Thyroid Pathology in PTEN -Hamartoma Tumor Syndrome: Characteristic Findings of a Distinct Entity

Abstract

Background:

Phosphatase and tensin homolog deleted on chromosome ten (PTEN)-hamartoma tumor syndrome (PHTS) is a complex disorder caused by germline inactivating mutations of the PTEN tumor suppressor gene. PHTS includes Cowden syndrome (CS), Bannayan-Riley-Ruvalcaba syndrome (BRRS), and Proteus-like syndromes. Affected individuals develop both benign and malignant tumors in a variety of tissues, including the thyroid. This study is to better characterize and describe the thyroid pathology within the different entities of this syndrome, and examine whether there is an association between specific thyroid findings and different PTEN mutations.

Methods:

Twenty patients with known PTEN mutations, and/or clinical diagnosis of PHTS, and thyroid pathology were identified: 14 with CS and 6 with BRRS.

Results:

Thyroid pathology findings were as follows: multiple adenomatous nodules in a background of lymphocytic thyroiditis (LT) in 75%, papillary carcinoma in 60%, LT alone in 55%, follicular carcinoma in 45%, C-cell hyperplasia in 55%, and follicular adenomas in 25%. Within the papillary carcinoma group, there were 6 microcarcinomas, 5 follicular variants, and 1 classical type.

Conclusions:

There were no morphologic differences between the thyroid findings in CS and BRRS. Also, there was no correlation between specific PTEN germline mutations (exons 5, 6, and 8) and pathologic findings. Distinctive and characteristic findings in PHTS include multiple unique adenomatous nodules in a background of LT, and C-cell hyperplasia; it is vital that pathologists recognize the classical histologic features of this syndrome to alert clinicians to the possibility of this syndrome in their patients.

Introduction

P hosphatase and tensin homolog deleted on chromosome ten (PTEN)-hamartoma tumor syndrome (PHTS) includes Cowden syndrome (CS), Bannayan-Riley-Ruvalcaba syndrome (BRRS), and Proteus-like syndromes. PHTS is a complex disorder caused by germline inactivating mutations of the PTEN tumor suppressor gene, which maps to 10q23.3 (1). Affected individuals with CS develop both benign and malignant tumors in a variety of tissues, such as breast, uterus, and thyroid (2 –5). Two-thirds of CS patients develop thyroid tumors; the pathologic findings in this syndrome have been described as involving the follicular cells, and include multinodular goiter, multiple adenomatous nodules (MAN), follicular adenoma, follicular carcinoma (FC), and, less frequently, papillary thyroid carcinoma (PTC) (3,6 –9). Although diagnostic criteria for CS have been established for more than a decade, there are no agreed upon international criteria for the diagnosis of BRRS, which is characterized by macrocephaly, pigmented macules of the penis, lipomas, vascular hamartomatous lesions, and hamartomatous intestinal polyposis. The rate of occurrence and histologic types of thyroid lesions in BRRS have not been widely reported, but have appeared similar to those seen in CS, suggesting a single entity. We investigate, classify, and describe the thyroid pathology in both CS and BRRS in a large number of patients and correlate the findings with clinical findings and specific PTEN mutations.

Materials and Methods

All thyroid lesions from patients with known PTEN mutations and/or clinical diagnosis of CS or BRRS were retrieved from the Departments of Pathology of Brigham and Women's Hospital, Children's Hospital Boston, Boston, MA, and Hopitaux Universitaires de Geneve, Switzerland. Specimen photographs, histopathology, and clinical and molecular findings were reviewed. Cases were classified according to the World Health Organization (WHO) classification of thyroid tumors (2). In addition, several studies addressing follicular lesions were reviewed (10 –19). Calcitonin immunostaining was performed using a prediluted polyclonal antibody (Cell Marque Corporation, Hot Springs, AZ). C-cell mapping was performed and population density was estimated using the following criteria, as set forth in previous studies (9,20,21): immunostained sections were scanned at low magnification (× 100) to detect clusters of C-cells, which were then enumerated at high magnification (× 400). When a cluster of at least 50 C-cells in one low power field (× 100) was identified, a diagnosis of C-cell hyperplasia (CCH) was made. Germline mutations of PTEN were analyzed by bidirectional DNA sequence analysis of the 9 exons of the PTEN gene and their respective splice sites (22).

To illustrate the clinical complexity of this syndrome, we have summarized the clinicopathologic findings in these patients in Tables 1 and 2.

Table 1.

