Medical Oncology

, Volume 29, Issue 4, pp 2613–2618

Primary neuroendocrine breast cancer, how much do we know so far?

Authors

    • Staten Island University Hospital
  • Kassem Harris
    • Staten Island University Hospital
  • Basem Azab
    • Staten Island University Hospital
  • Qun Dai
    • Staten Island University Hospital
Review Article

DOI: 10.1007/s12032-012-0222-z

Cite this article as:
Alkaied, H., Harris, K., Azab, B. et al. Med Oncol (2012) 29: 2613. doi:10.1007/s12032-012-0222-z

Abstract

Primary neuroendocrine cancer of the breast (NECB) is an extremely rare tumor. In 2003, the World Health Organization (WHO) recognized this category with three well-described subtypes: small cell, large cell, and carcinoid-like carcinoma; very few peer-review publications based on the WHO definition were encountered in the literature, and we conducted a literature search to investigate the reported incidence, diagnosis, prognosis, hormone receptor status, and treatment options for this rare tumor. Confirming the breast as an origin of neuroendocrine tumor represents a challenge. The diagnosis is mainly dependent on the exclusion of other extra-mammary organs based on clinical, radiological, and pathological data. Except for the very rare type small cell carcinoma, estrogen and progesterone receptors were reported to be expressed in 90 and 83 % of NECB, respectively. It is hypothesized that primary breast neuroendocrine carcinoma differentiates from the epithelial cells during the carcinogenesis process; the prognosis of non-small cell primary NECB seems to improve as the amount of mucinous component increases in the tumor specimen. Management similar to interventions utilized to manage the usual ductal-type carcinoma has been attempted in the past, such as chemotherapy and hormonal therapy; however, due to the rarity of the tumor, none of the published studies are randomized nor do they have a large number of patients. Additionally, none of reports analyzed NECB based on its distinct subtypes. These limitations make recommendations largely based on anecdotal and small observatory studies and call for the need for further research in this extremely rare tumor.

Keywords

Breast cancerNeuroendocrine cancerTreatmentPathologyPrognosisManagement

Introduction

Primary neuroendocrine cancer of the breast (NECB) was initially described in 1977 by Cubilla and Woodruff [1]. Later in 1982, Azzopardi described “Argyrophilic carcinoma” and its variant [2]. In 1989, Papotti et al. proposed seven histologic subcategories of primary NEBC [3]; thereafter, Spaino et al. [4] described five subtypes of NECB. In 2002, Sapino et al. [5] suggested that tumors that express any of the neuroendocrine markers such as chromogranin A, chromogranin B, and synaptophysin in more than 50 % of their cells can be classified as primary neuroendocrine breast cancer. A year later in 2003, the WHO adopted Sapino et al.’s definition to endorse primary NECBs as a separate unique category [6].

The prevalence of breast carcinoma with neuroendocrine (NE) differentiation was reported according to the methods used to detect NE markers, and it was found that the prevalence ranges from 12 to 19.5 % of cases based on different selection criteria, the source of analyzed tissue (whole section vs tissue microarray), and NE markers found [79]. However, based on the WHO definition, the true incidence of primary NECB is estimated to range between 0.3 and 0.5 % according to the reported review of 1368 and 1,845 breast cancer specimens, respectively [10, 11].

We conducted a literature review of primary NECB that meet the WHO definition of primary NECB published in 2003 and evaluated the incidence, diagnosis, pathology, hormone receptor status, risk factors, diagnosis, prognosis, and treatment options.

Seven peer-review publications were found that met our search criteria [7, 10, 1216].

Pathology

Morphology alone is not sufficient to confirm the diagnosis of primary NECB; NE expression by immunohistochemistry (IHC) represents the gold standard fashion to diagnose primary NECB [17]; chromogranin A (CGA), chromogranin B (CGB), and synaptophysin (SYP) are considered the most sensitive and specific NE markers for this condition [18]. These markers are related to the observed dense core granules (CGA and CGB) and the presynaptic clear vesicles (SYP) [19]. Less specific markers include neurospecific enolase (NSE) [7], CD56 [10], neurofilament triplprotein, and bombin or leu [7, 20]. The expression of NE markers is not consistent in the very rare small cell carcinoma subtype [21, 22]; in that condition, morphological behavior alone can be relied upon to confirm the diagnosis.

Histologic subtypes

In 1989, Papotti et al. [3] proposed seven histologic subcategories of primary NEBC, cohesive mucoid, mixed, trabecular, atypical, carcinoid-like, lobular, and small cell carcinoma. In 2000, Sapino et al. [4] described five subtypes of NECB, solid cohesive, alveolar, small cell, papillary and cellular mucinous carcinoma. In 2003, the WHO defines 3 separate histologic categories of NECB: solid carcinoid-like carcinoma, which is well-differentiated, large cell-, and small cell-type carcinoma, which are poorly differentiated.

