Article Text


Male breast cancer in Cowden syndrome patients with germlinePTEN mutations


Cowden syndrome (CS) (OMIM 158350) is a multiple hamartoma syndrome associated with germline mutations in thePTEN tumour suppressor gene. While CS is characterised most commonly by non-cancerous lesions (mucocutaneous trichilemmomas, acral and palmoplantar keratoses, and papillomatous papules), it is also associated with an increased susceptibility to breast cancer (in females) and thyroid cancer, as well as non-cancerous conditions of the breast and thyroid. Here we report two cases of male breast cancer occurring in patients with classical CS phenotypes and germline PTEN mutations. The first subject was diagnosed with CS indicated primarily by mucocutaneous papillomatosis, facial trichilemmomas, and macrocephaly with frontal bossing at the age of 31 years. He developed breast cancer at 41 years and subsequently died of the disease. A PTENmutation, c.802delG, was identified in this subject, yet none of his family members showed evidence of a CS phenotype, suggesting that thisPTEN mutation may be a de novo occurrence. The second subject had a CS phenotype including multiple trichilemmomas and thyroid adenoma, developed male breast cancer at 43 years, and died of the disease at 57 years. He was a carrier of aPTEN mutation c.347-351delACAAT that cosegregated with the CS phenotype in affected family members. These two cases of male breast cancer associated with germlinePTEN mutations and the CS phenotype suggest that CS may be associated with an increased risk of early onset male as well as female breast cancer.

  • PTEN
  • male breast cancer
  • Cowden syndrome

Statistics from

Cowden syndrome (CS) is an autosomal dominant cancer susceptibility syndrome characterised by multiple hamartomas.1-4 While CS is generally considered a rare condition, it is associated with highly variable, subtle, and protean symptoms and signs, suggesting that it may be underdiagnosed and under-reported.5 6 CS presentation usually occurs in patients in their twenties and is most commonly indicated by mucocutaneous lesions including trichilemmomas, acral and palmoplantar keratoses, and verucoid or papillomatous papules. Other common features include gastrointestinal hamartomas, neurological disorders including macrocephaly and mental retardation, and genitourinary abnormalities including tumours and malformations.5 7-11

The most common cancers associated with CS are breast (in females) and thyroid carcinomas.10Women in CS families have a lifetime risk of developing breast cancer of 25-50%, and an average age of diagnosis 10 years younger than sporadic breast cancer patients.8 10 As with other hereditary breast cancers, CS associated breast cancers are also more likely than sporadic breast cancers to be multifocal and bilateral in origin. Thyroid cancer occurs in about 3-10% of CS patients, and these cancers are predominantly follicular carcinomas.8 12 In addition to the hallmark breast and thyroid cancers, malignancies in other tissues have also been reported in CS patients, including glioblastomas, mucocutaneous carcinomas, non-small cell carcinomas of the lung, and various gastrointestinal and genitourinary carcinomas, as reviewed by Eng and Parsons.13

CS is associated with germline mutations in the tumour suppressor genePTEN (phosphatase and tensin homologue mutated on chromosometen; also known as MMAC1or TEP1).PTEN mutations are found in 13-81% of CS families and the PTEN locus is associated with loss of heterozygosity (LOH) in a subset of CS tumours, depending on ascertainment criteria.14-20 ThePTEN gene encodes a lipid phosphatase that dephosphorylates phosphatidylinositol 3,4 diphosphate (PtdIns(3,4)P2) and phosphatidylinositol 3,4,5 triphosphate (PtdIns(3,4,5)P3) to PtdIns(4)P and PtdIns(4,5)P2, respectively.21-23 This lipid phosphatase activity participates in regulating the phosphoinositide 3-kinase (PI3K) regulatory network. Thus, while it is not clear which of the many potential targets of the PTENphosphatase are most relevant to tumour suppression, its involvement in PI3K regulation suggests that it may play a role in both inhibiting cell cycle progression and promoting apoptosis.24-27

Since PTEN was characterised as a tumour suppressor, PTEN mutations have been reported in a number of syndromes other than CS. For example, germline mutations in PTEN are associated with 50-60% of cases of Bannayan-Riley-Ruvalcaba (BRR) syndrome, a multiple hamartoma syndrome showing clinical overlap with CS.19 28-33 Somatic PTENmutations have been found at high frequencies in sporadic glioblastomas and endometrial and prostate cancers and at lower frequencies in cancers of the breast, kidney, thyroid, and colon.18 34-37 PTEN is now considered one of the most frequently mutated genes in human cancers and may therefore represent a critical pathway required for tumour suppression in a variety of tissues. In this study, we report a case of male breast cancer in a subject with a novel germlinePTEN mutation and associated Cowden features and further investigate a second Cowden family with male breast cancer.16 These observations suggest thatPTEN may play a role in tumour suppression in both male and female breast epithelium.

Materials and methods

Family 97-80 was ascertained through the Cancer Risk Clinic at the University of Chicago, USA, and family CDst was ascertained at the Kolling Institute of Medical Research, Royal North Shore Hospital, Sydney, Australia. All subjects tested signed informed consent approved by the respective Institutional Review Boards.