Clinicopathologic Findings in Patients with PTEN-Hamartoma Tumor Syndrome

Case	Age ^a /sex	Diagnosis	Family history	Clinicopathologic findings
1	23/M	BRRS	PGF: gastric ca, GI polyps PGM: breast ca, lymphoma Paternal uncle (2): GI polyps	Macrocephaly, penile freckles, developmental delay, autoimmune thyroiditis. Gingival neuromas, dermal neuromas, storiform collagenomas, pilar cysts, colorectal ganglioneuromas and inflammatory/hamartomatous polyps, gastric fundic gland polyps.
2	38/F	CS	Son: BRRS Family members: thyroid ca, breast ca, uterine ca, macrocephaly, palmar skin tumors.	Ductal carcinoma in situ (ER/PR+)
3	34/F	CS	Mother: medullary thyroid ca, breast ca	Invasive ductal carcinoma [ER/PR+, Her2/neu (−)]
4	64/F	CS	Son: macrocephaly Mother: GI polyps, Alzheimer's disease Paternal uncles (2): gastric ca	Oral mucosal lesions, numerous skin “papillomas” Ductal carcinoma in situ, uterine adenocarcinoma in situ.
5	22/F	CS	Father: macrocephaly, cleft palate, single kidney PGM: brain tumor, bilateral breast ca Paternal aunt: breast ca MGM: lung ca	Macrocephaly, thickened tongue with prominent papillae. Pilar cysts, colonic ganglioneuromas.
6	48/F	CS	Breast ca, thyroid ca	N/A
7	12/M	BRRS	N/A	Penile freckles, papillomatosis of the lips and tongue.
8	13/M	BRRS	Mother, father: PTEN mutation negative PGF: gastric ca PGM: breast ca	Macrocephaly, penile freckles. Colonic ganglioneuromas and inflammatory polyps, upper GI polyps
9	76/F	CS	N/A	Invasive ductal carcinoma, pilar cysts, “hemolymphangioma,” basal cell carcinomas
10	9/F	BRRS	Mother: hypothyroidism, breast ca Maternal uncle: CS (same PTEN mutation)	Macrocephaly, short stature, premature thelarche. Upper GI hamartomatous/inflammatory polyps, lipomas
11	13/M	BRRS	N/A	Mucocutaneous lesions Lower extremity arteriovenous malformation
12	52/F	CS	Mother: CS	N/A
13	24/F	CS	Sister: macrocephaly. Father: lung ca, pancreatic ca. Mother: benign thyroid disease Maternal cousin: breast ca, thyroid ca	Macrocephaly
14	73/F	CS	Breast ca	Breast ca, endometrial ca
15	20/M	BRRS	N/A	Macrocephaly, palmar pits. Arteriovenous malformations, lipomas, renal cell carcinoma.
16	53/M	CS	Daughter: case 17	Lipomas, colon polyps, oral fibroepithelial lesions, neuromas, fordyce nodule, ethmoid papillary adenocarcinoma.
17	26/F	CS	Father: case 16	Lipomas, gingival papules, dermal nevi, melanoma
18	43/F	CS	Daughters: cases 19, 20	Macrocephaly. Breast ca. (bilateral)
19	18/F	CS	Mother: case 18 Sister: case 20	Cyanotic heart malformation, pulmonary hypertension. Multiple vertebral “hemangiomas”
20	13/F	CS	Mother: case 18 Sister: case 19	Macrocephaly, cortical frontal dysplasia. Ocular schwannoma

Age at diagnosis of CS/BRRS.

BRRS, Bannayan-Riley-Ruvalcaba syndrome; Ca, carcinoma; CS, Cowden syndrome; ER, estrogen receptor; F, female; GI, gastrointestinal; M, male; MGM, maternal grandmother; MGF, maternal grandfather; N/A, not available; PGF, paternal grandfather; PGM, paternal grandmother; PR, progesterone receptor; PTEN, phosphatase and tensin homolog deleted on chromosome ten.

Table 2.

Thyroid Pathology Findings in Patients with PTEN-Hamartoma Tumor Syndrome

Case	Gross findings	Pathologic findings	C-cell hyperplasia ^a
1	18 g; multinodular (0.1–2.5 cm)	FC (MI, 0.6 cm) MAN LT PTC (micro)	Diffuse
2	39 g; multinodular (0.1–2.7 cm)	LT Nodular hyperplasia PTC (multifocal, 0.3 and 0.5 cm)	None
3	11.6 g (lobe); dominant nodule (1.5 cm), multinodular background	MAN LT PTC (micro)	Diffuse
4	14 g; multinodular (0.3–0.9 cm)	MAN	None
5	Dominant nodule (6.1 cm)	MAN LT PTC (6.1 cm)	None
6	110 g	MAN PTC (multifocal, 8.0, 1.8, 1.8 cm)	None
7	N/A	FC (MI, 4.5 cm) MAN LT	Diffuse
8	41.6 g; multinodular (0.5–2.5 cm)	MAN LT	Focal
9	12.5 g (lobe); dominant nodule (0.8 cm), multinodular background (0.3–1.5 cm)	Follicular adenomas PTC (multifocal, 0.5, 0.8 cm)	None
10	8.2 g (lobe); dominant nodule (2.2 cm), multinodular background	Follicular adenomas (0.5, 2.2 cm) PTC (micro)	Focal
11	N/A	FC (MI, 1.8 cm) LT	Focal
12	92 g; multinodular (0.3–8.5 cm)	MAN (>100) LT Anomalous intrathyroidal vessels	None
13	97 g; multinodular (0.5–4.2 cm)	FC (MI, 0.5 cm) FC (MI, 1.2 cm) MAN (>20)	Diffuse
14	42.6 g; multinodular (0.4–2.2 cm)	MAN PTC (micro)	Focal
15	Age 7: N/A Age 20: 4 g (lobe); multinodular (0.2–0.9 cm)	Age 7: FC (MI, 2.8 cm) LT Age 20: PTC (multifocal, up to 0.7 cm) Nodular hyperplasia	None
16	25 g; multinodular(0.5–0.8 cm)	FC (MI, 0.5 cm) MAN Follicular adenoma (2 cm) Nodular hyperplasia PTC (micro)	Diffuse
17	20 g; multinodular (0.7–2.3 cm)	FC (MI, 1.1 cm) MAN LT Nodular hyperplasia	Focal
18	24 g; multinodular (up to 1.5 cm)	MAN Follicular adenoma Nodular hyperplasia	None
19	55 g; multinodular (up to 3.5 cm)	FC (MI, 3.5 cm) MAN LT PTC (2.4 cm) PTC (micro)	Diffuse
20	37 g; dominant nodule (3.0 cm)	MAN Follicular adenoma (3.0 cm)	None

Defined as >50 cells/100× field.

