maximum lod score of 8.92 [27]. In 1997, the PTEN gene (also called MMAC1) was isolated as a tumour suppressor frequently mutated in glioblastomas and in prostate and breast cancer cell lines [28], and Liaw, Marsh and colleagues, in Charis Eng and Ramon Parsons's group, identified germline PTEN variants in four of five families with Cowden disease [8]. Germline PTEN variants in Bannayan-Zonana syndrome were described in the same year [29], and in 1999 a study of 43 individuals with BRRS (a variant in 60%) and of families with both phenotypes demonstrated that Cowden and BRRS are a single allelic entity [30]. This was followed by the demonstration of germline and mosaic variants in Proteus-like syndrome [31], the attribution of almost all cases of adult Lhermitte-Duclos disease to germline PTEN variants [32] and the association between PTEN variants and autism spectrum disorder (ASD) with extreme macrocephaly [33]. The term PHTS was then proposed to encompass all carriers, with the corresponding management implications; the 2026 European guideline recommends abandoning historical eponyms because they cause confusion [1, 2].

Clinically, the International Cowden Consortium criteria were revised by Eng in 2000 [13] and, following a systematic literature review, by Pilarski and colleagues in 2013, a version adopted by the National Comprehensive Cancer Network (NCCN) [14, 34]. In 2011, the prospective Cleveland Clinic cohort of 3,042 probands produced a clinical scoring system for selecting patients for PTEN testing [15], and the same cohort provided the first prospective estimates of lifetime cancer risk in 2012 [35]. Surveillance recommendations evolved from expert opinion to formal consensus documents: the 2020 European ERN GENTURIS guideline [16], the 2025 international recommendations for cancer and overgrowth and for neurodevelopmental manifestations [17, 18], the 2025 American Association for Cancer Research (AACR) paediatric recommendations [19] and the 2026 ERN GENTURIS update [2]. The full revision of the GeneReviews chapter, published in July 2026, was dedicated to the memory of Charis Eng, whose work defined the syndrome [1].

This article organises PHTS as a single disease with several age-dependent presentations. ASD with macrocephaly associated with PTEN is addressed here only in terms of cancer risk and overall follow-up; the diagnostic approach to the autistic child with macrocephaly, the neurocognitive profile and the yield of PTEN testing in ASD belong to the general approach to autism spectrum disorder.

2. Molecular Genetics and Pathophysiology

2.1. The PTEN gene and protein

PTEN is located at 10q23.31 and encodes the PTEN phosphatase (phosphatidylinositol-3,4,5-trisphosphate 3-phosphatase and dual-specificity phosphatase), belonging to the subclass of phosphatases capable of removing phosphate groups from tyrosine, serine and threonine residues [1]. Its main lipid substrate is PIP3: by removing the phosphate at the D3 position, PTEN counteracts phosphatidylinositol-3 kinase (PI3K) activity and prevents activation of AKT kinase and, downstream, the mTORC1 complex (mammalian target of rapamycin), restraining cell growth, survival and migration [1, 36]. PTEN is located in the cytoplasm and nucleus; nuclear PTEN mediates cell cycle arrest and participates in maintaining genomic integrity and DNA repair, whereas cytoplasmic PTEN is required for apoptosis [1, 17]. The protein also inhibits cell migration and adhesion and negatively regulates several cyclins; loss of protein phosphatase activity also dysregulates the mitogen-activated protein kinase (MAPK) pathway [1].

2.2. Variant spectrum

Germline variants are distributed throughout the gene and include missense, nonsense and splicing variants, small deletions and insertions, and large deletions [1]. Approximately 40% are located in exon 5, which encodes the central phosphatase motif; most variants are private to individual families, although recurrent variants exist, such as c.388C>T p.(Arg130Ter), c.697C>T p.(Arg233Ter) and c.1003C>T p.(Arg335Ter) (NM_000314.8) [1]. In 76% of cases, the variant predicts a truncated, absent or dysfunctional protein; many missense variants are functionally null and some act as dominant negatives on the normal allele [1]. Promoter variants that reduce gene expression and exon or whole-gene deletions complete the spectrum; in approximately 10% of individuals with BRRS without a coding sequence variant, large intragenic deletions or deletions encompassing PTEN are found [3, 4]. Microdeletions of 10q23 encompassing PTEN and BMPR1A cause juvenile polyposis of infancy, with polyposis throughout the gastrointestinal tract, protein-losing enteropathy, macrocephaly and a poor prognosis [1, 37].

Variant interpretation follows the standards of the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP), with the specific specifications of the ClinGen PTEN expert panel (for example, for assessment of the phosphatase domain, functional data and the specific phenotype), which achieved 96.8% concordance with consensus ClinVar classifications in the pilot study [38, 39]. A variant of uncertain significance neither establishes nor excludes the diagnosis [1].

2.3. Disease mechanism and second somatic event

The mechanism is loss of function, with the maximum ClinGen haploinsufficiency score [1]. In the absence or deficiency of PTEN, AKT1 phosphorylation is no longer counteracted, preventing cell cycle arrest and apoptosis [1]. A second event frequently occurs in affected tissues: vascular malformations, which affect approximately half of patients, result from somatic loss of the normal allele in endothelial cells through copy-neutral loss of heterozygosity, generating somatic uniparental disomy of the mutated allele; in the mouse model, sirolimus and the AKT inhibitor capivasertib blocked lesion growth [9]. The same principle explains Proteus-like syndrome, in which tumours and overgrowth result from a second mutation in affected tissues, and SOLAMEN syndrome (segmental overgrowth, lipomatosis, arteriovenous malformation and epidermal nevus), associated with mosaic PTEN nullizygosity [40, 41]. In Lhermitte-Duclos disease, 75% of lesions show complete or partial loss of PTEN expression with increased phosphorylated AKT in dysplastic gangliocytoma cells [32].

2.4. Tissue-specific pathophysiology

Hyperactivation of the PI3K-AKT-mTOR pathway explains phenotypic convergence with other disorders of this pathway: activating variants in AKT1, AKT3, PIK3CA and PIK3R2, and inactivating variants in pathway inhibitors such as PTEN, cause overgrowth syndromes [36]. In the developing brain, PTEN regulates cell proliferation and differentiation, neuronal migration, neurite growth, synaptogenesis and myelination, providing the basis for megalencephaly, white matter changes, malformations of cortical development and the neurobehavioural phenotype [18]. Metabolically, PTEN haploinsufficiency increases insulin sensitivity despite greater adiposity, with lower fasting insulin levels, greater glucose utilisation and a reduced risk of type 2 diabetes, creating a paradox of increased cancer and obesity risk [1]. Dysregulation of mucosal B-cell homeostasis contributes to the lymphoid hyperplasia and autoimmunity observed in approximately 40% of patients [1].

2.5. Genotype–phenotype correlations

In large series, frameshift variants are overrepresented in thyroid cancer, nonsense variants in colorectal cancer, promoter variants in breast cancer and missense variants in ASD; individuals with PTEN-related ASD have, on average, functionally less severe variants [1, 35]. In the European cohort of 455 adults, truncating variants were associated with a two- to threefold higher breast cancer risk, and phosphatase domain variants with an approximately twofold higher risk [12]. In Japanese individuals, C2 domain variants (amino acids 186–351) were associated with a higher breast cancer risk [17]. These associations are insufficiently robust to support genotype-specific surveillance recommendations [17].

2.6. Individuals with a Cowden phenotype without a PTEN variant

In the Cleveland Clinic cohort, only 9.5% of the 3,042 probands meeting relaxed clinical criteria had a pathogenic PTEN variant [15]; historically, it was assumed that up to 85% of individuals meeting strict Cowden criteria and 65% of those with BRRS would have a detectable variant, but subsequent series found approximately 25% among those meeting strict criteria [4]. In the remaining individuals, germline variants in SDHB, SDHC and SDHD, PIK3CA, AKT1, SEC23B and WWP1, and KLLN epimutations have been described, conferring cancer susceptibility without constituting PHTS [1, 4, 42]. The European guideline emphasises that surveillance recommendations apply only to PTEN variant carriers; in individuals meeting clinical criteria without a variant, surveillance should be individualised according to personal and family history [2].

3. Clinical Spectrum and Manifestations by Organ System

3.1. Overview and age distribution

PHTS has a bimodal distribution of age at diagnosis: children are identified mainly because of macrocephaly, developmental delay, ASD or vascular malformations, whereas adults are diagnosed following cancer, intestinal polyposis or a family history [1, 18]. Regardless of the historical phenotype, all germline PTEN variant carriers are considered to have the same cancer risks [1]. The following table summarises the frequency of the main manifestations.

3.2. Cowden phenotype (adult presentation)

Cowden syndrome was described as an adult genodermatosis with increased cancer risk [1]. Mucocutaneous lesions are the most consistent clinical marker and historically had pathognomonic value: facial trichilemmomas (small benign hair follicle tumours forming centrofacial and periorificial papules), acral keratoses (verrucous lesions on the backs of the hands and feet and palmoplantar keratotic pits), facial papillomatous papules and mucosal lesions with oral papillomatosis giving the gums, tongue and palate a “cobblestone” appearance [1, 4]. Lipomas, fibromas and mucocutaneous neuromas complete the presentation [14]. Oesophageal glycogenic acanthosis — multiple whitish papules due to epithelial hyperplasia with glycogen accumulation — is a mucosal feature of the syndrome [1]. Men may have testicular lipomatosis, identifiable on ultrasonography as multiple hyperechoic lesions [1].

In the breast, up to 67% of women have benign disease (fibrocystic disease, fibroadenomas) [4]. In the thyroid, multinodular goitre, adenomatous nodules and follicular adenomas are common, affecting up to 75% of patients, and Hashimoto thyroiditis may coexist [1, 4]. In the uterus, leiomyomas are very common and tend to recur after myomectomy alone [1]. Urogenital findings include a bicornuate uterus and ureteral duplication, whose true frequency is unknown [1]. Cognitively, adults show a broad spectrum of abilities, with frequent deficits in motor and executive function and memory even without a history of developmental delay [1].

3.3. Bannayan-Riley-Ruvalcaba phenotype (paediatric presentation)

BRRS was considered a childhood-onset disease characterised by macrocephaly, intestinal hamartomatous polyposis, vascular malformations, lipomas, haemangiomas and pigmented genital macules [1]. Birth weight exceeds 4 kg and length exceeds the 97th centile, with normal final adult height [3]. Macrocephaly, with normal-sized ventricles, has a mean head circumference (HC) of +4.89 standard deviations (SD), often exceeding +4.5 SD, and appears constant in cases with an identified variant [3, 43]. Developmental features include hypotonia, gross motor and language delay in 50%, mild to severe intellectual disability in 15–20%, seizures in 25% and ASD [3]; the Oxford Desk Reference reports motor delay, learning difficulties and language delay in 70%, and mild proximal myopathy in 60% [43].

