Neuroendocrine Carcinoma of the Breast: Current Evidence and Future Perspectives

This review discusses the current evidence and the future perspectives on diagnosis and treatment of neuroendocrine carcinoma of the breast. A multimodality therapeutic strategy can be considered on an individual basis, with chemotherapy, endocrine therapy, peptide receptor radionuclide therapy, radiation therapy, surgery, or a combination. A better knowledge of the biology of these tumors will hopefully provide new therapeutic targets for personalized treatment.

Abstract

Neuroendocrine carcinoma of the breast is considered a rare entity, and for this reason there are no data from prospective clinical trials on its optimal management. Early stage tumors are usually treated with the same strategy used for the other types of invasive breast cancer. Anthracycline- and taxane-based regimens represent the most frequently administered chemotherapy in neoadjuvant and adjuvant setting, as well as for metastatic disease, although combinations of platinum compounds and etoposide have been widely used, in particular for small-cell histology and tumors with a high proliferation index. For metastatic disease, a multimodality therapeutic strategy can be considered on an individual basis, with chemotherapy, endocrine therapy, peptide receptor radionuclide therapy, radiation therapy, surgery, or a combination of the above. In the near future, a better knowledge of the biology of these tumors will hopefully provide new therapeutic targets for personalized treatment. In this review, we discuss the current evidence and the future perspectives on diagnosis and treatment of neuroendocrine carcinoma of the breast.

Implications for Practice:

Neuroendocrine carcinoma of the breast (NECB) is a distinct entity of breast cancer. Clinical features and morphology are not helpful to distinguish NECB from other subtypes of breast cancer; therefore, immunohistochemistry markers for neuroendocrine differentiation, mainly chromogranin and synaptophysin, should be routinely used to confirm the diagnosis, especially in cases of mucinous or solid papillary carcinoma in which the suspicion of NECB may be relevant. Adjuvant treatment should be offered according to the same recommendations given for the other types of invasive breast cancer. An accurate diagnosis of NECB is also important in the metastatic setting, in which a multimodality approach including specific therapies such as peptide receptor radionuclide therapy can be considered.

Introduction

Primary neuroendocrine carcinoma of the breast (NECB) includes a heterogeneous group of tumors with different biologic behavior and prognosis [1]. Their incidence has been reported to range from less than 1%–5% of breast cancers [2, 3], according to the different series and the different diagnostic criteria. However, some authors reported a neuroendocrine differentiation in up to 20% of breast carcinomas [4]. In fact, the true incidence of the disease is difficult to assess, because immunohistochemistry neuroendocrine markers are not routinely used in breast tumors [5]. In a recently published retrospective analysis performed by our research group, 1,232 cases of breast cancer were assessed with immunohistochemistry staining with synaptophysin and chromogranin A, and we found neuroendocrine differentiation in 10.4% of tumors [6]. The prognostic relevance of neuroendocrine differentiation of breast tumors is still debated. Because of its low incidence, there is also a scarcity of evidence regarding the optimal management of NECB, with the only available data resulting from case reports and series. In this review, we discuss the classification criteria, pathogenesis, diagnostic work-up, prognostic factors, and treatment of neuroendocrine tumors of the breast, also providing a hint on future perspectives based on a growing knowledge of tumor biology.

Materials and Methods

The currently available evidence on the diagnosis and treatment of NECB comes from retrospective case series or case reports. For this article, the MEDLINE database was searched for papers published before April 2015 using “neuroendocrine carcinoma” OR “neuroendocrine tumors” AND “breast” as search terms.

Pathological Classification

NECB was originally reported by Feyrter and Hartmann [7] in 1963 as an invasive breast cancer morphologically similar to intestinal carcinoids. In 1977, Cubilla and Woodruff [8] described eight other cases of breast cancer with a carcinoid growth pattern, providing for the first time a histopathological classification together with a clinical and prognostic analysis of NECB. It was only in 2003, however, that the World Health Organization (WHO) recognized neuroendocrine tumors of the breast as a separate entity of breast cancer, based on the definition provided by Sapino et al. [9]. Thus, neuroendocrine tumors of the breast were defined as tumors of epithelial origin, with morphology similar to gastrointestinal and pulmonary neuroendocrine tumors, expressing a neuroendocrine marker in at least 50% of the total cell population [2]. Chromogranin and synaptophysin show the best sensitivity and specificity as immunohistochemical neuroendocrine markers [10]. Also neuron-specific enolase can be occasionally positive in NECB [11], whereas CD56 and other immunohistochemical markers seem to be less sensitive and specific [12].