In 2010, Righi et al. [16] proposed 2 additional mucinous subtypes of NECB in addition to those recognized by the WHO, the solid papillary carcinoma that produces solid sheets of cells with variable amount of extracellular mucin production, and the cellular mucinous carcinoma where islands of tumor cells float in mucinous formed lakes. These 2 additional subtypes are distinguished by an expansive growth, an associated in situ ductal carcinoma component, specific cytological characteristics (plasmacytoid–signet ring cells), and mostly a low-histologic grade.

The most common subtype of diagnosed cases of NECB are of solid cohesive type that is similar to the well-differentiated NE tumor found somewhere else, followed by the solid papillary and mucin-producing subtype that is associated with both endocrine and mucinous differentiation and is linked more frequently to ductal carcinoma in situ (DCIS). This subtype is not considered by itself a hallmark of specific category by the WHO, but the WHO acknowledges that mucinous differentiation occurs in up to 26 % on NEBC. This dual (NE and mucin) divergent differentiation in the same tumor and even in the same cell is a unique aspect of primary NECB [10] that rarely takes place in other locations except the appendix [23]. Additionally, mucin-producing NECB may also co-express apocrine markers [24]. The solid and solid papillary pathological subtypes are the most common subtypes seen in NECB. In the most recent update by Luisella et al. among 89 diagnosed NEBC cases, 55 cases were reported to have either solid or solid papillary subtypes [16]. The least common form is the small cell carcinoma subtype that is poorly differentiated and usually seen in less than 13 % of diagnosed cases [16]. Unlike NE tumors found in other organs, benign NE tumor is yet to be discovered in the breast.

Tumor nuclear grade

Most primary NEBC are of nuclear grade 2 except the poorly differentiated small cell carcinoma (SmCC) that usually has a higher tumour nuclear grade [3, 16] (Table 1).
Table 1

Studies summarizing nuclear grade and hormone receptor status among reported studies of primary breast neuroendocrine carcinoma

Studies

Patients

NG 1 exclude SmCC

NG 2 exclude SmCC

ER

PR

HER 2

Righi et al. [16]

78

22

42

67

0

Zekioglu et al. [15]

12

11

11

1

Makretsov et al. [7]

10

2

7

7

7

1

Rovera et al. [14]

13

13

13

0

Lopez-Bonet et al. [10]

7

0

7

7

7

1

Richter-Ehrenstein et al. [13]

9

8

7

1

Wei et al. [12]

74

2

57

70

59

2

Total number

203

26

113

183

104

6

%

 

15.3

66.8

90

83

3

NG nuclear grade, SmCC small cell carcinoma, ER estrogen receptors, PR progesterone receptors

Hormone receptors status

More than 90 % of diagnosed primary NECB cases can express estrogen receptors (ER) while up to 75–80 % express progesterone receptors (PR) [12, 14]. Hormone receptor expression is less frequent in the small cell carcinoma type [16]. Up to 48 % of NECB cases can express androgen receptor, which is more commonly co-expressed with the apocrine marker GCDFP particularly in the more differentiated tumor type that is more commonly seen in elderly women [24].

Her 2 expression is nonexistent [25] or very rarely present in NECB [15]. The summary of published reports regarding NECB excluding SmCC subtype showed that the incidence of ER and PR expression in NCCB is 90 and 83 %, respectively (Table 1).

Age and menopausal status at diagnosis

Upalakalin et al. [26] reported that the average age at diagnosis of NECB is usually 10 years later than the usual type breast cancer; hence, the vast majority of cases are encountered in postmenopausal women [12].

Imaging characteristics

Primary NECB bares no unique finding on mammogram or MRI of the breast when compared with the primary ductal-type breast cancer [27, 28].