Craniofacial manifestations include downslanting palpebral fissures (60%), a high palate, strabismus or amblyopia (15%), and prominent Schwalbe lines and corneal nerves (35%) [3]. Hamartomatous polyps of the ileum and colon occur in 45% and often present in childhood with intussusception, rectal prolapse or rectal bleeding, although they may only be identified in middle age [3]. Hamartomas include lipomas (75%), which may become large, vascular malformations — most commonly high-flow AVMs (10%) and mixed forms (20%) — in subcutaneous, cranial (20%) or bony (10%) locations, and encapsulated or infiltrative cutaneous angiolipomas (50%) [3]. The flat brownish macules on the glans and penile shaft are not always present at birth and become more apparent in later childhood; they should always be sought [3]. Other abnormalities include joint hyperextensibility, pectus excavatum and scoliosis (50%); frontal bossing, pseudopapilloedema, diabetes, Hashimoto thyroiditis, acanthosis nigricans, lymphangiomyomas, angiokeratomas, oral papillomas, trichilemmomas, café-au-lait macules, tongue polyps, supernumerary nipples, enlarged testes and penis, broad thumbs and great toes, hypoglycaemia and multiple neurocutaneous neuromas are occasionally observed [3]. Neonatal overgrowth does not translate into tall adult stature, and developmental delays tend to improve with age [3].

3.4. Proteus-like syndrome, SOLAMEN and PTEN hamartoma of soft tissue

Some patients formerly classified as having Proteus syndrome had PTEN variants and are now considered to have PHTS; true Proteus syndrome is caused by somatic mosaicism for the activating AKT1 variant p.(Glu17Lys) [4, 40]. The PTEN-related Proteus-like form, initially described with hemihypertrophy, lower limb asymmetry, AVMs and lipomatosis due to germline or mosaic variants [31], includes SOLAMEN syndrome, associated with mosaic nullizygosity [41]. Almost all individuals with PTEN-related Proteus-like syndrome have a de novo variant [1]. PTEN hamartoma of soft tissue (PHOST) is a benign overgrowth involving vessels, fat, connective tissue and nerves and producing localised deformities resembling those of Proteus syndrome; linear naevi may accompany the overgrowth [1]. Vascular anomalies and PHOST cause chronic pain, progressive infiltrative overgrowth, functional limitations, recurrence after intervention and a need for repeated procedures [1].

3.5. Lhermitte-Duclos disease

Lhermitte-Duclos disease, or dysplastic gangliocytoma of the cerebellum, is a slow-growing hamartoma in which dysplastic ganglion cells replace the granular and Purkinje layers of the cerebellar cortex [32]. In adults, almost all cases, even without other signs of Cowden or BRRS, are attributable to PTEN variants: in a study of 18 unselected lesions, 15 (83%) had a PTEN variant, all with adult onset, and the six cases with available germline DNA had a germline variant; the three cases without a variant had been diagnosed at ages 1, 3 and 11 [4, 32]. For this reason, any adult with LDD should be counselled as having PHTS [32]. The lesion is usually diagnosed in the second or third decade, is generally unilateral and presents with headache, ataxia, vertigo and, through compression of the fourth ventricle, obstructive hydrocephalus or syringomyelia [18]. On magnetic resonance imaging (MRI), it appears as a cerebellar mass that is hypointense on T1 and hyperintense on T2, with a “tiger-striped” cerebellar cortex, minimal enhancement and no diffusion restriction [1, 18].

3.6. Autism spectrum disorder with macrocephaly

Children with PHTS frequently have macrocephaly due to megalencephaly, often exceeding +5 SD and more pronounced than in idiopathic ASD, developmental delay and ASD; approximately half have a neurodevelopmental disorder, and psychiatric disorders (anxiety, depression, bipolar disorder, attention deficit hyperactivity disorder, obsessive-compulsive disorder and psychosis) affect 11–44% of patients [1]. The prevalence of germline PTEN variants in ASD with macrocephaly is 10–20% [4]. Patients with PTEN-associated ASD have greater impairments in intelligence quotient, attention, impulsivity, reaction time, processing speed and motor function than patients with PHTS without ASD, and greater sensory difficulties [18]. Patients with PHTS and a neurodevelopmental disorder are not protected from cancer risk: in the prospective cohort of 701 patients, 11% of individuals with a neurodevelopmental disorder had had cancer, and their lifetime cancer risk exceeded that of patients without such a disorder (hazard ratio 2.7; 95% CI 1.7–4.2) [11].

3.7. Neurological and neuroimaging manifestations

Macrocephaly due to megalencephaly may be associated with cerebellar tonsillar ectopia (Chiari type 1 malformation), which may be asymptomatic or require surgery [18]. Enlarged perivascular spaces are the most frequent neuroimaging finding; other findings include periventricular white matter changes, arachnoid cysts, pseudotumour cerebri, focal cortical dysplasia, polymicrogyria, grey matter heterotopias, cavernomas, multiple developmental venous anomalies, dural arteriovenous fistulas and meningiomas, sometimes multiple [1, 18]. Generalised or focal epilepsy is associated with structural malformations and is more frequent in patients with ASD [1]. In a retrospective study of 149 patients at the Dutch reference centre (Radboud), the prevalence of epilepsy was 6%, ASD was associated with a higher epilepsy risk, the most frequent form was focal epilepsy with impaired-awareness seizures, electroencephalography showed mainly focal or multifocal discharges with variable localisation, and MRI changes did not correlate with electroencephalographic foci; the authors recommend clinical surveillance directed at focal seizures, selective rather than screening electroencephalography, and MRI in patients with epilepsy to identify malformations amenable to surgery [44]. Migraine is frequent in adolescents and adults, and secondary headaches include obstructive hydrocephalus due to LDD, syringomyelia, dural fistulas and idiopathic intracranial hypertension [18].

3.8. Gastrointestinal tract

More than 90% of carriers undergoing upper or lower gastrointestinal endoscopy have polyps of mixed histology: hamartomatous, ganglioneuromatous, juvenile, hyperplastic and adenomatous; upper gastrointestinal polyps also occur [1, 4]. PHTS hamartomatous polyps differ from those of Peutz-Jeghers syndrome, in which an arborising network of smooth muscle bundles arises from the muscularis mucosae and divides the epithelium into lobules [1]. In surveillance studies, serrated polyps and adenomas were found in 38% and 36% of individuals at the initial colonoscopy, and hamartomatous polyps coexisted with serrated polyps or conventional adenomas in almost half of patients in one series [17]. In children, gastrointestinal polyps affect 25–30%, are usually identified between mid-childhood and adolescence because of bleeding, anaemia or abdominal pain, and may worsen the gastrointestinal symptoms associated with ASD (pain, constipation, diarrhoea); constipation and feeding difficulties are the most prevalent gastrointestinal complaints [1, 17]. Eosinophilic oesophagitis and other eosinophilic gastrointestinal disorders have been described at increased prevalence in children [1].

3.9. Organ-specific cancer risks

Risk estimates varied according to study design. Early cohorts, consisting mainly of probands ascertained after a cancer diagnosis, tend to overestimate risk; the European cohort of Hendricks and colleagues (455 patients, 372 with a median prospective follow-up of 6 years) excluded index cases with cancer from the adjusted analysis and obtained lower estimates [2, 12]. The table presents estimates from the main sources.

Breast. Breast cancer is the most frequent cancer in women with PHTS, with a risk comparable to that in BRCA1 and BRCA2 variant carriers, a greater frequency of multifocal and bilateral disease, and the steepest increase in incidence between ages 30 and 40; 15% of patients developed breast cancer before age 29 in the European cohort [2, 17]. The Oxford Desk Reference classifies PTEN among high-risk genes (lifetime risk ≥ 40%) [46]. Male breast cancer has been described but was not observed in the series of more than 3,000 Cleveland Clinic probands, and the guidelines do not recommend breast surveillance in men [1, 17].

Thyroid. Follicular histology is overrepresented compared with the general population, in which papillary histology predominates, and medullary carcinoma has not been observed [1, 4]. Paediatric thyroid cancer occurs, with cases before age 10, but usually has a good prognosis; in one study, surveillance ultrasonography detected differentiated carcinoma in 2 of 43 individuals before age 18 [2, 17]. Thyroid nodules affect 44–51% of children with PHTS [17]. Thyroid cancer risk is higher in women [2].

Endometrium. Risk begins to increase from age 35, although exceptional very early cases occur [2, 17]. Uterine leiomyomas and endometrial hyperplasia are frequent; in a Dutch series of 25 women under surveillance, hyperplasia was detected in seven, six of whom were asymptomatic [17].

Kidney. Papillary renal cell carcinoma predominates, although chromophobe and clear cell carcinomas also occur [1, 47]. The marked difference between North American (18–42% by age 80) and European (3–9% by age 70) estimates reflects the inclusion of variants of uncertain significance and the lack of adjustment for ascertainment bias in the former [2].

Colon and rectum. Colorectal cancer usually arises in the setting of previous or concomitant polyps; patients with cancer generally had more than 50 polyps, and the combined risk of colorectal cancer and adenomas is approximately ten times that of the general population [1, 2, 17]. Isolated cases of gastric cancer and small bowel carcinoid tumour have been described [17].

Melanoma. Estimates are discordant, with SIRs ranging from 6.6 to 39 according to the study [2].

Prostate and sarcomas. In the cohort of 701 patients, 30% of those who had cancer developed at least one cancer outside the classical spectrum, and 70% of these also had a cancer within the spectrum [48]. The risk of prostate adenocarcinoma is increased (SIR 4.1; 95% CI 1.9–7.8), with a median age at diagnosis of 59 years (versus 67 years in the SEER programme) and a minimum of 45 years; soft tissue sarcoma risk is also increased (SIR 10.7; 95% CI 3.9–23.7), with a median age of 46 years (versus 62), including malignant giant cell tumour of the tendon sheath, chondrosarcoma, cutaneous leiomyosarcoma, Ewing sarcoma and liposarcoma [1, 48].