Genomic DNA was extracted from peripheral blood using standard protocols.38 Screening of thePTEN gene for point mutations, deletions, insertions, or small rearrangements was performed by denaturing gradient gel electrophoresis (DGGE) as previously described.39-42 PCR products showing an electrophoretic DGGE variant pattern were reamplified from the original genomic DNA, gel isolated, and purified using the Wizard PCR Preps DNA Purification System (Promega, Madison, WI). Purified PCR products were directly sequenced using the ABI Prism Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer, Norwalk, CT). The cycle sequenced products were electrophoresed on 6% Long Ranger gels (FMC Bioproducts, Rockland, ME) and analysed on an Applied Biosystems model 373A automated DNA sequencer (Perkin-Elmer).

Restriction digestion of PCR products was performed under the manufacturer's recommendations (New England Biolabs, Beverly, MA).


Family 97-80 is a 57 member, four generation kindred with a single affected subject, III.19, a 41 year old male, the youngest of a sibship of 11. He was diagnosed with CS at the age of 31 years when he presented with mucocutaneous papillomatosis, multiple facial trichilemmomas, and macrocephaly with frontal bossing (fig 1). Other phenotypic features included a high arched palate and hamartomas of multiple organs leading to severe bronchiectasis and digital clubbing. A diagnosis of a locally invasive infiltrating ductal carcinoma of the right breast was made at the age of 41 years. The tumour was nuclear and histological grade 2/3, oestrogen receptor positive (ER+), and with a low S phase fraction. Owing to his severe bronchiectasis requiring supplemental oxygen therapy, the patient was treated conservatively with a mastectomy and Tamoxifen. He died of widespread metastatic disease within two years of diagnosis.

Figure 1

Cowden syndrome in family member III.19. Note the mucocutaneous papillomatosis, facial trichilemmomas, and severe clubbing of the fingers.

A novel germline mutation in exon 7 of PTEN, c.802delG, predicted to affect splicing, was identified in patient III.19. This probably represents a de novo event because there were no stigmata of CS identified in any other family member (fig 2A). Unfortunately, both parents were dead and could not be tested; one of the proband's sisters, III.12, tested negative for this mutation, and other family members were unwilling to contribute a blood sample for research.

Figure 2

Pedigrees of families with CS and male breast cancer. Affected family members are represented by solid circles (females) or squares (males). Diamonds indicate gender unknown. Hatched circles indicate possible phenocopies. Half circles represent unconfirmed cancer. Numbers within symbols indicate multiple sibs, numbered as indicated above the symbols. +/- indicates that the subject is a carrier of a germline PTEN mutation. -/- indicates that the subject is wild type. Probands with CS are identified by an arrow. (A) Pedigree 97-80. The proband, III.19, was diagnosed with CS at 31 years and breast cancer at 41 years. He carries the germline PTEN c.802delG mutation. (B) Pedigree CDst. The proband, II.5, was diagnosed with breast cancer at 43 years and CS at 47. II.5 and II.4 were shown to be carriers of the PTEN germline mutation c.347-351delACAAT.

Family CDst is a 23 member, three generation kindred with three members, all sibs in the second generation, exhibiting multiple CS symptoms16 (fig 2B). Family member II.5 was diagnosed with breast cancer at the age of 43 years and thyroid adenoma at 47 years. This subject was also diagnosed with multiple trichilemmomas on the arms and legs and mildly dysplastic tubular adenoma of the bowel, but showed no signs of macrocephaly. His breast tumour was reported as invasive ductal carcinoma with none of 17 axillary lymph nodes showing any evidence of metastasis at the time of surgery. The patient subsequently developed hepatic metastases and died of breast cancer at the age of 57 years.

Patient II.4, a 63 year old female, was diagnosed with breast cancer, moderately differentiated adenocarcinoma of the proximal ascending colon, bowel polyps, possible neurofibromatosis, and a brain tumour. Unfortunately, we were unable to determine whether the neurofibromatosis in this patient represents a novel phenotype associated with CS or is associated with an unrelated mutation in a neurofibromatosis susceptibility gene. Neurofibromatosis is an autosomal dominant disorder that is completely penetrant (though variably expressive) and would be expected to occur multiple times within a family. However, there were no other reports of neurofibromatosis in this family. It is therefore intriguing to speculate whether the reported neurofibromatosis in this family is associated with CS as are other disorders of the nervous system in this syndrome.

II.2 was a 72 year old female with invasive ductal carcinoma of the breast, histological grade 2 and ER(+). Two of 18 axillary lymph nodes had shown metastasis. She also had endometrial carcinoma. A thyroidectomy had been performed at the age of 32 years from which no records were available. Patient II.3, a 65 year old female, was diagnosed with thyroid adenoma but has no other features of CS and is currently alive without cancer. Patient III.1 was diagnosed with breast cancer in her thirties and died of metastatic disease. Family member I.1 had thyroid cancer and died in her fifties of a cerebral haemorrhage.