In 2012, the WHO acknowledged that the 50% threshold of cells with neuroendocrine markers expression was arbitrary; therefore in the new classification, invasive carcinomas with neuroendocrine differentiation were included in the group of NECB regardless of the percentage of tumor cells expressing neuroendocrine markers [5]. According to the 2012 WHO classification, based on morphology, breast tumors with neuroendocrine features are divided into three groups: (a) neuroendocrine tumor, well-differentiated (carcinoid-like); (b) neuroendocrine carcinoma, poorly differentiated/small-cell carcinoma; and (c) invasive carcinoma with neuroendocrine differentiation. Well-differentiated neuroendocrine tumors are usually low-intermediate nuclear grade invasive carcinomas composed of spindle and plasmocytoid cells with occasionally clear cell features and resemble carcinoid tumors arising in the gastrointestinal tract and lung. Poorly differentiated/small-cell neuroendocrine carcinomas are morphologically identical to small-cell lung cancer, with a high nuclear/cytoplasmic ratio, dense chromatin, a brisk mitotic rate, and areas of necrosis. The third group comprises several morphologic subtypes of invasive breast carcinoma with neuroendocrine differentiation, mainly represented by the hypercellular variant of mucinous carcinoma and the invasive form of solid papillary carcinoma [13].

Histogenesis

The histogenesis of NECB is not clearly defined yet. One theory suggests the development of NECB from neuroendocrine cells constitutively present in the breast, but these cells have not been consistently found in normal breast tissue [14]. Furthermore, neuroendocrine cell hyperplasia or in situ neuroendocrine carcinoma that would support a progression from noninvasive to invasive neuroendocrine tumors have been rarely reported in the breast [15]. According to another theory, NECB would instead result from an early divergent differentiation of breast cancer stem cells into both neuroendocrine and epithelial lines [4]. This hypothesis is consistent with the observation that NECB is always a mixed tumor including endocrine and exocrine cells, and it is also consistent with molecular studies showing that in NECB the neuroendocrine cells are clonally related to the intraductal component of the tumor [16].

NECB is often positive for hormone receptors, whereas HER2 is almost always negative. Accordingly, gene expression profile analysis has revealed that NECB belongs to the luminal subtype [17]. A luminal B phenotype, defined by immunohistochemistry as hormone receptor-positive tumor with high proliferation index (i.e., Ki67 > 14%), accounts for approximately 50% of NECB [6]. However, there are also reports of HER2-positive NECB [18] and small-cell carcinoma of the breast with basal-like characteristics [19].

Clinical Presentation and Diagnostic Work-up

Primary NECB has no clinical features distinct from other types of breast cancer. It may present as an isolated hard breast lump with or without axillary lymphadenopathy. Sporadically, NECB may lead to clinical conditions related to hormonal hypersecretion, because of ectopic production of adrenocorticotropic hormone [20], norepinephrine [21], or calcitonin [22]. The disease is more commonly diagnosed in elderly women in their sixth or seventh decade of life [3]. Only a few cases of NECB have been reported in males [3, 23, 24].

The radiological features of primary NECB are nonspecific. However, some authors reported that NECB often appears on mammography as a round, sharply circumscribed, hyperdense mass and on breast ultrasound as a hypoechoic solid mass with ill-defined margins, increased vascularity, enhanced posterior echo, and a cystic component. Magnetic resonance imaging shows homogeneous low signal intensity with heterogeneous rapid initial enhancement on the T1-weighted image [25].

Fine-needle aspiration cytology may not be adequate for the diagnosis of NECB because the cytological features of such tumors are similar to those of invasive ductal carcinoma and intraductal papilloma. Therefore, definitive diagnosis is made with core needle biopsy or with surgical specimen [26, 27]. Differential diagnoses include Merkel cell carcinoma, lymphoma, melanoma, and, above all, neuroendocrine tumors metastatic to the breast. The presence of an associated ductal carcinoma in situ component is a valid proof of the primary nature of the tumor [28]. However, when a NECB is diagnosed, a breast metastasis from other primary sites should be ruled out by appropriate imaging, such as a chest and abdomen computed tomography (CT) scan. Somatostatin receptor scintigraphy (SRS) or positron emission tomography (PET)-CT with ⁶⁸Gallium-labeled somatostatin analogs also may be useful to exclude a different primary site in the case of well-differentiated neuroendocrine carcinomas, whereas 18-fluorodeoxyglucose PET-CT could be used with the same purpose in the case of poorly differentiated neuroendocrine carcinoma with a high proliferation rate [29]. When the differentiation between primary and secondary neuroendocrine tumor remains enigmatic, some practical criteria have been suggested for differential diagnosis. In particular, a larger tumor, the absence of in situ component, negativity of estrogen and progesterone receptors, and the absence of axillary nodal metastasis are highly suggestive of a secondary rather than a primary lesion [30].