FC, follicular carcinoma; PTC, papillary thyroid carcinoma; PTC (micro), papillary thyroid microcarcinoma; MI, minimally invasive; LT, lymphocytic thyroiditis; MAN, multiple adenomatous nodules.

Results

Pathologic findings

Of the 20 patients with PHTS (CS [n = 14]; BRRS [n = 6]), 14 were female and 6 were male. The majority (11/20, 55%) had a family history of PHTS, and the remainder (9/20, 45%) were classified as simplex (no apparent family history); all patients fulfilled the diagnostic criteria for PHTS. The mean age at diagnosis of PHTS was 33.7 years (9–76 years). When subclassified into the two syndromes, the mean age at diagnosis of BRRS was 15 years (9–23 years) and 41.7 years (13–76 years) for CS. The mean age at diagnosis of a thyroid lesion was 33.3 years (range 7–73 years); 15 years (7–24 years) for BRRS, and 41.1 years (22–73 years) for CS. Thus, identification of a thyroid lesion and a diagnosis of PHTS were often concurrent (Tables 1 and 2). Further inspection revealed that in 15 patients (cases 2–7, 11–14, and 16–20) the thyroid diagnosis was made at the same time as the PHTS diagnosis. In three patients, the PHTS diagnosis preceded the thyroid diagnosis (case 1 by 1 year, case 8 by 10 years, and case 10 by 2 years), and in two patients, the thyroid pathology diagnosis preceded the ascertainment of PHTS (case 9 by 8 years, and case 15 by 13 years). For some patients, thyroid pathology was diagnosed during evaluation and screening for PHTS, and for others, the PHTS diagnosis and work-up, including PTEN mutation analysis, was triggered by pathologic findings.

Seventeen patients underwent a total thyroidectomy, and the remaining three opted for lobectomy. Tumors were classified according to the World Health Organization classification system (2), and all thyroids had multiple pathologic findings (Table 2). The most common diagnosis was MAN in 75% (n = 15; Figs. 1 –3), followed by PTC in 60% (n = 12; 4 multifocal follicular variants, 1 follicular variant, 1 classic type, and 6 microcarcinomas), lymphocytic thyroiditis (LT) in 55% (n = 11; Fig. 2b), CCH in 55% (n = 11, Fig. 4), FC in 45% (n = 9), nodular hyperplasia in 25% (n = 5), and multiple and solitary follicular adenomas in 25% (FA, n = 5).

FIG. 1.

Thyroid cut surface. Multiple adenomatous nodules (MAN), with the characteristic macroscopic appearance showing well-circumscribed, solid, tan nodules. Close-up view of the cut surface.

FIG. 2.

Scanning view of thyroid with MAN. MAN with little intervening normal thyroid (a; H&E, original magnification), and in a background of lymphocytic thyroiditis (b; H&E, original magnification).

FIG. 3.

MAN. Nodules are well-circumscribed but unencapsulated. They have a solid and cellular appearance and are separated by scattered residual follicles (H&E; 100 ×).

FIG. 4.

C-cell hyperplasia (a, HE, 400 ×; b, same area, calcitonin immunostain, 200 ×).

MAN in a background of LT was the most frequent histologic finding; MAN were diagnosed in 15 specimens, which represented 50% (3/6 cases) of BRRS patients and 86% (12/14 cases) of CS patients in this series. The size of the adenomatous nodules ranged from 0.3 to 8.5 cm in greatest dimension, and the number of nodules per gland ranged from over 20 to more than 100 (Table 2). Grossly, these nodules were numerous, firm, white/tan, and homogenous, without gelatinous colloid or prominent secondary change, distinguishing them from the more common nodular hyperplasia. Microscopically, these adenomatous nodules had a solid architecture with a cellular composition, frequently lacking colloid, and without prominent cystic degeneration or heterogeneity among nodules. The majority of nodules were well-circumscribed and unencapsulated, though a few had a discontinuous rim of thin fibrous tissue simulating a capsule. Adjacent normal thyroid tissue was scanty, compressed, and consisted of a thin rim of occasional follicles in a background of LT. There were no histologic differences between the adenomatous nodules in CS versus BRRS.

CCH, as shown by calcitonin immunostaining (Fig. 4), was also a frequent finding; it was seen in 55% (11/18) of all patients. Of these, five were in BRRS patients (5/6, 83%) and six in CS patients (6/14, 43%). Thus, CCH can be seen in any PHTS patient. Although CCH is not as well defined in the pediatric age group (23,24), it appears to be relatively more common in BRRS patients.

Thyroid carcinoma was identified in the majority of patients in our study (15/20, 75%) (Table 2). For FC specifically, the average age at diagnosis in PHTS was 22.1 years (7–53 years); 32 years (18–53 years) in CS patients, and 14 years (7–24 years) in BRRS. Of the patients with CS, 35% (5/14 cases) had FC, only one of which was multicentric. In addition, we found FC in four patients with BRRS (4/6 cases; 67%), indicating that FC was frequent in the pediatric BRRS population. None of the malignancies were diagnosed preoperatively; all carcinomas were diagnosed by careful pathological examination.