Cancer in children and young adults. Testicular (seminoma, mixed tumour) and ovarian (dysgerminoma, immature teratoma) germ cell tumours, lymphomas, angiosarcoma and neuroblastoma have rarely been described [1, 48]. In the cohort of 701 patients, 15% of those who had cancer were diagnosed by age 29 [11].

Second primary cancers. In the cohort of 701 patients, 42% of those who had cancer developed a second primary cancer [11]. In an earlier prospective multicentre study, 40% of the 114 carriers with cancer had a second cancer, with an SIR of 7.74 for all cancers and a 10-year cumulative risk of second breast cancer of 29% [49]. In the European cohort, the 5- and 10-year risks of a second PHTS-related cancer were 24.5% and 45.7% in women and 14.5% and 19.8% in men; the risk of a second breast cancer after a first breast cancer was 23.3% at 5 years and 45.6% at 10 years [50].

4. Clinical Recognition and Red Flags

Recognition of PHTS rests on three simple clinical actions: measuring head circumference, examining the skin and oral mucosa, and constructing a targeted three-generation pedigree [4]. The family history should include benign breast and thyroid disease, gastrointestinal polyps, a large head and learning difficulties, as well as breast, thyroid, endometrial and renal cancers [4]. Examination should seek macrocephaly (usually exceeding +2 SD), oral “cobblestone” lesions, acral keratoses and palmoplantar pits, lipomas and vascular malformations, goitre or thyroid nodularity, and breast lumps [4]. When assessing macrocephaly, parental HCs should be recorded and adjusted for height; so-called benign familial macrocephaly should be accepted cautiously because PTEN and PTCH1 variants may present with few syndromic signs [43].

Skin lesions in patients with Birt-Hogg-Dubé syndrome (fibrofolliculomas) may be mistaken for trichilemmomas, and gastrointestinal ganglioneuromas may lead to an incorrect diagnosis of neurofibromatosis type 1; both errors should be considered in an adult with facial papules or intestinal ganglioneuromatosis [1].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic principle

The diagnosis of PHTS is established in a proband with suggestive findings and a heterozygous pathogenic or likely pathogenic PTEN variant; clinical criteria cannot establish the diagnosis on their own and serve to select individuals for testing [1]. No single biochemical, histological or imaging finding is confirmatory [1].

5.2. Historical clinical criteria (International Cowden Consortium and NCCN up to 2013)

The consensus criteria for Cowden syndrome, updated annually by the NCCN until the 2013 revision, were divided into pathognomonic, major and minor criteria [4, 13].

Source: Oxford Desk Reference, Box 4.7, based on GeneReviews and the NCCN [4].

5.3. Revised Pilarski criteria (2013), adopted by the NCCN

The systematic review by Pilarski and colleagues eliminated the pathognomonic category and defined quantitative thresholds; it found insufficient evidence to retain benign breast disease, uterine leiomyomas and genitourinary malformations as criteria and added ASD, colon cancer, oesophageal glycogenic acanthosis, penile macules, renal cell carcinoma, testicular lipomatosis and vascular anomalies [14, 34].

An operational clinical diagnosis in an individual requires three or more major criteria, one of which must be macrocephaly, LDD or gastrointestinal hamartomas, or two major and three minor criteria; in a family in which one member meets the criteria or has a PTEN variant, two major criteria (with or without minor criteria), one major and two minor criteria, or three minor criteria suffice for relatives [14, 34]. The NCCN genetic testing criteria derive from these clinical criteria [34].

5.4. Cleveland Clinic score and paediatric criteria

The Cleveland Clinic score, developed in 3,042 prospective probands, estimates the pre-test probability of finding a PTEN variant from phenotype and age [15]. In adults, a score ≥ 10 leads to a recommendation for referral to a genetics professional [4]. The score correlated with reduced PTEN protein expression, unlike the 2010 NCCN criteria, and improved sensitivity and positive predictive value compared with those criteria in both adults and children [15]. In children, PHTS should be considered when macrocephaly is associated with at least one of the following: ASD or developmental delay; dermatological findings (lipomas, trichilemmomas, oral papillomas or penile freckling); vascular findings (AVMs or haemangiomas); gastrointestinal polyps [4]. The Cleveland Clinic calculator is also used in resource-limited settings to justify counselling when testing is unavailable [52].

5.5. Molecular strategy

Source: GeneReviews, based on Yehia and colleagues and the HGMD [1].

The targeted approach begins with PTEN sequencing (exons 1 to 9 and adjacent intronic regions); if negative, this is followed by deletion/duplication analysis and then promoter sequencing [1, 4]. A multigene panel including PTEN and differential diagnosis genes (AKT1, PIK3CA, SDHB, SDHC, SDHD, SEC23B, WWP1) is an alternative that limits incidental findings [1]. Exome or genome sequencing is recommended by the ACMG and the American Academy of Pediatrics as a first- or second-line test in children with developmental delay, intellectual disability or multiple congenital anomalies, but sensitivity for deletions of one or a few exons may be lower than that of targeted methods [1]. The presence of the PTENP1 pseudogene at 9p21 requires specific primer design to avoid co-amplification [1]. When mosaic PI3K-AKT pathway disease is suspected (Proteus-like syndrome, segmental overgrowth), high-depth sequencing (> 1,000×) of blood and affected tissue DNA should be considered [36, 43]. In juvenile polyposis with signs of PHTS, PTEN should be studied first, followed by SMAD4 and BMPR1A; without signs of PHTS, the order is reversed, and the threshold for chromosomal microarray should be low [37].

In oncology, detection of PTEN alterations in the tumour may reveal a germline variant; the ESMO Handbook notes that PTEN deletions are difficult to detect in liquid biopsy, and therefore recommends next-generation sequencing of tumour tissue [53]. Tumour findings suggestive of germline origin should lead to confirmation in constitutional DNA.

5.6. Proposed algorithm

1. Identification of an individual with clinical criteria (Pilarski/NCCN), a high Cleveland Clinic score, paediatric criteria (macrocephaly with an associated finding), adult LDD or ASD with macrocephaly [4, 14, 15, 32].
2. Pre-test genetic counselling and a three-generation pedigree [1, 4].
3. PTEN sequencing with deletion/duplication analysis (alone or within a panel); if negative, promoter analysis; for neurodevelopmental phenotypes, exome or genome sequencing [1].
4. Pathogenic or likely pathogenic variant: diagnosis of PHTS, complete baseline assessment (section 7.1), cascade testing and surveillance [1].
5. Variant of uncertain significance: do not use for clinical decisions or predictive testing; reassess using ClinGen criteria, segregation studies and functional data [1, 39].
6. No variant: consider mosaicism (tissue analysis), the differential diagnosis and alternative genes; individualised surveillance based on personal and family history [2, 42].

5.7. Limitations of genetic testing and report content

Test coverage determines the interpretation of a negative result. Sequencing of the coding region and exon–intron junctions detects up to 80% of variants, copy number analysis 3–11%, and promoter sequencing approximately 10% (table in section 5.5) [1]; a panel that does not include deletion/duplication analysis or the promoter does not exclude PHTS. Deep intronic variants escape exome-targeted DNA sequencing: in a patient with a clinical diagnosis of Cowden syndrome and a negative DNA panel, targeted PTEN RNA sequencing revealed a pathogenic deep intronic variant, and the authors propose combined DNA and RNA analysis in patients meeting clinical criteria or RNA reassessment of those with negative DNA testing [54]. The PTENP1 pseudogene can generate false results if primers are not specific [1], and somatic mosaicism (Proteus-like syndrome, SOLAMEN, segmental overgrowth) requires high-depth sequencing of affected tissue [36, 41, 43]. PTEN promoter variants lie in a bidirectional region shared with KLLN, and their interpretation, like that of KLLN epimutations, requires caution [1, 42]. In the presence of a convincing phenotype, a negative result does not exclude the disease: the diagnosis of PHTS cannot then be established, but surveillance should be individualised according to personal and family history [2].

The report should include the gene, transcript with version, Human Genome Variation Society (HGVS) nomenclature at coding DNA and protein levels, zygosity, ACMG/AMP classification with the ClinGen PTEN expert panel specifications and the criteria applied, and technical limitations [38, 39]. The MANE Select transcript is NM_000314.8, encoding the 403-amino-acid protein NP_000305.3 [7]. The classical literature frequently uses abbreviated names (for example, R130X for c.388C>T, p.(Arg130Ter)) or describes variants only by exon; when citing these sources, the historical designation should be given alongside the HGVS form normalised according to ClinVar. Deletions with unknown breakpoints are described using uncertainty notation, as in the case in section 10.1: c.(492+1_493-1)_(634+1_635-1)del [51]. Detection of PTEN loss or a PTEN variant in a tumour (including the immunohistochemistry used to select patients for capivasertib in prostate cancer) does not establish germline origin, which requires confirmation in constitutional DNA [23, 53]. A variant of uncertain significance does not become pathogenic through phenotypic concordance, response to an mTOR or AKT inhibitor, or partial segregation: in a Brazilian family with intestinal polyposis in several members and early breast cancer in a sister, the intronic variant c.1026+32T>G was presented as causal on the basis of an in silico prediction of a new splicing donor site [55]; without RNA or functional analysis and without complete segregation, a variant of this kind should be classified using ACMG/AMP and ClinGen criteria before guiding predictive testing of relatives, and segregation analysis should be described as research [1, 39].

6. Differential Diagnosis

7. Treatment and Risk Management Strategies

7.1. Organisation of care and baseline assessment

Management should be provided by a multidisciplinary team, ideally experienced in PHTS, including medical genetics, gastroenterology, endocrinology, gynaecology, urology, oncology, surgery, dermatology, neurology and neurodevelopment [1, 17]. The reference guidelines are those of GeneReviews, which refers to the NCCN (version 3.2026), the 2025 international recommendations and the 2026 European ERN GENTURIS update [1, 2, 17, 18]. ClinGen actionability is strong in adults and moderate in children because the interventions with the greatest impact (breast, endometrial, renal and colorectal surveillance and risk-reducing surgery) begin in adulthood [20].