Prognosis

The prognostic relevance of neuroendocrine differentiation in breast carcinoma is still debated because several studies have been published with mixed results [3, 6, 31–35]. These conflicting results might be explained by the limited number of cases reported in each series, by the different inclusion criteria based on whether the 2003 or 2012 WHO definitions were applied to identify NECB, and also by the analysis performed considering NECB as a whole, without analyzing the outcomes according to the different histologic subtypes.

In a population-based study from the surveillance, epidemiology, and end results (SEER) database, from 2003 to 2009, 142 cases of NECB were identified according to the 2003 WHO criteria of the 50% threshold of tumor cells positivity for neuroendocrine markers [3]. In this large study, patients with NECB had a shorter overall survival (OS) and disease-specific survival compared with invasive breast carcinoma not otherwise specified at the same stage. Moreover, in a multivariate analysis including other known prognostic factors such as age, tumor size, nodal status, histologic grade, hormone receptor status, and therapy, neuroendocrine differentiation was an independent adverse prognostic factor [3].

In a retrospective analysis of 55 early NECB compared with 115 matched non-NECB control cases, disease-free survival (DFS) but not cancer-specific survival was significantly worse for patients with NECB [6]. In patients with NECB, larger tumor size (>20 mm), higher tumor stage, negative hormone receptor status, and a lack of surgical treatment were associated with shorter OS [3]. Also the proliferation index and histologic subgroup represent relevant prognostic factors in NECB. In fact, in a large series of Chinese patients with NECB, Ki67 > 14% was significantly associated with worse DFS and OS [36]. When the specific histologic subgroup was taken into account, small-cell carcinoma was significantly associated with worse prognosis in comparison with all the other neuroendocrine tumors of the breast [1].

When the specific histologic subgroup was taken into account, small-cell carcinoma was significantly associated with worse prognosis in comparison with all the other neuroendocrine tumors of the breast.

Treatment

Surgery is the mainstay of the treatment for early NECB, and the choice of surgical procedure depends on the location of the tumor and on the clinical stage [37]. There are no specific studies with adjuvant radiation therapy for NECB, but radiotherapy should be considered according to the recommendations given for the other types of invasive breast cancer.

The choice of adjuvant systemic therapy should be made on individual basis, considering the biological features of the disease and the risk of recurrence. Tumor size and nodal metastases are the main prognostic factors for evaluating risk of relapse for NECB, as for other types of breast cancer [3, 38]. Chemotherapy can be used as adjuvant therapy in patients with a high risk of relapse or as neoadjuvant therapy in cases of locally advanced or inoperable NECB. Patients with hormone receptor-positive NECB are candidates to receive adjuvant endocrine therapy [27]. Given the negative prognostic role of proliferation index, patients with hormone receptor-positive and high Ki67 may also benefit from adjuvant chemotherapy in addition to endocrine therapy [39]. Several chemotherapy regimens have been reported in literature, including anthracyclines and/or taxanes containing combinations routinely used for other types of breast cancer and also combinations of platinum agents and etoposide, usually administered for small-cell lung cancer. Some examples of the chemotherapy regimens reported in the literature include fluorouracil/epirubicin/cyclophosphamide followed by docetaxel; docetaxel/epirubicin/cyclophosphamide; cyclophosphamide and doxorubicin; cyclophosphamide/methotrexate/fluorouracil; paclitaxel alone; carboplatin/paclitaxel; carboplatin or cisplatin and etoposide; and cisplatin and irinotecan [27, 40]. In a small series, the type of adjuvant chemotherapy was decided on the bases of the Ki67 index, according to a modified therapeutic algorithm initially developed for gastrointestinal neuroendocrine tumors [41]. In this series, anthracycline-based chemotherapy was given to patients with Ki67 < 15%, and cisplatin/etoposide to patients with Ki67 ≥ 15% [29]. However, in the absence of robust data on the role of platinum compounds and etoposide for the adjuvant treatment of NECB, we suggest treating these tumors with the same principles adopted for ductal breast carcinoma; therefore if chemotherapy is indicated, a regimen including anthracyclines and/or taxanes should be preferred.

As reported above, HER2 is only sporadically expressed by NECB [6, 18, 36]. The prognostic role of HER2 in NECB is not clear, but it is conceivable that it is analog to that observed for other invasive breast carcinomas; therefore anti-HER2 therapy should be considered for HER2-positive NECB.