Diagnosis

A core biopsy or fine-needle aspiration can confirm the diagnosis of mammary neuroendocrine carcinoma [29]; an important challenge remains to be confirmed that the breast represents the source of such a rare tumor. When investigated usually, the coexisting of intra-ductal component [30] and the absence of other primary sites establish the breast as an organ of carcinoma origin. The presence of hormone receptor expression cannot be suggested to confirm that the breast represents the source of such a tumor since ER and PR expressions are found in tumors originating from other organs such as the lung. In one report, 42 % of tumor specimens expressed primary pulmonary NE tumors, 40 % of specimens expressed ER, and 70 % of which expressed PR [31]. In these instances, the expression of either GCDFP-15 or mammogram is more specific to mammary tissue origin and in most cases confirms the diagnosis of primary neuroendocrine breast carcinoma [13]. GCDFP-15 was also suggested as a blood marker of breast cancer micro-metastasis when skin contamination is excluded [32]. In small cell carcinoma subtype, it is important to exclude primary lung or Merckle cell cancer. In these instances, IHC stain may be helpful, usually primary breast NECB stains are positive for CK7 and negative for CK20 [33], while both stains are negative in SmCC of the lung. On the other hand, Merckle cell carcinoma will usually stain negative for CK7 and positive for CK [20, 34]. In summary, the combination of clinical data, radiological studies such as PET/CT [13], and pathological review by IHC stain will provide in most cases a sufficient evidence to define the source of NE carcinoma.

Origin hypothesis

In 1994, Maluf et al. [17] proposed that NECB do not originate from preexisting NE cells but rather from a dual differentiation of neoplastic precursor stem cells along both epithelial and endocrine lines; Eyden et al. [35] reported a common stem mammary cell origin due to the existence of double exocrine–endocrine differentiation within the tumor cells. Miremadi et al. [8] also suggested that since hyperplastic lesions or benign neuroendocrine tumors were never found in the breast, breast NE carcinoma differentiates from the epithelial cells during the carcinogenesis process. Most recently, gene expression profiling analysis has led to the evolution of a working model for a breast cancer molecular taxonomy consisting of 5 molecular subtypes derived from ductal carcinoma (i.e., luminal A, luminal B, basal like, HER2, and normal breast like) [36]. Weigelt et al. [37] found that NECB belongs to the luminal subtype according to the gene expression analysis.

Prognosis/clinical evolution

Focal breast NE differentiation bears no significance in terms of patient’s outcome [8], unless it is seen in at least 50 % of tumor cells. Even then, reports have conflicting outcomes associated with these rare tumors. Miremadi et al. [8] reported that primary breast cancers with neuroendocrine markers expression do not differ from the non-NE counterpart in terms of mean age, size, histologic grade, nodal status, and prognosis. Similar reports had suggested that the prognosis of primary NECB does not differ from the conventional non-NECB [9]. Rovera et al. [14] found that the overall survival of patients affected by primary breast NE carcinoma of non-small cell type is better than that of patients affected by infiltrating ductal or lobular carcinoma in 13 studied cases. On the other hand, most recently, Tian et al. [38] and Wei et al. [12] reported a worse outcome of primary NECB in comparison with the usual ductal breast cancer not-otherwise classified (NOS) after 74 cases of primary NECB were retrieved for analysis. However, their reports contained only 3 NECB of cellular mucinous carcinomas subtype (Table 2) that has been reported to correlate with a better prognosis [24, 39]. Sapino et al. [24] found that patients with solid papillary carcinomas (producing some degree of mucin) or mucinous carcinoma had a significantly longer survival time than did patients with tumors of the other histologic subtypes. In addition to mucin production, breast cancer’s apocrine differentiation was found to correlate with a better prognosis [40, 41].
Table 2

Management of primary neuroendocrine breast carcinoma according to the published studies

Studies

 

Mastectomy

Lumpectomy

LN

Metastatic disease

Prognosis of NEBCs versus usual type IDC

Righi et al. [16]

78

26

4

Same

Zekioglu et al. [15]

12

6

6

1

Better

Makretsov et al. [7]

10

3

Same

Rovera et al. [14]

13

2

Better

Lopez-Bonet et al. [10]

7

5

2

3

Better

Wei et al. [12]

74

30

37

41

6

Worse

LN lymph nodes, NECB neuroendocrine carcinoma of the breast, IDC invasive ductal carcinoma

In terms of histologic grade, Sapino et al. [24] reported in his study that all patients diagnosed with grade 1 tumors were alive at 13 years of follow-up, while patients with the grade 3 NE tumors had a bad prognosis, and all of them died within 6 years of diagnosis. In Zhen et al.’s [38] study, histologic grade was associated with a trend in predicting distant recurrence-free survival (DRFS) in the univariate analysis but had no prognostic significance for OS. In the same analysis, large tumor size, high nuclear grade, and the presence of regional lymph node metastasis were found to have an adverse prognostic factors for both OS and DRFS. However, with multivariate analysis lymph node status and Ki67 stood as risk factors for OS [38]. Zhen et al. also found that patients with NECB showed a trend toward better OS in the ER-positive cases. In the study by Sapino et al. [24], both ER and PR expressions correlated significantly with survival in the univariate analyses.

Primary breast SmCC subtype is considered to be as aggressive as its pulmonary counterpart. Its prognosis and clinical outcome were reported to be very poor [33, 42]; however, in a study by Shin et al. [22], the prognosis of 9 cases of primary SMCC subtype was found to be more favorable in comparison with SmCC of the lung.