After diagnosis, the following are recommended: a complete clinical and family history; a complete physical examination; education on cancer signs and symptoms from age 18 or five years before the earliest cancer in the family; dermatological examination (including melanoma); thyroid examination and ultrasonography from age 12; in women, education on self-examination from age 18, clinical breast examination from age 25 and breast imaging from age 30 (age 25 in the international and European guidelines); renal ultrasonography (or computed tomography/MRI) from age 40; education on abnormal uterine bleeding and consideration of endometrial biopsy from age 35; assessment of gastrointestinal manifestations and colonoscopy from age 35, earlier if symptomatic; developmental assessment (motor, adaptive, cognitive and language); ASD screening from age 12 months; a neurological examination directed at signs of intracranial AVMs; assessment of hypotonia and scoliosis in the BRRS phenotype; genetic counselling and psychosocial support [1]. Routine neuroimaging is not recommended in patients without signs or symptoms; contrast-enhanced brain MRI and arterial and venous MR angiography are indicated for positional or projectile vomiting, new neurological deficits, signs of intracranial hypertension or new-onset headache in a child under age 5, with haemosiderin-sensitive sequences (gradient echo and susceptibility-weighted imaging) for cavernomas [1, 18]. HC should be measured until age 5 to confirm a stable trajectory; acceleration should raise suspicion of hydrocephalus [18].

7.2. Cancer surveillance: comparison of guidelines

The guidelines converge on intensive breast surveillance and differ regarding the age at which thyroid ultrasonography begins, endometrial and renal surveillance, and the colonoscopy interval. The table compares recommendations; for individuals with a family history of a cancer at an early age, surveillance of that organ should begin 5 to 10 years before the earliest age at diagnosis in the family [1, 34].

Evidence of surveillance benefit is strongest for breast cancer: annual MRI had a sensitivity of 100% versus 50% for mammography in a series of 39 women, mammography added no benefit, and cancers detected outside surveillance programmes were more advanced [2, 17]. In premenopausal women with regular cycles, MRI should be performed between cycle days 5 and 12 to reduce background parenchymal enhancement [17]. In the thyroid, patients without nodules on the initial ultrasound have a greater than 90% probability of remaining free of actionable nodules at 3 years and 85% at 6 years, supporting longer intervals and concern about overdiagnosis of small indolent carcinomas [2, 17]. Actionable nodules are those meeting American Thyroid Association criteria for fine-needle aspiration cytology; with benign cytology, repeating surveillance in 1–2 years is reasonable [17]. In the endometrium, transvaginal ultrasonography alone is less sensitive than biopsy for hyperplasia, ultrasonography is neither sensitive nor specific before menopause, and no survival benefit has been demonstrated; the international guideline therefore reserves it for clinical trials, whereas the European guideline accepts it as an individual option [2, 17]. The European guideline does not recommend faecal occult blood testing (faecal immunochemical testing) as sole surveillance and considers adenomas, serrated lesions, juvenile polyps with dysplasia and mixed lesions clinically relevant and requiring complete removal [2]. The sensitivity of multicancer detection using circulating tumour DNA for stage I tumours (16.8%) is insufficient to replace conventional surveillance [17].

7.3. Risk-reducing surgery

Mastectomy. Bilateral risk-reducing mastectomy should be discussed with a high-risk breast team as an alternative to surveillance, considering family history, age (generally after age 25), life expectancy, residual risk, reconstruction options and psychosocial aspects; it may be preferred when surveillance is difficult to access, breast tissue is dense or repeated biopsies have been required [1, 17]. In high-risk women, prophylactic mastectomy reduces breast cancer risk by approximately 90% [1, 17]. The risk of a second breast cancer (approximately 46% at 10 years after a first cancer in the European cohort) reinforces this discussion at the time of the first diagnosis [2, 50].

Thyroidectomy. Routine risk-reducing thyroidectomy is not recommended; the decision is individual and shared with thyroid surgery, endocrinology and genetics [17]. When cytology is suspicious for malignancy, total thyroidectomy is indicated because subtotal thyroidectomy leaves tissue that regrows with a similar cancer risk [1].

Hysterectomy. Routine risk-reducing hysterectomy is not recommended, but it should be discussed individually after completion of childbearing, particularly in the presence of multiple uterine polyps, atypical hyperplasia, multiple leiomyomas or a marked family history of endometrial cancer [1, 2, 17]. Multiple leiomyomas are treated with total hysterectomy because myomectomy alone is accompanied by frequent recurrence [1]. Because ovarian cancer risk does not appear to exceed that of the general population, ovarian-preserving approaches may be considered; in atypical hyperplasia, options include hysteroscopy with curettage and a progestogen-releasing intrauterine device [17].

Colectomy. Risk-reducing colectomy is not recommended in PHTS with polyposis; management is endoscopic [17].

7.4. Chemoprevention

There is no direct evidence of benefit from tamoxifen or raloxifene in PHTS; if risk-reducing drugs (tamoxifen, raloxifene, exemestane or anastrozole) are considered, the absence of evidence and the increase in endometrial cancer risk associated with tamoxifen in an already at-risk population should be discussed, generally arguing against this drug in women with a uterus; in many European countries, this use is not included in the guidelines [1, 17]. Cyclo-oxygenase inhibitors (for example, acetylsalicylic acid) and mTOR inhibitors are not recommended for chemoprevention of polyposis or colorectal cancer [17]. There is no evidence that lifestyle factors modify risk differently from the general population, but avoidance of known carcinogens (tobacco, obesity, alcohol) and excessive sun exposure is recommended [2, 17].

7.5. Treatment of cancer and other manifestations

Cancer. PHTS-associated cancers should be treated according to current guidelines for the corresponding sporadic tumour; there is no direct evidence of increased second malignancies induced by chemotherapy or radiotherapy, and this concern should not prevent standard treatment [4, 17]. Nevertheless, caution is recommended with diagnostic and therapeutic radiation exposure, weighing the benefit against the theoretical risk of a second malignancy (typically 10 to 20 years later); MRI does not carry this risk [1, 18]. In advanced hormone receptor-positive, HER2-negative breast cancer, PTEN alterations (including germline variants detected in the tumour) confer eligibility for capivasertib (Truqap®), an AKT inhibitor, in combination with fulvestrant after progression on endocrine therapy; FDA approval (16 November 2023) and European Commission authorisation (17 June 2024, following the CHMP opinion of 25 April 2024) require a PIK3CA, AKT1 or PTEN alteration identified by a validated test; since June 2026, the US label has included treatment-naïve metastatic or androgen pathway modulation-sensitive prostate cancer with PTEN deficiency detected by tumour immunohistochemistry, in combination with abiraterone and prednisone, an indication absent from the European authorisation as of 4 October 2026 [23, 24, 53, 59]. In a woman with PHTS, a germline PTEN variant detected in tumour tissue meets the breast cancer label's PTEN alteration criterion, so use is on-label when the remaining label criteria are met. The recommended dose is 400 mg orally twice daily for 4 days, followed by a 3-day break; the most frequent adverse effects are diarrhoea, skin reactions, hyperglycaemia (including severe diabetic ketoacidosis), nausea, fatigue and stomatitis, requiring blood glucose and glycated haemoglobin assessment before and during treatment [23]. The CAPItello-291 trial showed a progression-free survival benefit without a demonstrated overall survival benefit [53]; detailed results and the limitations of their interpretation are in section 7.7. Two cases of breast cancer in patients with PHTS had a durable response to capivasertib [17].

7.6. Targeted therapies under investigation

No drug is approved for PHTS. GeneReviews and the international guidelines limit mTOR inhibitor use to clinical trials, except in selected clinical situations determined by the treating physician [1, 17, 18].

Sirolimus (Rapamune®). An mTOR inhibitor approved by the FDA and EMA for prophylaxis of renal transplant rejection and for lymphangioleiomyomatosis; its use in PHTS is off-label [61, 62]. In the National Cancer Institute phase II pilot study (NCT00971789), 18 adults with PHTS received a 6 mg oral loading dose followed by 2 mg daily in 28-day cycles for 56 days (extendable in patients with cancer); treatment was well tolerated and associated with regression of skin and gastrointestinal lesions, improved cerebellar function after one month and reduced S6 phosphorylation in tissues [63, 64]. In a one-year open-label trial in adults with colonic polyposis (NCT04094675), only 2 of the 5 participants completed treatment, both with reduced polyp burden, and adverse effects limited use [65, 66]. In a retrospective series of seven children with vascular anomalies, treatment started at a median age of 10 years, and six (86%) had significant clinical improvement after 2.5 years, with mild mouth ulcers and acne as the only adverse effects [60]; compassionate use reduced vascular overgrowth and pain in two patients with malformations due to somatic loss of the second allele [9]. In juvenile polyposis of infancy with a PTEN variant, mTOR inhibitors reduced the need for colectomy and improved protein-losing enteropathy and chronic bleeding [17]. The label includes a boxed warning on immunosuppression, infections and lymphoma, and adverse effects such as hypersensitivity, angioedema, fluid retention and delayed wound healing, hypertriglyceridaemia and hypercholesterolaemia, proteinuria and declining renal function [61]. The international guideline advises against sirolimus for colorectal cancer chemoprevention or polyp management outside juvenile polyposis of infancy [17].

Everolimus (Afinitor®, Votubia®). Approved for tumours associated with tuberous sclerosis complex, among other indications; use in PHTS is off-label [67, 68]. In the randomised, double-blind, placebo-controlled phase II trial (NCT02991807), 46 patients aged 5 to 45 received everolimus 4.5 mg/m² for 6 months; the primary neurocognitive endpoint did not improve, gastrointestinal adverse effects were more frequent in the active group, and several secondary measures (non-verbal IQ, verbal learning, autistic symptoms, motor function, adaptive behaviour) moved towards improvement [69]. A 2026 sensitivity analysis confirmed the absence of an effect on the composite endpoint and suggested benefit in fine motor skills and social functioning [70]. A new phase II/III everolimus trial for social skills is planned (NCT07218575; 60 participants aged 5 to 45; primary endpoint: change in the total Social Responsiveness Scale score at 6 months; estimated start in September 2026), which was not yet recruiting at the last registry update (4 June 2026); an earlier sirolimus trial for the same endpoint (NCT06080165) was withdrawn before starting [71, 72].

Alpelisib (Vijoice®). A PI3Kα inhibitor approved by the FDA on 5 April 2022 (accelerated approval) and authorised by the European Commission on 15 July 2026 for severe manifestations of PIK3CA-related overgrowth spectrum (PROS) in patients aged 2 years and older, not for PHTS [73, 74]. In PHTS, data are preclinical only: in PTEN-haploinsufficient lipoma cells from three patients, alpelisib reduced proliferation, induced senescence, decreased AKT, mTOR and S6 phosphorylation, and reversed the sirolimus-induced increase in phosphorylated AKT [22]. No alpelisib clinical trials in PHTS were identified in the ClinicalTrials.gov registry. For reference in the differential diagnosis with PROS, the labelled dose is 250 mg daily in adults and 50 mg daily from ages 2 to 18 (which may be increased to 125 mg from age 6 after 24 weeks), with diarrhoea, stomatitis and hyperglycaemia as the most frequent adverse effects and a risk of severe hypersensitivity, severe skin reactions and pneumonitis [73].