Unexpectedly, we also identified PTCs (12/20 cases, 60%), which have rarely been described in these patients; PTCs occurred in patients with CS (9/14 cases, 64%) and BRRS (3/6 cases, 50%). Six were microcarcinomas (cases 1, 3, 10, 14, 16, and 19) ranging in size from 0.15 to 0.5 cm, four were multifocal follicular variant of PTC (cases 2, 6, 9, and 15), one case was a classic type PTC (case 5), and one was a follicular variant of PTC (case 19) (the thyroid pathology findings and size for all malignant tumors is included on Table 2). The mean age at diagnosis of PTC in PHTS was 32.4 years (11–73 years), 31.6 years in CS (18–73 years), and 17.5 years in BRRS (11–24 years).

In our series, as in the literature, the most commonly reported extrathyroidal malignancy was ductal breast carcinoma (5/20 cases), all of which occurred in CS patients. Two tumors were invasive ductal carcinoma (cases 3 and 9), two were ductal carcinoma in situ (cases 2 and 4), and additional information was not available in one case (case 14). The mean age at diagnosis of patients with ductal breast carcinoma was 48.9 years (33–69 years). One extrathyroidal malignancy was reported in our BRRS patients (case 15, renal cell carcinoma at age 11). No uterine malignancy was documented in our series.

Molecular genetics findings

Analysis of germline PTEN mutations was available for review in 13 patients; no mutation was identified in one patient, and six patients (although they fulfilled the diagnostic criteria for PHTS clinically) either declined mutation testing or requested that their results be withheld. The results of the mutation analyses are summarized in Table 3.

Table 3.

Molecular Genetic Findings Patients with PTEN-Hamartoma Tumor Syndrome

Case	Sex	Diagnosis	PTEN mutation	Functional effect (predicted)
1	M	BRRS	c.389G > A/p.R130Q, exon 5	Missense mutation: Arginine to glutamine at AA 130
2	F	CS	c.540T > A/p.Y180X, exon 5	Nonsense mutation
3	F	CS	Negative	—
4	F	CS	N/A	—
5	F	CS	c.609_611delTCCinsATAAAT, exon 6	PT, at AA 204
6	F	CS	N/A	—
7	M	BRRS	c.968dupA, exon 8	FS → PT 1 residue downstream
8	M	BRRS	c.389G > A/p.R130Q, exon 5	Missense mutation: Arginine to glutamine at AA 130
9	F	CS	N/A	—
10	F	BRRS	c.512-513insA, exon 6	FS → PT 8 residues downstream
11	M	BRRS	c.499_505delACTATTC, exon 6	FS → PT 13 residues downstream
12	F	CS	N/A	—
13	F	CS	N/A	—
14	M	CS	N/A	—
15	M	BRRS	c.491delA, exon 5	FS → PT 2 residues downstream
16	M	CS	c.959T > A/p.L320X, exon 8	Nonsense mutation
17	F	CS	c.959T > A/p.L320X, exon 8	Nonsense mutation
18	F	CS	c.406T > C/p.C136R, exon 5	Missense mutation: Cysteine to arginine at AA 136
19	F	CS	c.406T > C/p.C136R, exon 5	Missense mutation: Cysteine to arginine at AA 136
20	F	CS	c.406T > C/p.C136R, exon 5	Missense mutation: Cysteine to arginine at AA 136

→ , leading to; AA, amino acid; FS, frame shift; PT, premature termination.

The majority of mutations occurred in exon 5 (7/13, 54%), followed by exon 6 (3/13, 23%) and exon 8 (3/13, 23%), confirming, as previously reported, that the most frequent site of germline mutations in PHTS is exon 5. Mutations identified in exon 5 were as follows: five missense mutations, one nonsense mutation, and one deletion that produced a frame shift leading to a premature termination. Exon 6 showed deletions and/or insertions, and exon 8 revealed two nonsense mutations, and a duplication that produced a frame shift leading to a premature termination.

Germline PTEN mutations and their associated pathologic findings are summarized as follows: exon 5 mutations (patients 1, 2, 8, 15, and 18–20) had FC, MAN, PTC, nodular hyperplasia (NH), and FA; exon 6 mutations (patients 5, 10, and 11) had FC, PTC, and FA; and exon 8 mutations (patients 7, 16, and 17) had FC, MAN, NH, and FA. Thus, we found no correlation between specific germline mutations and thyroid pathologic findings.

Discussion

CS and BRRS (Tables 4 and 5) are the major entities comprising the PHTS, which is caused by germline mutations of the PTEN gene and inherited in an autosomal dominant fashion (4).

Table 4.

Diagnostic Criteria for Cowden Syndrome

Definitive diagnosis

PTEN mutation identified

Operational diagnosis

Mucocutaneous lesions (one of the following)

Fibrous papules (≥6); at least 3 must be trichilemmomas

Cutaneous facial papules and oral mucosal papillomatosis

Oral mucosal papillomatosis and acral keratoses

Palmo-plantar keratoses (≥6)

≥2 Major criteria (must include macrocephaly or LDD)

Macrocephaly (≥97th percentile)

LDD (dysplastic gangliocytoma of the cerebellum)

Thyroid FC

Breast adenocarcinoma

Endometrial cancer

1 Major and 3 minor criteria

≥4 Minor criteria

Developmental delay or mental retardation

Mucocutaneous lipomas or fibromas

Breast fibroadenomas or fibrocystic change

Multinodular goiter

Thyroid follicular adenoma

Hamartomatous gastrointestinal polyps

Uterine leiomyomas

Bicornuate uterus

Horseshoe kidney

Renal cell carcinoma

Operational diagnosis in family members of patient with a diagnosis of CS

Meets above criteria, or

Any single major criterion, or

2–3 minor criteria, or

Personal history of BRRS

Diagnostic criteria refer to the current operational diagnostic criteria for CS, which are available online at the National Comprehensive Cancer Network (www.nccn.org) (5,19,20,23).