Staging and lymph node status at the time of diagnosis and its impact on OS/DRFS

Righi et al. [16] reported on 89 cases of primary breast NEBCs, 78 of which were of non-SmCC subtype and 11 cases were of SmCC subtype. Among patients diagnosed with non-SmCC histology, 26 cases (33 %) and 4 (5 %) cases had lymph nodes (LN) involvement and distal metastasis, respectively. On the other hand, 60 % and 27 % of 11 cases of SmCC subtype had LN involvement and distal metastasis respectively, indicating more frequent LN involvement and distal metastasis in SmCC than non-SmCC neuroendocrine breast carcinomas at the time of diagnosis.

In terms of staging and its impact on DRFS and OS, in the largest study of Zhen et al. [38], the 5-year OS was 100, 75, and 80 % in stage I (TNM), stage II, and stage III and IV of the disease, respectively. The 5-year DRFS rate was 86, 60, and 47 % in stage I, stage II, and stage III and IV of the disease, respectively. Stages III and IV patients were found to have a significantly poorer DRFS rate than patients with stage I and stage II disease. In addition, patients with stage II disease had significantly lower OS and DRFS rates than stage I, and patients with stage III/IV had significantly lower OS than stage I and II.

Treatment

In general, the treatment reported in the literatures for NECB is similar to the treatment for the usual ductal-type carcinoma not-otherwise specified. Surgical management had included lumpectomy with or without adjuvant radiation therapy or mastectomy [10, 12, 15, 38].

Different chemotherapy regimens have been utilized in the neo-adjuvant, adjuvant, and metastatic settings. Chemo-regimens used to treat the usual type ductal breast carcinoma are also utilized in similar setting in NECB [10, 12]. On the other hand, chemo-regimens similar to those used for pulmonary small cell carcinoma have been attempted in the primary breast SmCC subtype [43]. Because hormone receptors are very commonly expressed in NECB, hormonal therapy for such patients was explored in the adjuvant [10, 12] and the neo-adjuvant settings [12], and its effectiveness has also been reported in the metastatic settings both in small cell and in non-small cell NEBCs carcinomas subtypes [44].

The benefit of chemotherapy versus hormonal therapy versus adjuvant radiation therapy has been debated by Bing et al. [12] who reported on 74 cases of NECB excluding SmCC subtype. In that study, only endocrine and radiation therapy showed a trend in survival benefit, while standard chemotherapies produced poorer outcomes. The poor response to chemotherapy is consistent with the observation that NE carcinoma of the breast is similar to NE carcinoma of other sites such as the lungs and the gastrointestinal tract that tend to be resistant to chemotherapy [45]. It was also found in Bang et al. analysis that any combination of 2 or 3 (hormonal therapy, radiation therapy, chemotherapy) methods did not appear superior to each other nor to any single therapeutic modality. But Bing et al. acknowledged that the small number of patients in his report along with the short-term follow-up for some patients make it difficult to make recommendations regarding the best treatment approach to manage such patients. Despite the limitation of the study by Bang et al., it is the only report to date that such number of patients with NECB were investigated for different treatment options and their associated outcomes in comparison with the usual breast ductal carcinoma NOS.

Unsolved problems

First, all the studies conducted to date so far had analyzed NECB as a single entity without subdividing it into various distinct subtypes, with SmCC representing the most aggressive subtype, while mucinous subtype representing the other end of the spectrum with the best reported prognosis. Second, conventional prognostic parameters such as histologic grade were not consistently taken into account when cases were compared with non-NECB, with notable exception of Bing et al.’s report. Finally, due to the rarity of NECB among others, no randomized control trials were conducted to date to compare the different treatment strategies and their outcomes based on variety of different risk factors.

Conclusion

Primary neuroendocrine carcinoma of the breast is a very rare tumor. The WHO recognized this category in 2003. It seems that different subtypes of NECB have different prognosis. Very few reports have been published to date to compare the different histologic subtypes and their treatment options. In light of the limited available data, treatment algorithm similar to what is used to manage the usual type ductal carcinoma offers the most reasonable approach to treat such patients, with the caveat that there is not enough evidence to support chemotherapy or any chemo-regimen versus other to manage this rare carcinoma. On the other hand, there is some evidence to suggest that hormonal therapy should be the utilized in the management of NEBC in the adjuvant or metastatic setting according to the hormone receptor status. In summary, more research is needed for this rare tumor.

Conflict of interest

The authors declare that they have no conflict of interest.

Copyright information

© Springer Science+Business Media, LLC 2012