Sequence analysis of the proband, II.5, showed that thePTEN mutation associated with CS in this family is c.347-351delACAAT.16 This mutation creates anNsiI restriction site that can be used for rapid detection in at risk family members using a PCR assay (fig 3). This deletion occurs in exon 5 upstream of the PTPase core motif, and is predicted to lead to premature truncation of the protein. Four additional family members were screened by PCR and restriction analysis: II.4 was found to share the same mutation as the proband, while II.3, III.14, and III.15 did not have this mutation. As expected, the PTEN mutation cosegregates with the CS phenotype. However, the thyroid adenoma in patient II.3 is most likely a sporadic tumour unrelated to CS as this family member does not carry the familial PTENmutation.

Figure 3

NsiI digestion of PTEN exon 5 PCR product of affected (II.4, II.5) and unaffected (II.3, III.14, III.15 ) subjects in pedigree CDst. The NsiI restriction site is created in those carrying the c.347-351delACAAT germline PTEN mutation.


It is currently estimated that 30-86% of inherited breast cancers in women are caused by germline mutations in highly penetrant susceptibility genes such as BRCA1 andBRCA2.43 Rarer genetic conditions, such as the Li-Fraumeni syndrome, Muir-Torre syndrome, Peutz- Jeghers syndrome, ataxia-telangiectasia, and Cowden syndrome also contribute to a small percentage of breast cancer in women. By contrast, it is not known what proportion of male breast cancer is caused by germline mutations in highly penetrant breast cancer susceptibility genes. Male breast cancer typically occurs at 0.5-2.4% the frequency of female breast cancer,44 depending upon geographical location. In the USA, approximately 1000 males are diagnosed with breast cancer each year.45 Outside the USA and Europe, Japan has the lowest incidence rate (0.6 cases of male breast cancer per 100 000 men),46 while some African nations have higher frequencies. In Zambia, for example, male breast cancer accounts for 15% of all breast cancer cases.47

The risk of male breast cancer can increase in cases of increased levels of oestrogen or decreased levels of androgen, such as in cases of testicular injury48 or cirrhosis of the liver, which is associated with increased levels of oestrogen.49 Men with Klinefelter's syndrome have a 14-50 fold increased risk of developing male breast cancer.50 Clinical presentations of male and female breast cancers are similar, but median age of onset in males is later than in females (60 v 53 years) and tumours in male breast cancers are generally of higher grade.51-53 All histological tumour types that have been characterised in females have also been seen in males with breast cancer. Although genetic markers in male breast cancer remain poorly characterised, a few studies have examined the contribution of markers such as HER2/neu andp53 . Amplification of theHER2/neu oncogene has been observed, but reports vary widely concerning its frequency and prognostic value.54-56

Germline mutations in the BRCA2 gene have been identified in men with breast cancer and a founder mutation,BRCA2999del5, has been reported to account for nearly 44% of male breast cancers in Iceland.57-58Overall, males with BRCA2 germline mutations have a 6.3% probability of developing breast cancer by the age of 70 years,59 whereas mutations in the other major female breast cancer susceptibility gene, BRCA1, are not thought to be associated with significant male breast cancer risk. More recently, one primary male breast cancer case was reported in a subject who also had colon cancer and an inheritedMLH1 mutation from a family with hereditary non-polyposis colorectal cancer (HNPCC) syndrome.60 This suggests that male breast cancer may also occur as an integral tumour in the HNPCC syndrome.

The data presented in this study suggest that germlinePTEN mutations contribute to the development of both male and female breast cancer within CS families. Given thatPTEN mutations were identified in the germline of the two male breast cancer patients described here, and because germline PTEN mutations are known to increase the risk of breast cancer in females, we believe that thePTEN mutation contributed to male breast cancer development in these CS patients. While we cannot formally exclude the possibility that the cases of male breast cancer in these CS families are phenocopies, we believe that this is highly unlikely. Reasons include the low frequency of male breast cancer in the general population, the young age at diagnoses in these two cases compared to the general population, the presence of other CS manifestations in these two men with breast cancer, and, at least in family CDst, segregation of the germline PTEN mutation in all affected subjects.

To our knowledge, there has been only one previous report of a possible male breast cancer case associated with aPTEN mutation in a CS family. This subject's family showed characteristic cases of skin and uterine abnormalities, along with myomatosis, goitre, and macrocephaly. It was not clear from this study whether the reported male breast involvement was malignant or benign.17 Thus, germline mutation ofBRCA2, rarelyBRCA1 and MLH1, and now PTEN would appear to be associated with the development of both male and female breast cancer. Both men in this report died of metastatic breast cancer. We believe that males with PTEN mutations may have an increased risk for breast cancer and clinicians taking care of CS patients should now become aware of this risk. Further studies to identify a putative role for PTEN mutations in a larger male breast cancer cohort are continuing.


We are very grateful to the patients who participated in this study. This study was partially supported by grants from the Department of Defence (DAMD17-99-1-9123), the Falk Medical Research Trust (to OIO), the NCRR (MOI RR00055) and CA14599-24 to the University of Chicago Clinical Research Center and Comprehensive Cancer Center, the American Cancer Society (RPG98-211-01CCE to CE), and the National Cancer Institute (P30CA16058 to The Ohio State University Comprehensive Cancer Center).


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.