AKT inhibitors. A phase II trial of the allosteric AKT inhibitor TAS-117 in advanced solid tumours with germline inactivating PTEN variants was terminated because of insufficient recruitment [17, 75]. Capivasertib blocked vascular malformation growth in the mouse model [9]. A 2026 case report described a 34-year-old woman with PHTS and stage IV triple-negative metaplastic breast carcinoma, progressing after chemotherapy and radiotherapy, with disease stabilisation maintained at 3.8 years on TAS-117 and a complete circulating cell-free DNA marker response; this is evidence from a single patient and cannot be generalised [76].

Off-label use and extrapolation. Off-label use is defined by the label wording, age and jurisdiction and should be distinguished from evidence extrapolation. Oral sirolimus is approved in the US from age 13 and in the EU in adults for renal transplantation and lymphangioleiomyomatosis, so its use in children with PHTS-associated vascular anomalies is off-label in terms of both disease and age; the rationale extrapolates the common PI3K-AKT-mTOR pathway mechanism and, in PHTS, rests on a retrospective series of seven children [60–62]. Everolimus has paediatric indications only in tuberous sclerosis complex (TSC); in PHTS, its use is off-label, and extrapolation from TSC was not confirmed by the randomised neurocognitive trial [67–69]. Alpelisib is approved for PROS, a PIK3CA mosaic disorder distinct from germline PHTS; its use in PHTS would be off-label and without clinical data [22, 73, 74]. During pregnancy, use of these drugs for non-oncological PHTS manifestations is not supported by labels or studies.

7.7. Regulatory status, trials and hierarchy of evidence

Evidence on PHTS treatment has distinct levels that should not be confused. Surveillance recommendations arise from expert consensus (2025 international and 2026 European guidelines) supported by observational cohorts; pharmacological efficacy data in PHTS come from a single randomised trial (everolimus, negative for the primary endpoint), uncontrolled pilot studies and case reports; trial registry information (ClinicalTrials.gov) is sponsor-reported and not peer-reviewed; and regulatory actions (FDA approval or a European Commission decision following the opinion of the Committee for Medicinal Products for Human Use, CHMP) concern indications other than PHTS, except when the oncological indication is defined by the tumour PTEN alteration [2, 17, 18, 69]. A positive CHMP opinion is not yet an authorisation, and an authorisation does not guarantee marketing, reimbursement or access in each country: in Portugal, these depend on marketing and the INFARMED funding assessment, and in Brazil on ANVISA registration and incorporation into the Unified Health System following CONITEC assessment.

Table 7.1. Regulatory status of the drugs cited in PHTS (4 October 2026)

Table 7.2. Trials and intervention studies relevant to PHTS

These studies do not permit drug ranking by indirect comparison. The endpoints differ (safety, polyp burden, neurocognitive composite, progression-free survival), as do the populations (PHTS carriers without cancer on the one hand, and oncology patients selected by tumour alterations, among whom germline carriers are an unquantified minority, on the other), and PHTS studies lack a control group except for the everolimus trial, which was negative for the primary endpoint; improvements in that trial's secondary endpoints generate hypotheses but do not confirm them. In oncology trials, median progression-free survival describes the median patient in each arm rather than the expected gain for an individual patient, and the hazard ratio is a relative measure: in CAPItello-291, the median of 7.2 months means that half of the patients treated with capivasertib had progressed or died by that time, and no overall survival benefit was demonstrated [53, 79]. Trials terminated because of insufficient recruitment (TAS-117) or withdrawn before starting (sirolimus for social skills) should be interpreted as an absence of results rather than evidence of inefficacy.

8. Reproductive Options and Family Counselling

8.1. Transmission and risk to relatives

PHTS has autosomal dominant inheritance, and each child of a carrier has a 50% probability of inheriting the variant [1, 4]. Approximately 10–44% of diagnosed individuals have a de novo variant, and almost all patients with PTEN-related Proteus-like syndrome are de novo cases [1]. In the Cleveland Clinic experience, approximately half of patients have an affected relative; the Oxford Desk Reference estimates that 10–50% have an affected parent [1, 4]. Because manifestations may be subtle, a family diagnosis is often made only after identification of a more severely affected member, and an apparently isolated case may result from failure to recognise the disease in the parents, early parental death or parental mosaicism [1]. Molecular testing of both parents of a simplex case is therefore recommended; if the variant is not detected in either parent's leucocytes, the recurrence risk for siblings is estimated at approximately 1% because of gonadal mosaicism, and a parent with somatic and gonadal mosaicism may have minimal manifestations [1]. Penetrance approaches 99% in adulthood, with no known non-penetrant carriers [1]. Relatives who did not inherit the variant, and their descendants, have cancer risks similar to those of the general population [1].

8.2. Cascade testing and age for predictive testing

Testing for the familial variant is recommended for all at-risk relatives [1]. There is no consensus on the ideal age, but testing in early childhood is often appropriate because it allows early detection of thyroid disease (the earliest PTEN-related thyroid cancer was diagnosed at age 4 and breast cancer at age 17) and guides early intervention for neurodevelopmental disorders [1]. This position differs from practice in exclusively adult-onset syndromes, in which predictive testing of minors is usually deferred. Predictive testing requires prior identification of the pathogenic variant in an affected relative, and a variant of uncertain significance should not be used for this purpose [1].

8.3. Reproductive options

Discussion of genetic risk and reproductive options should ideally take place before pregnancy and be offered to affected or at-risk young adults [1]. Once the familial variant has been identified, prenatal diagnosis (chorionic villus sampling or amniocentesis) and preimplantation genetic testing in the context of in vitro fertilisation are possible [1]. Other options include accepting the risk with targeted paediatric follow-up from birth, gamete donation and adoption. Professionals and families have differing perspectives on prenatal diagnosis in a disease with highly variable expressivity and predominantly adult-onset cancer risks, so the decision is personal and counselling non-directive [1]. Risk-reducing hysterectomy should be discussed after completion of childbearing, considering age, the desire for future fertility and overall health [1, 17].

8.4. Psychosocial aspects

The recommended surveillance requires substantial patient commitment and a coordinated multidisciplinary approach [2]. Neurocognitive deficits, ASD (approximately 25% of patients meet criteria) and psychiatric disorders should be considered when planning surveillance and consent, and psychosocial and social work support should be offered [1, 2, 17]. Patient groups (PTEN Foundation, PTEN Research, PTEN Italia) and the PTEN patient registry support education on warning signs and research [1]. The international guideline includes specific recommendations for transgender people: individuals assigned female at birth with remaining breast tissue should continue breast surveillance, and in those assigned male at birth receiving gender-affirming hormone therapy, breast surveillance should begin 5 years after treatment starts and not before age 25–30 [17].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence

The classical estimate of 1 in 200,000 to 250,000 came from clinical series and is considered an underestimate [2, 4]. In a cross-sectional study identifying ClinVar pathogenic or likely pathogenic variants in 245,394 participants in the US All of Us programme and 469,558 UK Biobank participants, PHTS prevalence was 1 in 8,764 and 1 in 13,043, respectively — approximately 10 to 20 times higher than previous estimates — and most carriers did not have thyroid cancer [10]. The European guideline adopts a prevalence between 1 in 9,000 and 1 in 15,000 [2]. The Dutch reference centre estimated a PHTS prevalence of 1 in 20,000 in the Netherlands, a figure referring to diagnosed disease in that country rather than the carrier frequency measured in biobanks [44]. These figures derive from volunteer cohorts with limited representation of ancestries and ages and capture mildly affected carriers missed by clinical series.

9.2. Population diversity and founder effect

No PTEN founder variants of population relevance have been described; most variants are private to individual families, with some recurrent variants at CpG codons (p.Arg130Ter, p.Arg233Ter, p.Arg335Ter) observed in diverse populations [1]. The prospective Cleveland Clinic cohort included patients from North America, South America, Europe, Australia and Asia but does not provide ancestry-stratified estimates [11]. In Japanese individuals, C2 domain variants were associated with increased breast cancer risk [17]. The international guideline identifies as research priorities determining whether cancer risks vary by ancestry and assessing outcome disparities associated with genetic factors, comorbidities and social determinants [17]. Cancer risks are higher in women: in the European cohort, the risk of PHTS-related cancer by age 60 was 68.4–86.3% in women and 16.4–20.8% in men, and thyroid cancer risk is also higher in females [2, 12].

9.3. Portuguese-speaking and regional data

We identified no population-based studies of prevalence or cancer risk in Portugal, Brazil or other Portuguese-speaking countries, nor stratified Latin American data. Isolated reports exist: a family of Brazilian origin included in a Spanish series of 14 families with Cowden syndrome, nine (64%) of which had a PTEN variant [80]; a 56-year-old Brazilian woman with “cobblestone” gingival papillomatosis and histologically confirmed cutaneous trichilemmomas [81]; and a Portuguese child with ASD, progressive macrocephaly and a novel PTEN frameshift variant [82]. In Brazil, published series come mainly from oral pathology services: a series from Piracicaba (State University of Campinas) described 10 patients in whom oral papillomatosis was a constant finding and decisive for diagnosis [83]; in São Paulo, histological examination of oral lesions in nine clinically diagnosed patients showed features suggestive of viral warts at different stages, an interpretation discussed by the authors that does not alter the diagnostic criteria [84]; and a study of an isolated case and a family identified a novel nonsense variant, c.940G>T, and an intronic variant, c.1026+32T>G, whose classification is discussed in section 5.7 [55]. These hospital series describe phenotypes and do not measure prevalence. In the absence of regional data, European and North American estimates are applied, acknowledging their ascertainment bias. Access to breast MRI, ultrasonography, colonoscopy and genetic testing determines implementation of the guidelines in resource-limited health systems; the international guideline allows local adaptation of recommendations, and the European guideline prioritises cost-effectiveness and universal access [2, 17]. In settings without access to testing, the Cleveland Clinic calculator has been used to estimate variant probability and support counselling and surveillance [52].

9.4. Special populations

Children with ASD or intellectual disability constitute the earliest-diagnosed group and have the greatest risk of losing cancer follow-up during the transition to adulthood; their lifetime cancer risk exceeds that of carriers without a neurodevelopmental disorder [11]. Patients identified through secondary findings in exome or genome sequencing — PTEN is included in the ACMG secondary findings list — frequently lack a suggestive personal history and benefit from the same surveillance programme [10, 21]. Patients with thyroid, breast or endometrial cancer without a previous PHTS diagnosis are another target population because of the risk of a second cancer [50].