LDD, Lhermitte-Duclos disease.

Table 5.

Diagnostic Criteria for Bannayan-Riley-Ruvalcaba Syndrome

There are no agreed-upon international criteria for the diagnosis of BRRS, but there are two proposed diagnostic protocols

At least 3 of the following 4 features: macrocephaly, lipomatosis, hemangiomas, and speckled penis

At least 2 of the 3 features: macrocephaly, hamartomas (including at least 1 lipoma, hemangioma, or intestinal polyp), and speckled penis

BRRS generally presents in childhood with delayed motor and intellectual development, and other described features include thyroid adenomas, LT/Hashimoto's thyroiditis, lymphatic/vascular malformations, high arched palate, scoliosis, seizures, lipid storage myopathy, and joint hyperextensibility (5,19,43).

PTEN is a tumor suppressor gene located on 10q23.3 (1). The function of PTEN is not entirely understood, but it is a major phosphatase for phosphoinositide-3,4,5-triphosphate. By downregulating the levels of phosphoinositide-3,4,5-triphosphate, PTEN produces an inhibitory (tumor suppressor) effect on the PI3P/Akt pathway, an important carcinogenesis pathway. It is proposed that PTEN has important activity both in the cytoplasm and nucleus. Nuclear PTEN might be required for cell cycle arrest by downregulating cyclin D1 and preventing phosphorylation of mitogen-activated protein kinase pathway, whereas cytoplasmic PTEN seems to be required for apoptosis by downregulating the phosphorylation of AKT and upregulating p27. Therefore, loss of PTEN function results in escape from programmed cell death and G1 arrest in the cell cycle. The gene is composed of 9 exons, with a 1209-bp coding sequence, which is predicted to encode a 403-amino acid protein. Mutations are dispersed throughout the 9 exons of the PTEN gene, with ∼40% of mutations located in exon 5, which represents 20% of the coding sequence and is known as the mutational “hot spot” in PHTS. Very few germline mutations have been reported in exon 1, and none in exon 9 (25). Seventy-five percent of the germline mutations result in truncated protein, lack of protein, or dysfunctional protein (26 –31).

In contrast to medullary thyroid carcinoma (MTC), where up to a quarter is associated with a familial syndrome, only 5% of non-MTCs are associated with a heritable condition. These familial non-MTC can be divided into two distinct groups (8); those occurring in tumor syndromes where thyroid involvement is a minor component and those occurring in syndromes where non-MTCs are the predominant tumor encountered. The first group includes familial adenomatous polyposis, Carney complex type 1, Werner syndrome, and the focus of this study, PHTS. The second group of familial non-MTC syndromes, in which non-MTCs are the major finding, includes the following rare entities; pure familial PTC (fPTC), with or without oxyphilia, fPTC with papillary renal cell carcinoma, and fPTC with multinodular goiter (2,4,6,8).

Although PTEN mutations define this syndrome, they are rare in sporadic thyroid carcinomas (32). In CS, germline intragenic PTEN mutations have been found in up to 81% of classically affected patients and gene deletions have been described in very rare cases (5,30). In BRRS, intragenic mutations are seen in 60% of patients (5,30). In our series, the majority of mutations occurred in exon 5 (7/13 cases, 54%), followed by exon 6 (3/13 cases, 23%) and exon 8 (3/13 cases, 23%), confirming that the most frequent site of germline mutations in PHTS is exon 5. Case 3 (CS) had a negative mutation test, which is expected to occur in ∼10%–15% of individuals who meet diagnostic criteria for CS (25). However, the specific germline mutation does not correlate with specific thyroid pathology (Tables 2 and 3), including CCH.

Studies have failed to demonstrate a consistent genotype–phenotype relationship in PHTS (22,25,30,33,34). The earliest phenotypic features associated with PTEN mutations are macrocephaly and mucocutaneous lesions. Hamartomas, including gastrointestinal polyps, are often seen, and malignancies develop over time (33). BRRS often presents during childhood, and as of yet, an international consensus on the criteria for diagnosis has not been reached. Marsh et al. (22) defined the criteria as least three of the following four features: macrocephaly, lipomatosis, hemangiomas, and speckled penis. Parisi et al. (35) defined BRRS as at least two of three features, including macrocephaly, hamartomas (at least one lipoma, hemangioma, or intestinal polyp), and penile macules. Of note, it has recently been shown that the hemangiomas in BRRS are intramuscular hamartomatous proliferations of abnormal blood vessels and fibroadipose tissue (36). Other features described in BRRS include thyroid adenomas, Hashimoto thyroiditis, and lymphatic malformations, among others (30,37) (Table 5). In contrast, CS has well-established diagnostic criteria (Table 4). Careful phenotype analysis will provide support for the assertion that BRRS and CS are actually one condition, presenting at different stages (33). This study helps to support this observation by the lack of genotype–phenotype correlation in the thyroid gland.

