A novel deleterious PTEN mutation in a patient with early-onset bilateral breast cancer

A young age at Breast Cancer (BC) onset may be a hallmark feature of inherited predisposition. Indeed, germ-line mutations in the two major BC genes BRCA1 and BRCA2 have been reported in 15 to 23% of younger Italian BC patients [1, 2], consistently with frequencies described in other western populations, which range from 6 to 23% [3‐5].

Although the majority of the remaining cases might be explained by a multifactorial etiology, mutations in rarer BC predisposing genes should also be considered. Rare, high penetrance BC genes include TP53, STK11, CDH1 and PTEN. Collectively, they are generally thought to account for less than 1% of inherited BC; nevertheless, an accurate clinical and family history evaluation may provide significant clues to the identification of patients carrying such uncommon mutations. Recognizing mutation carriers is crucial to plan targeted risk-management according to the specific gene, as is routinely done for BRCA mutation carriers.

The PTEN gene encodes a negative regulator of the PI3K/Akt/mTOR pathway and is one of the most frequently mutated genes in cancer, with loss of heterozygosity at the PTEN locus being reported in about 40% of invasive BC [6, 7]. Conversely, germ-line PTEN mutations are rare, and cause several syndromes with variable clinical manifestations that are collectively labeled PTEN Hamartoma Tumor Syndrome (PHTS). The prototypic syndrome, Cowden Syndrome (CS), is featured by macrocephaly, gastrointestinal lesions and cerebellar gangliocytoma, as well as by benign and malignant tumors of the thyroid, the endometrium and the breast, with a lifetime risk of BC estimated to be as high as 85.2% [8]. Nevertheless, the most typical features are specific mucocutaneous lesions, which include trichilemmomas, acral keratoses and oral papillomatous papules, and occur in 90-100% of cases.

Women harboring a predisposing gene mutation face a high risk to develop BC at a young age and to experience multiple primary BCs. These risks have been extensively studied in women with BRCA1/2 mutations, whose average age at BC onset is around 45 years, while the risk of contralateral BC 25 years after first BC is 47.4%, according to recent figures [11].

For rarer conditions, such as CS, these risks and the relative management are not so clearly established. Until recently, lifetime BC risk in CS female patients was largely accepted to be around 25-50%; however, recent reports on two large series raised the estimated lifetime risk to 77-85% [8, 12]. Consistently, a pooled analysis of literature data and records from the Mayo Clinic provided an estimate of the cumulative BC risk as high as 81% [13]. The age at BC onset is generally believed to be young (38–50 years) [14]. Nevertheless, data on the risk for bilateral BC are scarce; in the French series, out of 23 BC cases, 11 (48%) were reported as bilateral [12], while in the study by Riegert-Johnson and colleagues 34% of patients diagnosed with BC cancer had bilateral disease [13].

Missense variants of uncertain significance are a relatively common finding that complicates the interpretation of gene test results. In our patient, we found the novel missense mutation p.Asp24Val, which had not been reported before. Anyway, different mutations at the same codon had been described: the germ-line p.Asp24Tyr mutation, in a Bannayan-Riley-Ruvalcaba patient [15], and the p.Asp24Gly, detected in the germline of a CS patient [16] as well as in sporadic tumors of the endometrium and of the central nervous system as a somatic mutation (COSM5170, http://​www.​sanger.​ac.​uk/​genetics/​CGP/​cosmic/​), supporting the functional relevance of the highly conserved Asp24. These data, together with the in silico prediction, strongly suggested a functional role of this variant, which was confirmed by demonstrating in vitro that the mutant PTEN, unlike the wild-type protein, was unable to inhibit the PI3K/Akt/mTOR pathway.

Whether PTEN acts as a classical tumor suppressor gene following Knudson’s two-hit hypothesis is still controversial: on one hand, in a mouse model of prostate cancer, the complete loss of PTEN, unlike its dose-reduction, was demonstrated to induce senescence instead of cancer, unless the loss of TP53 co-occurred [17]. On the other hand, PTEN immunostaining was proven to be negative in 13 out of 15 BC samples from CS patients, suggesting loss or inactivation of the wild-type PTEN allele in the tumor [18]. To explain its heterogeneous behavior, PTEN has been recently appointed as a haploinsufficient gene, characterized, however, by tissue specificity and context dependency [19].

In the present case, the mutation was demonstrated to be at the heterozygous state in both normal and tumor tissue and no additional PTEN sequence mutations were detected in the tumor, which is in line with what we found in non-breast tumors from other CS patients, where loss or mutations of the wild-type PTEN allele were excluded. In such cases, however, somatic mutations of different genes were found, with an apparent correlation between the specific gene and the type of tumor developed [20, 21]. In the present case, the markedly increased Akt phosphorylation detected in BC, if compared to normal tissue, led us to infer that a somatic event had cooperated in deregulating the PI3K/Akt/mTOR pathway and, presumably, in inducing carcinogenesis. Hence, activating mutations commonly found to impair PI3K/Akt/mTOR regulation in cancer were analyzed in the breast tumor, with none of them being detected. Such findings suggest the involvement of a different, uncommon somatic event, such as the possible epigenetic inactivation of the wild-type PTEN allele (which could not be ruled out) or an unusual mutation in genes involved in the PI3K/Akt/mTOR pathway, thus demonstrating once again how carcinogenesis in PTEN mutation carriers is a complex, still elusive process, which requires major efforts to be elucidated.