10. Illustrative Clinical Case

10.1. Published case: mother and son with oral papillomatosis and a PTEN exon 6 deletion

Sutera and colleagues, from an Italian oral medicine and genetics centre, described in an open-access article a 25-year-old man referred to exclude acanthosis nigricans because of asymptomatic verrucous lesions at multiple sites of the oral mucosa and lip vermilion [51]. Examination showed small whitish and pinkish lesions on the gums, vermilion, hard palate and buccal mucosa, and “cobblestone” papillomatous lesions along the lateral borders of a fissured tongue; HC was 61.5 cm (above the 99th centile; +4.46 SD) (Figure 1 of Sutera et al. 2022) [51]. With two major criteria (multiple mucocutaneous lesions and macrocephaly), he met testing criteria, although not the clinical diagnostic criteria [51]. His 55-year-old mother, who accompanied him, had previously disregarded tongue lesions; examination revealed multiple “cobblestone” oral papillomas, oral melanosis, cutaneous keratotic pits and acral hyperkeratotic papules, and an HC of 61 cm (+6.04 SD) (Figure 2 of Sutera et al. 2022) [51]. Her history included a uterine malignancy of an unspecified nature, cerebral ischaemia, benign thyroid nodules treated with partial thyroidectomy, a hepatic angioma and four lipomas; even without counting the uterine tumour, she met two major and three minor criteria (lipomas, structural thyroid lesions and vascular anomalies), and her son then met the familial criteria [51].

PTEN sequencing in the son, including the promoter, identified no pathogenic variant, only a rare intronic substitution inherited from his mother without a predicted splicing effect; MLPA (multiplex ligation-dependent probe amplification) analysis revealed in both a heterozygous deletion of the entire exon 6, c.(492+1_493-1)_(634+1_635-1)del, with a predicted effect of p.(Gly165Ilefs*9), demonstrated in the MLPA analysis reproduced in the article (Figure 3 of Sutera et al. 2022) [51]. Both entered cancer surveillance programmes, and the authors emphasised that diffuse gingival hyperplasia, without proportionate gingivitis and involving several mucosal subregions, may reveal the syndrome; oral lesions warrant excision only when they compromise oral hygiene or appearance [51].

Teaching points. The case illustrates four principles: oral lesions and macrocephaly are early signs accessible to any clinician; examination of accompanying relatives may reveal the index case; negative sequencing does not exclude PHTS, and deletion/duplication analysis is mandatory (accounting for 3–11% of variants); and surgery for mucocutaneous hamartomas should be conservative [1, 51].

10.2. Composite case: Lhermitte-Duclos disease as the route to diagnosis

The following is a composite case, constructed for teaching from the features described in the cited sources; it does not correspond to a published patient, and the management described applies the guidelines without replacing them. A 34-year-old woman is seen in neurology with progressive occipital headache, gait imbalance and morning vomiting of three months' duration. Her history includes total thyroidectomy at age 29 for follicular thyroid carcinoma and myomectomy at age 31. Brain MRI shows a right cerebellar hemisphere lesion, hypointense on T1 and hyperintense on T2, with a “tiger-striped” pattern, without significant enhancement or diffusion restriction, with fourth ventricle compression and mild ventricular dilatation — a typical appearance of Lhermitte-Duclos disease with cerebrospinal fluid obstruction [1, 18]. Examination reveals an HC of 59 cm, multiple centrofacial and periorificial facial papules, gingival papillomas and palmar keratotic pits. Her mother had breast cancer at age 41, and a maternal uncle underwent nephrectomy for renal cell carcinoma at age 50.

Because of CSF obstruction, the patient is referred for neurosurgical assessment and undergoes resection, with histological confirmation of dysplastic gangliocytoma of the cerebellum [1, 18]. Even before applying any clinical criterion, adult LDD is itself an indication for PTEN testing and counselling as PHTS [32]. Clinically, she meets major criteria (adult LDD, follicular thyroid cancer, macrocephaly and multiple mucocutaneous lesions, to be confirmed by trichilemmoma biopsy) [14]. Sequencing identifies the heterozygous variant c.388C>T p.(Arg130Ter), one of the recurrent variants described in GeneReviews [1].

The follow-up plan includes: brain MRI every 6–12 months after resection [1, 18]; immediate annual breast MRI, with mammography from age 40 (or immediately, according to the North American guideline) and discussion of risk-reducing mastectomy in a high-risk breast clinic [1, 2, 17]; thyroid bed surveillance as determined by endocrinology; education on abnormal uterine bleeding and individualised discussion of endometrial surveillance and hysterectomy after completion of childbearing [2, 17]; baseline colonoscopy at age 35 [1, 2]; renal imaging (baseline MRI at age 35–40 in the European guideline, or periodic ultrasonography according to the North American and international guidelines), brought forward because of the family history [1, 2, 17]; and dermatological assessment [1, 2]. Cascade testing is offered to the patient's mother, siblings and two children, aged 6 and 3; if positive, the children will begin developmental surveillance and thyroid ultrasonography from age 12 (or age 10–18 according to the European guideline) [1, 2, 17].

11. Prognosis and Long-Term Follow-Up

Prognosis is determined mainly by cancers and, in children, by neurodevelopmental disorders and vascular malformations [1]. In the European adult cohort, 10-year overall survival after breast cancer was 77%, ranging from 90% at stage 0 to 0% at stage IV; relative survival was comparable to that of sporadic breast cancer in the first two years and higher thereafter, and 10-year overall survival was 87% for thyroid cancer and 64% for endometrial cancer [85]. The prognosis of PHTS-associated cancers is therefore comparable to that of the general population, but overall and cancer mortality in women with PHTS exceed population rates (standardised mortality ratios of 3.7 and 2.7), attributed to the higher cancer incidence [85]. The high risk of second primary cancers — 42% of patients with cancer in the North American cohort and a 45.7% 10-year risk in European women — justifies continuing surveillance after the first cancer [11, 50].

Lifelong follow-up follows the guidelines in section 7, adjusted to age, sex, family history and patient preferences, ideally coordinated by an experienced centre (for example, ERN GENTURIS members) with prospective data registration [2]. After age 75–80, breast surveillance should be individualised [17]. In postmenopausal women, the need to continue breast MRI should be reassessed [2]. In childhood and adolescence, follow-up includes annual monitoring of development and educational needs, ASD screening, annual assessment of scoliosis and asymmetry, and thyroid ultrasonography from age 12 [1]. Transition from paediatric to adult care is a critical point, particularly for young people with intellectual disability, whose cancer surveillance should be planned with caregivers [11, 18]. Current knowledge gaps include penetrance after age 60–70, the effectiveness of renal, colorectal, skin and endometrial surveillance, the effect of risk-reducing breast surgery, and response to systemic cancer treatments, which research in international registries seeks to address [2, 17].

12. Summary and Concluding Remarks

PTEN hamartoma tumour syndrome is a single molecular entity with autosomal dominant inheritance, uniting historically distinct phenotypes — Cowden syndrome, Bannayan-Riley-Ruvalcaba syndrome, Proteus-like syndrome, adult Lhermitte-Duclos disease and ASD with macrocephaly — under the same mechanism: PTEN loss of function and unopposed activation of the PI3K-AKT-mTOR pathway, frequently completed by a second somatic event in affected tissues. Prevalence, estimated at 1 in 9,000 to 1 in 13,000 in genotype-first studies, is much higher than was accepted until a few years ago, indicating substantial underdiagnosis.

Clinical recognition rests on macrocephaly, mucocutaneous lesions (trichilemmomas, oral papillomas, acral keratoses), polyposis of mixed histology, benign thyroid and breast disease and, in children, developmental delay, ASD, lipomas, vascular malformations and penile macules. The Pilarski/NCCN criteria and Cleveland Clinic score guide selection for testing, but diagnosis is molecular and requires assessment of sequence variants, deletions/duplications and the promoter.

The dominant cancer risks are breast cancer in women (60–91% depending on the cohort and method), endometrial and thyroid cancer (predominantly follicular), with moderate risks of renal cancer (mainly papillary), colorectal cancer and melanoma, and increased risks of prostate cancer and sarcomas; second primary cancers are frequent. The 2025–2026 guidelines converge on annual breast MRI from age 25–30, discussion of risk-reducing mastectomy, thyroid ultrasonography from the second decade with intervals adapted to findings, and baseline colonoscopy at age 35–40, while differing on endometrial and renal surveillance. Adult Lhermitte-Duclos disease is itself an indication for PTEN testing and is followed with MRI, reserving surgery for atypical or obstructive forms.

No targeted therapy is approved: sirolimus showed regression of skin, gastrointestinal and vascular lesions in small series, everolimus did not improve the primary neurocognitive endpoint, and alpelisib has only preclinical data in PHTS, so these drugs should be used in clinical trials or through individualised decisions. Capivasertib, approved for advanced breast cancer with PTEN alterations, is relevant to patients with PHTS and metastatic breast cancer. Cascade testing, frequently offered in early childhood, and lifelong multidisciplinary coordination are the interventions with the greatest prognostic impact.