More than 90% of individuals affected with CS manifest a phenotype by age 20 (6,38). By the end of their third decade, nearly all (99%) develop the pathognomonic mucocutaneous lesions, although any feature may be present, including thyroid lesions (Table 4). The spectrum of disorders related to PHTS has continued to evolve since 1995, when diagnostic criteria for CS were introduced. The current diagnostic criteria for CS consider FC (present in 10%–15%) a major criterion, whereas multinodular goiter and follicular adenomas are minor criteria (present in 50%–67%) (5,38,39). Our histologic findings in this study suggest that the multinodular goiter seen in PHTS patients is commonly due to characteristic, distinct MAN, rather than the adenomatous changes of old age. Nodular hyperplasia, a common thyroid disease, has a heterogenous, hyperemic gross appearance with nodules containing gelatinous colloid, and frequent secondary changes such as calcification and cystic degeneration. Histologically, they are unencapsulated, heterogenous, and composed of large, irregular, dilated follicles with hyperplastic epithelium and abundant colloid; only occasional nodules are densely cellular. In contrast, the adenomatous nodules in PHTS are firm, yellow-tan, lack gelatinous colloid and do not exhibit secondary changes. Histologically, the adenomatous nodules are solid, cellular, and composed of small follicles lacking abundant colloid. Some nodules may have a thin, discontinuous rim of fibrous tissue simulating a capsule.

In our experience, the MAN seen in PHTS are bilateral, more numerous, and distinct from age-related nodules, with only a scant atrophic rim of normal thyroid follicles. The number of these nodules far exceeds those seen in old age, or in any other inherited syndrome; in this series we report MAN in 75% of PHTS patients (50% of those with BRRS patients and 86% of those with CS). Nodule size (0.3–8.5 cm) and number per thyroid (20–>100) varied widely, but morphologic differences between the lesions of CS and BRRS were not identified. We are of the opinion that the International Cowden's Consortium CD Diagnosis Criteria/National Comprehensive Cancer Network consider adding MAN as a major criterion for the diagnosis of CS and BRRS (5,34,38 –41).

FC is an important feature of both CS and BRRS. It has been reported that these tumors are frequently multicentric, and are believed to arise from pre-existing follicular adenomas (2,3) We too have found that the majority of carcinomas arise in a background of MAN, though in our series only one FC (case 13) had a multicentric pattern. Management for CS patients has focused on cancer screening, and surveillance recommendations include annual thyroid ultrasounds after the age of 18 (22). If a thyroid nodule measures ≥1 cm, it is recommended that it be evaluated with fine needle aspiration, followed by standard management with surgery and adjunctive therapy if indicated (42,43). In contrast to other familial syndromes (44), prophylactic thyroidectomy in PHTS is not recommended. A cancer risk in BRRS patients has previously been suspected, though unknown (25,33,37); we now report a high frequency of FC (4/6 cases, 67%) in our series, with a mean age at diagnosis of 14 years. Given these findings, as well as the identification of renal cell carcinoma in a young BRRS patient (case 15), we believe that BRRS patients should be followed with similarly rigorous surveillance.

The majority of thyroid lesions occurring in PHTS are of follicular origin; papillary carcinoma has been only rarely associated with this entity (3,7,9). In our study, we identified PTC in 60% of patients (12/20 cases). However, given the prevalence of PTC in the general adult population (25%–34%) (45,46), it is difficult to conclude from these data that PTC is characteristic of PHTS. We do not endorse the addition of PTC as a diagnostic criterion for either CS or BRRS.

MTC is not considered part of the spectrum of PHTS; however, studies have identified CCH in individuals with this syndrome (3,9). C-cell proliferations may be either neoplastic or physiologic; within the neoplastic spectrum, calcitonin producing C-cells have been categorized as a precursor of familial MTC (21,44,47). Physiologic CCH is a reactive process associated with hypercalcemia, hypergastrinemia, goitrous hypothyroidism, Hashimoto thyroiditis, and attributed to peritumoral effect (adjacent to large papillary or follicular neoplasms, and non-Hodgkin lymphoma of the thyroid) (20,23,24). Interestingly, CCH was identified in 11 patients in this series: 5 with BRRS and 6 with CS. The pattern and degree of hyperplasia varied from focal and patchy to diffuse, and was associated with a variety of lesions (Table 2). These findings were most prevalent in BRRS patients, which could represent an age-related phenomenon (9,23,24,40). However, the high number (>>50 per 100 × field) of C-cells, often exceeding the number of follicular cells, more plausibly points to an abnormality of C-cells in PHTS, although the criteria for CCH are not well defined for the pediatric age group. CCH was not associated with age in our study. We recommend that pathologists include PHTS in the differential diagnosis of CCH of the thyroid.

The diagnosis of a thyroid lesion preceded the diagnosis of PHTS by many years in some patients in this study, whereas in others, a work-up for PHTS (including PTEN mutation analysis) was triggered by pathologic findings in the thyroid (specifically, MAN). We propose an algorithmic approach for the pathologist adressing unusual thyroid pathology in a young patient (Fig. 5). MAN, associated with other thyroid lesions, CCH, and/or FC in young patients should alert the pathologist to the possibility of an inherited tumor syndrome.

FIG. 5.

Algorithmic approach to unusual thyroid pathology in young patients.