NECB can metastasize to multiple sites even years after the treatment for primary tumor, and therefore a long-term follow up is advisable. Metastatic sites include liver, bones, lung, pancreas, soft tissues, and brain [42]. When metastatic disease is suspected, especially in the case of metachronous metastases developed after a long latency period from the treatment of the primary tumor, a biopsy of the metastatic site would be useful to confirm the diagnosis and also to analyze the molecular characteristics of the disease, including immunohistochemistry markers of neuroendocrine differentiation, HER2, and hormone receptor status, given the possible heterogeneity between primary breast cancer and metastases [43].

When metastatic disease is diagnosed, the treatment strategy should be chosen on an individual basis, taking into account the general conditions and comorbidities of the patient, the extent and the aggressiveness of the disease, and the biological features of the tumor. Surgical resection may represent a valid option for liver metastases in selected cases, because prolonged survival has been reported after hepatectomy [42, 44]. Liver surgery should be pursued if a R0 resection is feasible and when there is no evidence of extrahepatic disease, except for bone metastases, which can be controlled by radiotherapy. An alternative to surgery for oligometastatic disease can be represented by stereotactic radiotherapy, as for other breast carcinomas [45].

When metastatic disease is located in multiple sites, the option of choice should be systemic treatment. In case of hormone receptor positivity and indolent disease with skeletal or lymphnodal metastases, some authors have administered endocrine therapy with long-lasting disease control [46, 47]. For symptomatic, aggressive disease and in cases of visceral metastases, chemotherapy should be considered. As for the adjuvant setting, also for metastatic disease, the chemotherapy regimens that have been reported in the literature are the regimens usually given for other types of metastatic breast cancer and those given for small-cell lung cancer.

For NECB expressing somatostatin receptors, peptide receptor radionuclide therapy (PPRT) has been reported to be an active and well-tolerated approach. In patients with a positive SRS or ⁶⁸Gallium PET-CT, PPRT may be used after failure of conventional chemotherapy or also as the first or second line within an integrated therapeutic strategy [48].

Future Perspectives

Because of its rarity, data on molecular characteristics of NECB are limited. A better knowledge of the molecular profile of NECB could help to identify novel targets for a tailored treatment.

In a recent study, primary or metastatic NECB were screened for mutations in common cancer genes [49]. Mutations were found in 5 of 15 tumors. PIK3CA mutations were identified in 3 tumors, 2 of which also harbored mutations in fibroblast growth factor receptor (FGFR) family members. This is an interesting finding, because phosphatidylinositol 3-kinase (PI3K) inhibitors and FGFR inhibitors are currently in development [50, 51]. In the study of Ang et al., an activating mutation of vascular endothelial growth factor receptor 2 (VEGFR2), was found [49], and in another study using immunohistochemistry, VEGF-C was reported to be strongly expressed in NECB [52]. Taken together, the activating mutation of VEGFR-2 and the high level of expression of VEGF-C detected in some cases of NECB may provide a rationale for the investigation of antiangiogenic agents in this disease.

Taken together, the activating mutation of VEGFR-2 and the high level of expression of VEGF-C detected in some cases of NECB may provide a rationale for the investigation of antiangiogenic agents in this disease.

The PI3K/AKT/mammalian target of rapamycin (mTOR) pathway plays a role in the development and progression of pancreatic neuroendocrine tumors (pNETs) [53]. Everolimus, a mTOR inhibitor, has demonstrated efficacy in well and moderately differentiated pNETs [54], and it is currently investigated in other types of neuroendocrine tumors. Interestingly, the PI3K/AKT/mTOR pathway seems implicated also in acquired resistance to hormone therapy in estrogen receptor (ER)-positive breast cancer [55]. In fact, in a phase 3 study, the addition of everolimus to the aromatase inhibitor exemestane, significantly prolonged progression-free survival in metastastic ER-positive, HER2-negative breast cancer patients who had had recurred or progressed while receiving previous therapy with a nonsteroidal aromatase inhibitor [56, 57]. Targeting mTOR in metastatic NECB may represent an intriguing therapeutic strategy that deserves specific investigation.

This article is available for continuing medical education credit at CME.TheOncologist.com.

Author Contributions

Conception/Design: Alessandro Inno, Giuseppe Bogina, Monica Turazza, Laura Bortesi, Simona Duranti, Alberto Massocco, Giuseppe Zamboni, Giovanni Carbognin, Filippo Alongi, Matteo Salgarello, Stefania Gori

Collection and/or assembly of data: Alessandro Inno, Stefania Gori

Manuscript writing: Alessandro Inno, Stefania Gori

Final approval of manuscript: Alessandro Inno, Giuseppe Bogina, Monica Turazza, Laura Bortesi, Simona Duranti, Alberto Massocco, Giuseppe Zamboni, Giovanni Carbognin, Filippo Alongi, Matteo Salgarello, Stefania Gori

Disclosures

Filippo Alongi: Astellas (H). The other authors indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board