13. References

1. Tramontana T, Yehia L. PTEN hamartoma tumor syndrome. 2001 Nov 29 [updated 2026 Jul 28]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. PMID: 20301661. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1488/
2. 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.
3. 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.
4. 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.
5. Online Mendelian Inheritance in Man, OMIM®. Cowden syndrome 1 (#158350); Bannayan-Riley-Ruvalcaba syndrome (#153480); Macrocephaly/autism syndrome (#605309); PTEN (*601728). Baltimore (MD): Johns Hopkins University; [accessed 2026 Oct 4]. Available at: https://omim.org/entry/158350
6. Orphanet. PTEN hamartoma tumor syndrome (ORPHA:306498); Cowden syndrome (ORPHA:201); Bannayan-Riley-Ruvalcaba syndrome (ORPHA:109); Proteus-like syndrome (ORPHA:2969); Lhermitte-Duclos disease (ORPHA:65285); SOLAMEN (ORPHA:137608); PTEN hamartoma of soft tissue (ORPHA:726807); Macrocephaly-intellectual disability-autism syndrome (ORPHA:210548). Paris: INSERM; [accessed 2026 Oct 4]. Available at: https://www.orpha.net/en/disease/detail/306498
7. National Center for Biotechnology Information. PTEN (Gene ID 5728): MANE Select transcript NM_000314.8 [Internet]. Bethesda (MD): NCBI Datasets, Gene product report; [accessed 2026 Oct 4]. Available at: https://www.ncbi.nlm.nih.gov/datasets/gene/5728/
8. 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.
9. Castillo SD, Perosanz X, Ressler AK, Ivars M, Rodríguez J, Rovira C, et al. Somatic uniparental disomy of PTEN in endothelial cells causes vascular malformations in patients with PTEN hamartoma tumor syndrome. Cancer Discov. 2025;15(7):1350-62. doi:10.1158/2159-8290.CD-24-0807. PMID: 40152704.
10. 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):e4049-54. doi:10.1210/clinem/dgaf236. PMID: 40231587.
11. Yehia L, Plitt G, Tushar AM, Joo J, Burke CA, Campbell SC, et al. Longitudinal analysis of cancer risk in children and adults with germline PTEN variants. JAMA Netw Open. 2023;6(4):e239705. doi:10.1001/jamanetworkopen.2023.9705. PMID: 37093598.
12. Hendricks LAJ, Hoogerbrugge N, Mensenkamp AR, Brunet J, Lleuger-Pujol R, Høberg-Vetti H, et al. Cancer risks by sex and variant type in PTEN hamartoma tumor syndrome. J Natl Cancer Inst. 2023;115(1):93-103. doi:10.1093/jnci/djac188. PMID: 36171661.
13. Eng C. Will the real Cowden syndrome please stand up: revised diagnostic criteria. J Med Genet. 2000;37(11):828-30. doi:10.1136/jmg.37.11.828. PMID: 11073535.
14. Pilarski R, Burt R, Kohlman W, Pho L, Shannon KM, Swisher E. Cowden syndrome and the PTEN hamartoma tumor syndrome: systematic review and revised diagnostic criteria. J Natl Cancer Inst. 2013;105(21):1607-16. doi:10.1093/jnci/djt277. PMID: 24136893.
15. Tan MH, Mester J, Peterson C, Yang Y, Chen JL, Rybicki LA, et al. A clinical scoring system for selection of patients for PTEN mutation testing is proposed on the basis of a prospective study of 3042 probands. Am J Hum Genet. 2011;88(1):42-56. doi:10.1016/j.ajhg.2010.11.013. PMID: 21194675.
16. Tischkowitz M, Colas C, Pouwels S, Hoogerbrugge N; PHTS Guideline Development Group; European Reference Network GENTURIS. Cancer surveillance guideline for individuals with PTEN hamartoma tumour syndrome. Eur J Hum Genet. 2020;28(10):1387-93. doi:10.1038/s41431-020-0651-7. PMID: 32533092.
17. 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):1754-65. doi:10.1158/1078-0432.CCR-24-3819. PMID: 39937242.
18. Dhawan A, Liu D, Baitamouni S, Anthony K, Srivastava S, Hardan AY, et al. Neurodevelopmental and neurologic manifestations of PTEN hamartoma tumor syndrome: management recommendations. Neurol Genet. 2025;11(5):e200299. doi:10.1212/NXG.0000000000200299. PMID: 41080814.
19. 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):234-44. doi:10.1158/1078-0432.CCR-24-1947. PMID: 39540884.
20. ClinGen Actionability Working Group. PTEN hamartoma tumor syndrome — Cowden syndrome (report AC025), adult and paediatric contexts. Clinical Genome Resource; [accessed 2026 Oct 4]. Available at: https://actionability.clinicalgenome.org/ac/
21. Miller DT, Lee K, Abul-Husn NS, Amendola LM, Brothers K, Chung WK, et al. ACMG SF v3.2 list for reporting of secondary findings in clinical exome and genome sequencing: a policy statement of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2023;25(8):100866. doi:10.1016/j.gim.2023.100866. PMID: 37347242.
22. Kirstein AS, Augustin A, Penke M, Cea M, Körner A, Kiess W, et al. The novel phosphatidylinositol-3-kinase (PI3K) inhibitor alpelisib effectively inhibits growth of PTEN-haploinsufficient lipoma cells. Cancers (Basel). 2019;11(10):1586. doi:10.3390/cancers11101586. PMID: 31627436.
23. AstraZeneca Pharmaceuticals LP. TRUQAP (capivasertib) tablets: prescribing information. Revised 2026 Sep. DailyMed setid d698c106-2322-401e-b738-cbd83c843ecf. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d698c106-2322-401e-b738-cbd83c843ecf
24. European Medicines Agency. Truqap (capivasertib): EPAR — medicine overview. Marketing authorisation dated 17 June 2024. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/truqap
25. European Commission. European Reference Networks: ERN GENTURIS, VASCERN and ERN ITHACA [Internet]. Brussels: European Commission; [accessed 2026 Oct 4]. Available at: https://health.ec.europa.eu/rare-diseases-and-european-reference-networks/european-reference-networks_en
26. Lloyd KM, Dennis M. Cowden's disease. A possible new symptom complex with multiple system involvement. Ann Intern Med. 1963;58:136-42. doi:10.7326/0003-4819-58-1-136. PMID: 13931122.
27. Nelen MR, Padberg GW, Peeters EA, Lin AY, van den Helm B, Frants RR, et al. Localization of the gene for Cowden disease to chromosome 10q22-23. Nat Genet. 1996;13(1):114-6. doi:10.1038/ng0596-114. PMID: 8673088.
28. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, et al. PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science. 1997;275(5308):1943-7. doi:10.1126/science.275.5308.1943. PMID: 9072974.
29. Marsh DJ, Dahia PL, Zheng Z, Liaw D, Parsons R, Gorlin RJ, et al. Germline mutations in PTEN are present in Bannayan-Zonana syndrome. Nat Genet. 1997;16(4):333-4. doi:10.1038/ng0897-333. PMID: 9241266.
30. Marsh DJ, Kum JB, Lunetta KL, Bennett MJ, Gorlin RJ, Ahmed SF, et al. PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome. Hum Mol Genet. 1999;8(8):1461-72. doi:10.1093/hmg/8.8.1461. PMID: 10400993.
31. Zhou XP, Marsh DJ, Hampel H, Mulliken JB, Gimm O, Eng C. Germline and germline mosaic PTEN mutations associated with a Proteus-like syndrome of hemihypertrophy, lower limb asymmetry, arteriovenous malformations and lipomatosis. Hum Mol Genet. 2000;9(5):765-8. doi:10.1093/hmg/9.5.765. PMID: 10749983.
32. Zhou XP, Marsh DJ, Morrison CD, Chaudhury AR, Maxwell M, Reifenberger G, et al. Germline inactivation of PTEN and dysregulation of the phosphoinositol-3-kinase/Akt pathway cause human Lhermitte-Duclos disease in adults. Am J Hum Genet. 2003;73(5):1191-8. doi:10.1086/379382. PMID: 14566704.
33. Butler MG, Dasouki MJ, Zhou XP, Talebizadeh Z, Brown M, Takahashi TN, et al. Subset of individuals with autism spectrum disorders and extreme macrocephaly associated with germline PTEN tumour suppressor gene mutations. J Med Genet. 2005;42(4):318-21. doi:10.1136/jmg.2004.024646. PMID: 15805158.
34. Dragoo DD, Taher A, Wong VK, Elsaiey A, Consul N, Mahmoud HS, et al. PTEN hamartoma tumor syndrome/Cowden syndrome: genomics, oncogenesis, and imaging review for associated lesions and malignancy. Cancers (Basel). 2021;13(13):3120. doi:10.3390/cancers13133120. PMID: 34206559.
35. Tan MH, Mester JL, Ngeow J, Rybicki LA, Orloff MS, Eng C. Lifetime cancer risks in individuals with germline PTEN mutations. Clin Cancer Res. 2012;18(2):400-7. doi:10.1158/1078-0432.CCR-11-2283. PMID: 22252256.
36. Firth HV, Hurst JA. Overgrowth. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 272-5.
37. Firth HV, Hurst JA. Juvenile polyposis syndrome (JPS). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 578-80.
38. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-24. doi:10.1038/gim.2015.30. PMID: 25741868.
39. Mester JL, Ghosh R, Pesaran T, Huether R, Karam R, Hruska KS, et al. Gene-specific criteria for PTEN variant curation: recommendations from the ClinGen PTEN Expert Panel. Hum Mutat. 2018;39(11):1581-92. doi:10.1002/humu.23636. PMID: 30311380.
40. Jones KL, Jones MC, del Campo M. Proteus syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 737-8.
41. Caux F, Plauchu H, Chibon F, Faivre L, Fain O, Vabres P, et al. Segmental overgrowth, lipomatosis, arteriovenous malformation and epidermal nevus (SOLAMEN) syndrome is related to mosaic PTEN nullizygosity. Eur J Hum Genet. 2007;15(7):767-73. doi:10.1038/sj.ejhg.5201823. PMID: 17392703.
42. Yehia L, Eng C. PTEN hamartoma tumour syndrome: what happens when there is no PTEN germline mutation? Hum Mol Genet. 2020;29(R2):R150-7. doi:10.1093/hmg/ddaa127. PMID: 32568377.
43. Firth HV, Hurst JA. Macrocephaly. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 224-6.
44. Jonker R, Schieving J. Epilepsy in pediatric patients with PTEN hamartoma tumor syndrome: first step in recommendations for clinical management. Eur J Paediatr Neurol. 2026;61:36-41. doi:10.1016/j.ejpn.2026.03.004. PMID: 41825102.