In conclusion, characteristic multicentric, bilateral, benign, and malignant thyroid lesions are observed in PHTS. Morphologic differences between thyroid lesions in CS and BRRS were not observed, although CCH was more frequent in the younger BRRS patients. There was no correlation between specific germline mutations and pathologic findings identified. This supports the belief that CS and BRRS represent different points along the spectrum of PHTS. MAN in a background of LT and/or CCH are distinctive findings in this syndrome, and should alert pathologists to notify clinicians of the possibility of PHTS.

Footnotes

Disclosure Statement

The authors declare that no competing financial interests exist.

References

Nelen

, Padberg

, Peeters

, Lin

, van den Helm

, Frants

, Coulon

, Goldstein

, van Reen

, Easton

, Eeles

, Hodgsen

, Mulvihill

, Murday

, Tucker

, MMariman

, Starink

, Ponder

, Ropers

, Kremer

, Longy

, Eng

. 1996. Localization of the gene for Cowden disease to chromosome 10q22-23. Nat Genet, 13:114–116.

DeLellis

, Loyd

, Heitz

. 2004. World Health Organization Classification of Tumours., Pathology and Genetics of Tumours of Endocrine Organs. IARC: Lyon.

Harach

, Soubeyran

, Brown

, Bonneau

, Longy

. 1999. Thyroid pathologic findings in patients with Cowden disease. Ann Diagn Pathol, 3:331–340.

Malchoff

, Malchoff

. 2006. Familial nonmedullary thyroid carcinoma. Cancer Control, 13:106–110.

Zbuk

, Eng

. 2007. Cancer phenomics: RET and PTEN as illustrative models. Nat Rev Cancer, 7:35–45.

Dotto

, Nosé

. 2008. Familial thyroid carcinoma: a diagnostic algorithm. Adv Anat Pathol, 15:332–349.

Kameyama

, Takami

, Miyajima

, Mimura

, Hosoda

, Ito

. 2001. Papillary carcinoma occurring within an adenomatous goiter of the thyroid gland in Cowden's disease. Endocr Pathol, 12:73–76.

Nosé

. 2008. Familial non-medullary thyroid carcinoma: an update. Endocr Pathol, 19:226–240.

Zambrano

, Holm

, Glickman

, Huang

, Perez-Atayde

, Kozakewich

, Shamberger

, Nosé

. 2004. Abnormal distribution and hyperplasia of thyroid C-cells in PTEN-associated tumor syndromes. Endocr Pathol, 15:55–64.

10.

Baloch

, LiVolsi

. 2007. Our approach to follicular-patterned lesions of the thyroid. J Clin Pathol, 60:244–250.

11.

Collini

, Sampietro

, Rosai

, Pilotti

. 2003. Minimally invasive (encapsulated) follicular carcinoma of the thyroid gland is the low-risk counterpart of widely invasive follicular carcinoma but not of insular carcinoma. Virchows Arch, 442:71–76.

12.

Hirokawa

, Carney

, Goellner

, DeLellis

, Heffess

, Katoh

, Tsujimoto

, Kakudo

. 2002. Observer variation of encapsulated follicular lesions of the thyroid gland. Am J Surg Pathol, 26:1508–1514.

13.

LiVolsi

, Baloch

. 2004. Follicular neoplasms of the thyroid: view, biases, and experiences. Adv Anat Pathol, 11:279–287.

14.

Lloyd

, Erickson

, Casey

, Lam

, Lohse

, Asa

, Chan

, DeLellis

, Harach

, Kakudo

, LiVolsi

, Rosai

, Sebo

, Sobrinho-Simoes

, Wenig

, Lae

. 2004. Observer variation in the diagnosis of follicular variant of papillary thyroid carcinoma. Am J Surg Pathol, 28:1336–1340.

15.

Rosai

. 2005. Handling of thyroid follicular patterned lesions. Endocr Pathol, 16:279–283.

16.

Rosai

. 1991. Follicular thyroid neoplasms. Am J Clin Pathol, 95:758–759.

17.

Rosai

, Kuhn

, Carcangiu

. 2006. Pitfalls in thyroid tumour pathology. Histopathology, 49:107–120.

18.

Tischler

, DeLellis

. 2002. Tumors of thyroid follicular epithelium: where have we been and where are we going? Endocr Pathol, 13:267–269.

19.

Fonseca

, Soares

, Cardoso-Oliveira

, Sobrinho-Simoes

. 2006. Diagnostic criteria in well-differentiated thyroid carcinomas. Endocr Pathol, 17:109–117.

20.

Albores-Saavedra

, Monforte

, Nadji

, Morales

. 1988. C-cell hyperplasia in thyroid tissue adjacent to follicular cell tumors. Hum Pathol, 19:795–799.

21.

Wolfe

, Delellis

. 1981. Familial medullary thyroid carcinoma and C cell hyperplasia. Clin Endocrinol Metab, 10:351–365.

22.

Marsh

, Kum

, Lunetta

, Bennett

, Gorlin

, Ahmed

, Bodurtha

, Crow

, Curtis

, Dasouki

, Dunn

, Feit

, Geraghty

, Graham

Jr. , Hodgson

, Hunter

, Korf

, Manchester

, Miesfeldt

, Murday

, Nathanson

, Parisi

, Pober

, Romano

, Eng

. 1999. PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome. Hum Mol Genet, 8:1461–1472.

23.

LiVolsi

. 1997. C cell hyperplasia/neoplasia. J Clin Endocrinol Metab, 82:39–41.

24.