45. Bubien V, Bonnet F, Brouste V, Hoppe S, Barouk-Simonet E, David A, et al. High cumulative risks of cancer in patients with PTEN hamartoma tumour syndrome. J Med Genet. 2013;50(4):255-63. doi:10.1136/jmedgenet-2012-101339. PMID: 23335809.
46. Firth HV, Hurst JA. Breast cancer. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 546-7.
47. Mester JL, Zhou M, Prescott N, Eng C. Papillary renal cell carcinoma is associated with PTEN hamartoma tumor syndrome. Urology. 2012;79(5):1187.e1-7. doi:10.1016/j.urology.2011.12.025. PMID: 22381246.
48. Yehia L, Plitt G, Tushar AM, Liu D, Joo J, Ni Y, et al. Extended spectrum of cancers in PTEN hamartoma tumor syndrome. NPJ Precis Oncol. 2025;9(1):61. doi:10.1038/s41698-025-00847-3. PMID: 40050354.
49. Ngeow J, Stanuch K, Mester JL, Barnholtz-Sloan JS, Eng C. Second malignant neoplasms in patients with Cowden syndrome with underlying germline PTEN mutations. J Clin Oncol. 2014;32(17):1818-24. doi:10.1200/JCO.2013.53.6656. PMID: 24778394.
50. Hendricks LAJ, Verbeek KCJ, Schuurs-Hoeijmakers JHM, de Jong MM, Links TP, Brems H, et al. The risk of a second primary cancer in PTEN hamartoma tumor syndrome (PHTS). Genet Med. 2025;27(10):101467. doi:10.1016/j.gim.2025.101467. PMID: 40433764.
51. Sutera S, Giachino DF, Pelle A, Zuntini R, Pentenero M. Gingival overgrowths revealing PTEN hamartoma tumor syndrome: report of novel PTEN pathogenic variants. Biomedicines. 2022;11(1):81. doi:10.3390/biomedicines11010081. PMID: 36672590.
52. Al-Noman A, Mokbul MI, Hossain N, Rana MS, Hasan MM, Islam MS, et al. A young female of Cowden syndrome presenting with Lhermitte-Duclos disease: an illustrative case. Surg Neurol Int. 2023;14:296. doi:10.25259/SNI_325_2023. PMID: 37680909.
53. ESMO Handbook — Cancer Genomics. Breast cancer: genomics. Lugano: ESMO Press. p. 83-7.
54. Hall MJ, Kim DW, Namjoshi D, McSweeny M, Richardson ME, Marsh AP, et al. Paired DNA/RNA testing uncovers a deep intronic PTEN pathogenic variant associated with clinical Cowden syndrome: a case report. Front Oncol. 2025;15:1679432. doi:10.3389/fonc.2025.1679432. PMID: 41142619.
55. Machado RA, Paranaíba LMR, Martins L, Melo-Filho MR, de Souza TT, Picciani BLS. Variable expressivity and novel PTEN mutations in Cowden syndrome. Oral Surg Oral Med Oral Pathol Oral Radiol. 2019;127(1):55-61. doi:10.1016/j.oooo.2018.08.016. PMID: 30316882.
56. Firth HV, Hurst JA. Breast cancer: clinical approach. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 547-8.
57. Firth HV, Hurst JA. Multiple endocrine neoplasia (MEN): differential diagnosis and thyroid cancer. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 593.
58. Dhawan A, Baitamouni S, Liu D, Anthony K, Salamone FP, Neri I, et al. Dermatologic manifestations of PTEN hamartoma tumor syndrome: management recommendations from the PTEN Hamartoma Tumor Syndrome Dermatology Working Group. J Am Acad Dermatol. 2026. doi:10.1016/j.jaad.2026.07.095. PMID: 42532428.
59. European Medicines Agency. Medicines data — human medicines dataset (EPAR; updated on 2026-10-04) [Internet]. Amsterdam: EMA; [accessed 2026 Oct 4]. Available at: https://www.ema.europa.eu/en/medicines/download-medicine-data
60. Zabeida A, Brzezinski JJ, Wasserman JD, Cytrynbaum C, Weksberg R, Zwicker K, et al. Sirolimus for vascular anomalies associated with PTEN hamartoma tumor syndrome. Pediatr Blood Cancer. 2024;71(11):e31282. doi:10.1002/pbc.31282. PMID: 39166269.
61. U.S. Food and Drug Administration. Sirolimus tablets and oral solution: prescribing information (openFDA drug label, 2025). Available at: https://api.fda.gov/drug/label.json?search=openfda.generic_name:sirolimus
62. European Medicines Agency. Rapamune (sirolimus): EPAR — medicine overview. Marketing authorisation dated 13 March 2001. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/rapamune
63. Komiya T, Blumenthal GM, DeChowdhury R, Fioravanti S, Ballas MS, Morris J, et al. A pilot study of sirolimus in subjects with Cowden syndrome or other syndromes characterized by germline mutations in PTEN. Oncologist. 2019;24(12):1510-e1265. doi:10.1634/theoncologist.2019-0514. PMID: 31350329.
64. National Cancer Institute. Sirolimus to treat Cowden syndrome and other PTEN hamartomatous tumor syndromes (NCT00971789). ClinicalTrials.gov; [accessed 2026 Oct 4]. Available at: https://clinicaltrials.gov/study/NCT00971789
65. Stanich PP, Gofar K, Roberts ME, Hussan H. Sirolimus for colon polyposis in PTEN hamartoma tumor syndrome. Clin Transl Gastroenterol. 2025;16(8):e00871. doi:10.14309/ctg.0000000000000871. PMID: 40478255.
66. Ohio State University. Sirolimus for Cowden syndrome with colon polyposis (NCT04094675). ClinicalTrials.gov; [accessed 2026 Oct 4]. Available at: https://clinicaltrials.gov/study/NCT04094675
67. Novartis Pharmaceuticals Corporation. AFINITOR and AFINITOR DISPERZ (everolimus): prescribing information (openFDA drug label, revised 2026 Jun). Available at: https://api.fda.gov/drug/label.json?search=openfda.brand_name:AFINITOR
68. European Medicines Agency. Votubia (everolimus): EPAR — medicine overview. Marketing authorisation dated 2 September 2011. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/votubia
69. Srivastava S, Jo B, Zhang B, Frazier T, Gallagher AS, Peck F, et al. A randomized controlled trial of everolimus for neurocognitive symptoms in PTEN hamartoma tumor syndrome. Hum Mol Genet. 2022;31(20):3393-404. doi:10.1093/hmg/ddac111. PMID: 35594551.
70. Liu Y, Wang R, Srivastava S, Jo B, Frazier T, Filip-Dhima R, et al. Sensitivity analysis of the efficacy of everolimus for neurocognitive symptoms in PTEN hamartoma tumor syndrome. Adv Ther. 2026;43(2):666-87. doi:10.1007/s12325-025-03441-y. PMID: 41428178.
71. Boston Children's Hospital. A randomized, placebo controlled double-blind trial of everolimus for improving social abilities in PTEN germline mutations (NCT07218575). ClinicalTrials.gov; [accessed 2026 Oct 4]. Available at: https://clinicaltrials.gov/study/NCT07218575
72. Boston Children's Hospital. Sirolimus for improving social abilities in people with PTEN germline mutations (NCT06080165; withdrawn) [Internet]. ClinicalTrials.gov; [accessed 2026 Oct 4]. Available at: https://clinicaltrials.gov/study/NCT06080165
73. Novartis Pharmaceuticals Corporation. VIJOICE (alpelisib) tablets and oral granules: prescribing information (openFDA drug label, revised 2025 Nov). Available at: https://api.fda.gov/drug/label.json?search=openfda.brand_name:VIJOICE
74. European Medicines Agency. Vijoice (alpelisib): EPAR — medicine overview. Marketing authorisation dated 15 July 2026. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/vijoice
75. Rodón J, Funchain P, Laetsch TW, Arkenau HT, Hervieu A, Singer CF, et al. A phase II study of TAS-117 in patients with advanced solid tumors harboring germline PTEN-inactivating mutations. Future Oncol. 2022;18(30):3377-87. doi:10.2217/fon-2022-0305. PMID: 36039910.
76. Godina E, Fernández Carcaño M, Uzunparmak B, Pepper T, Shah P, Rodon J. Remarkable response of metaplastic breast cancer to a novel targeted AKT inhibitor TAS-117 in a Cowden syndrome patient: a case report. BJC Rep. 2026;4(1). doi:10.1038/s44276-026-00216-y. PMID: 41986640.
77. U.S. Food and Drug Administration. HYFTOR (sirolimus topical gel) — prescribing information and Drugs@FDA (NDA 213478; initial approval 2022-03-22) [Internet]. openFDA; [accessed 2026 Oct 4]. Available at: https://api.fda.gov/drug/label.json?search=openfda.brand_name:"HYFTOR"
78. Novartis Pharmaceuticals. Study assessing the efficacy, safety and PK of alpelisib (BYL719) in pediatric and adult patients with PIK3CA-related overgrowth spectrum — EPIK-P2 (NCT04589650) [Internet]. ClinicalTrials.gov; results posted in the registry in 2025 [accessed 2026 Oct 4]. Available at: https://clinicaltrials.gov/study/NCT04589650
79. Turner NC, Oliveira M, Howell SJ, Dalenc F, Cortes J, Gomez Moreno HL, et al. Capivasertib in hormone receptor-positive advanced breast cancer. N Engl J Med. 2023;388(22):2058-70. doi:10.1056/NEJMoa2214131. PMID: 37256976.
80. Bussaglia E, Pujol RM, Gil MJ, Martí RM, Tuneu A, Febrer MI, et al. PTEN mutations in eight Spanish families and one Brazilian family with Cowden syndrome. J Invest Dermatol. 2002;118(4):639-44. doi:10.1046/j.1523-1747.2002.01728.x. PMID: 11918710.
81. de Arruda JAA, Freire CH, Leite TDB, Noce CW, Cavalcante IL, Vogel JO, et al. Orofacial manifestations in a middle-aged woman with Cowden syndrome: a case image. Head Neck Pathol. 2023;17(4):1071-4. doi:10.1007/s12105-023-01586-2. PMID: 37735289.
82. Moreno Fernandes M, Rodrigues Neto M, Sá Pinto M, Pena Fernandes T, Maia AC, Ayres Pereira I, et al. A novel PTEN frameshift variant in a child with autism spectrum disorder and macrocephaly: a case report. Cureus. 2026;18(1):e100834. doi:10.7759/cureus.100834. PMID: 41646565.
83. Flores IL, Romo SA, Tejeda Nava FJ, Roger dos Santos Silva A, Vargas PA, Paes de Almeida O. Oral presentation of 10 patients with Cowden syndrome. Oral Surg Oral Med Oral Pathol Oral Radiol. 2014;117(4):e301-10. doi:10.1016/j.oooo.2014.01.015. PMID: 24560406.
84. Hammerschmidt M, Lourenço SV, Nico MMS. A clinicopathological study of the oral lesions of Cowden disease. J Oral Pathol Med. 2017;46(8):637-43. doi:10.1111/jop.12519. PMID: 27889943.
85. Hendricks LAJ, Verbeek KCJ, Schuurs-Hoeijmakers JHM, de Putter R, Brems H, Van Daele SH, et al. Cancer prognosis and treatment results in patients with PTEN hamartoma tumour syndrome (PHTS) — a European cohort study. BJC Rep. 2025;3(1):42. doi:10.1038/s44276-025-00157-y. PMID: 40468016.