Perry

, Molberg

, Albores-Saavedra

. 1996. Physiologic versus neoplastic C-cell hyperplasia of the thyroid: separation of distinct histologic and biologic entities. Cancer, 77:750–756.

25.

Eng

. 2003. PTEN: one gene, many syndromes. Hum Mutat, 22:183–198.

26.

Starink

, van der Veen

, Arwert

, de Waal

, de Lange

, Gille

, Eriksson

. 1986. The Cowden syndrome: a clinical and genetic study in 21 patients. Clin Genet, 29:222–233.

27.

Fackenthal

, Marsh

, Richardson

, Cummings

, Eng

, Robinson

, Olopade

. 2001. Male breast cancer in Cowden syndrome patients with germline PTEN mutations. J Med Genet, 38:159–164.

28.

Michaels

, Shakir

. 1993. Association of multinodular goiter with breast carcinoma: Cowden's disease. J Endocrinol Invest, 16:909–911.

29.

Sabate

, Gomez

, Torrubia

, Blancas

, Sanchez

, Alonso

, Lerma

. 2006. Evaluation of breast involvement in relation to Cowden syndrome: a radiological and clinicopathological study of patients with PTEN germ-line mutations. Eur Radiol, 16:702–706.

30.

Marsh

, Coulon

, Lunetta

, Rocca-Serra

, Dahia

, Zheng

, Liaw

, Caron

, Duboue

, Lin

, Richardson

, Bonnetblanc

, Bressieux

, Cabarrot-Moreau

, Chompret

, Demange

, Eeles

, Yahanda

, Fearon

, Fricker

, Gorlin

, Hodgson

, Huson

, Lacombe

, Eng

. 1998. Mutation spectrum and genotype-phenotype analyses in Cowden disease and Bannayan-Zonana syndrome, two hamartoma syndromes with germline PTEN mutation. Hum Mol Genet, 7:507–515.

31.

Soysal

, Tate

, Polat

, Aktepe

, Sivaci

, Imirzalioglu

. 2009. Analysis of PTEN gene mutations in a Turkish patient with Cowden syndrome. Genet Test Mol Biomarkers, 13:547–551.

32.

DeLellis

. 2006. Pathology and genetics of thyroid carcinoma. J Surg Oncol, 94:662–669.

33.

Lachlan

, Lucassen

, Bunyan

, Temple

. 2007. Cowden syndrome and Bannayan Riley Ruvalcaba syndrome represent one condition with variable expression and age-related penetrance: results of a clinical study of PTEN mutation carriers. J Med Genet, 44:579–585.

34.

Pilarski

. 2009. Cowden syndrome: a critical review of the clinical literature. J Genet Couns, 18:3–27.

35.

Parisi

, Dinulos

, Leppig

, Sybert

, Eng

, Hudgins

. 2001. The spectrum and evolution of phenotypic findings in PTEN mutation positive cases of Bannayan-Riley-Ruvalcaba syndrome. J Med Genet, 38:52–58.

36.

Tan

, Baris

, Burrows

, Robson

, Alomari

, Mulliken

, Fishman

, Irons

. 2007. The spectrum of vascular anomalies in patients with PTEN mutations: implications for diagnosis and management. J Med Genet, 44:594–602.

37.

Longy

, Coulon

, Duboue

, David

, Larregue

, Eng

, Amati

, Kraimps

, Bottani

, Lacombe

, Bonneau

. 1998. Mutations of PTEN in patients with Bannayan-Riley-Ruvalcaba phenotype. J Med Genet, 35:886–889.

38.

Eng

. 2000. Will the real Cowden syndrome please stand up: revised diagnostic criteria. J Med Genet, 37:828–830.

39.

Liaw

, Marsh

, Li

, Dahia

, Wang

, Zheng

, Bose

, Call

, Tsou

, Peacocke

, Eng

, Parsons

. 1997. Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet, 16:64–67.

40.

Gibson

, Peng

, Croker

. 1982. Age-associated C-cell hyperplasia in the human thyroid. Am J Pathol, 106:388–393.

41.

Pilarski

, Eng

. 2004. Will the real Cowden syndrome please stand up (again)? Expanding mutational and clinical spectra of the PTEN hamartoma tumour syndrome. J Med Genet, 41:323–326.

42.

Marqusee

, Benson

, Frates

, Doubilet

, Larsen

, Cibas

, Mandel

. 2000. Usefulness of ultrasonography in the management of nodular thyroid disease. Ann Intern Med, 133:696–700.

43.

Cooper

, Doherty

, Haugen

, Kloos

, Lee

, Mandel

, Mazzaferri

, Mclver

, Sherman

, Tuttle

. American Thyroid Association Guidelines Taskforce 2006 Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid, 16:109–142.

44.

Skinner

, Moley

, Dilley

, Owzar

, Debenedetti

, Wells

Jr . 2005. Prophylactic thyroidectomy in multiple endocrine neoplasia type 2A. N Engl J Med, 353:1105–1113.

45.

Bondeson

, Ljungberg

. 1984. Occult papillary thyroid carcinoma in the young and the aged. Cancer, 53:1790–1792.

46.

Harach

, Franssila

, Wasenius

. 1985. Occult papillary carcinoma of the thyroid. A “normal” finding in Finland. A systematic autopsy study. Cancer, 56:531–538.

47.

Mears

, Diaz-Cano

. 2003. Difference between familial and sporadic medullary thyroid carcinomas. Am J Surg Pathol, 27:266–267.