Abstract
Aim: Primary neuroendocrine carcinomas of the breast (PNCB) are very rare and tumor markers for this indication are not well defined. We aim at reporting a case and providing a marker useful for prognosis and prediction of tumor recurrence for patients with PNCB. Case Report: A 75-year-old woman presented with a slight painful lump in her left breast of more than 6 months duration. Prior to surgery, the serum level of carcinoembryonic antigen (CEA) (54.4 ng/ml; normal limit <5.0 ng/ml) was significantly elevated. Ultrasonography identified a hypoechoic lesion. Mammography revealed a hyperdense lesion with a well-circumscribed margin. The patient underwent a modified radical mastectomy with axillary lymph node dissection. Pathology showed tumor cells with neuroendocrine features, with diffuse immunopositivity for chromogranin and synaptophysin. The tumor cells were also strongly positive for progesterone and estrogen receptor, but negative for HER-2/neu expression. The CEA value gradually decreased to the normal range within one month after surgery. Neither recurrence nor distant metastasis has been detected at 20 months after surgery and hormone therapy with letrozole. The serial CEA levels were within normal limits in the follow-up period. Conclusion: The serum CEA level after surgery may be a potential marker for evaluating tumor recurrence or prognosis of patients with PNCB.
Primary neuroendocrine carcinomas of the breast (PNCB) are very rare and have been reported mainly occurred in elderly women (1). They were first described by Cubilla and Woodruff in 1977 (2) and the incidence rate ranged from 0.27-0.5% in breast carcinomas (3, 4). Due to its rarity, tumor detection, treatment and prognosis of patients having PNCB is not well defined. We report the case of PNCB in a 75-year-old woman in Taiwan and make a comprehensive review of tumor detection, treatment and prognosis of patients having PNCB.
Case Report
A 75-year-old woman presenting with a slight painful lump in her left breast of more than 6 months duration visited our hospital. Upon examination, an ill-defined hard lump measuring about 4.0 cm was felt in central portion of her left breast, near the nipple-areolar complex. There was no axillary lymphoadenopathy. Ultrasonography showed a 4.0 cm hypoechoic lesion with a slightly irregular contour and heterogenous content (Figure 1). Mammography revealed a hyperdense lesion with a well-circumscribed margin, without microcalcification (Figure 2). Ultrasound-guided core needle biopsy for the mass demonstrated a mucinous carcinoma. Prior to surgery for removal of the lesion, the serum level of the tumor marker carcinoembryonic antigen (CEA), 54.4 ng/ml, was significantly elevated (normal limit <5 ng/ml), while the cancer antigen (CA) 15.3 value was within the normal range.
The patient underwent a modified radical mastectomy of the left breast with axillary lymph node dissection. Grossly, the breast tumor was a well-circumscribed, tan to grayish, firm mass, measuring 4.0×3.0×3.0 cm. Upon microscopic examination, the tumor was composed of round to oval neoplastic cells arranged in a solid-nested, trabecular or ribbon growth pattern and separated by a delicate fibrovascular stroma, with abundant extracellular mucin production (Figure 3). These tumor cells contained relatively uniform nuclei, with fine chromatin and moderate granular eosinophilic cytoplasm. There was evidence of increased mitotic activity, with 5 mitotic features per 10 high-power fields, marked tumor necrosis and hemorrhage. No lymph node metastasis was noted. Immunohistochemical staining of tumor cells revealed a diffusely cytoplasmic positivity for chromogranin and synaptophysin (Figure 4), which are neuroendocrine markers. Tumor cells were also strongly positive for progesterone and estrogen receptors but negative for HER-2/neu expression. We had examined the head, neck, lung, gastrointestinal tract and bone in order to exclude possible nonmammarian primary sites or distant metastasis. The breast tumor pathology was revised to PNCB, solid type, of histological grade 1 and stage IIA (pT2N0M0).
The patient received adjuvant hormone therapy with letrozole after surgery. The CEA value was decreased to 20.5 ng/ml at 8 days after surgery, and then decreased to within the normal range (CEA value 4.38 ng/ml) at one month after surgery. There was no evidence of recurrence or distant metastasis 20 months after surgical treatment and hormone therapy with letrozole. The serial serum CEA levels were within normal limits in the follow-up period.
Discussion
Neuroendocrine carcinomas, including carcinoid tumors are malignant tumors with neuroendocrine differentiation, which arise mainly from endocrine organs and non-endocrine organs, such as the gastrointestinal system and lungs (5), but rarely from breast. Unlike such lesions in gastrointestinal tract or lungs, benign neuroendocrine tumors have, to our knowledge, never been reported in breast; therefore, all breast neuroendocrine lesions have previously been considered as carcinomas (6).
PNCB was first described by Cubilla and Woodruff in 1977, and only eight cases of breast carcinoid tumors with argyrophilic granules were identified (2). In 1988, chromogranin and secretogranin were found to be localizedly expressed in some argyrophilic breast carcinomas (7). In 2000, Sapino et al. firstly defined differentiated neuroendocrine breast carcinomas as tumors with specific morphological features which expressed neuroendocrine markers in more than 50% of the tumor cells (8), compatible with the 2003 WHO classification of PNCB based on two characteristics: morphologic features and neuroendocrine differentiation (9). Chromogranin and synaptophysin have been identified as specific markers for neuroendocrine differentiation (10, 11). By definition, breast carcinomas showing neuroendocrine differentiation only in scattered or some tumor cells are not included in PNCB. Accordingly, the incidence of PNCB is very low. For instance, the incidence in the report of Gunhan-Bilgen et al. was only 0.27% in 1,845 cases of breast carcinoma (3). The incidence in the report of López-Bonet et al. was about 0.5% in a series of 1,368 histopathologically proven breast carcinomas (4). Moreover, the disease occurrs mainly in elderly women in sixth to seventh decades of life (1, 12, 13).
Miremadi and colleagues observed that the amount of neuroendocrine differentiation in breast carcinomas bears no relation to prognostic factors or patient outcome (14). Histological grade is one of the most important prognostic factors (15). Solid neuroendocrine carcinoma and atypical carcinoids are regarded as well-differentiated tumors, with better prognosis with adequate surgical and adjuvant therapy (16); in contrast, small cell and large cell neuroendocrine carcinomas are poorly differentiated, giving a negative prognosis (17, 18). Since our patient was a case of solid neuroendocrine carcinoma of the breast with strong positivity for progesterone and estrogen receptors and a lack of lymph node involvement, we believe the patient should have a good prognosis, which is supported by no evidence of recurrence or distant metastasis at 20 months after surgery and hormone therapy.
Several tumor markers can be detected in the serum of patients with malignancy. The concept of serum tumor marker represents a quantifiable assessment of the tumor burden at that time. The use of tumor markers concern several different aspects, such as determination of cancer risk, screening, diagnosis, prognosis, prediction of response to therapy, and monitoring the course of disease (19-21). The most common serum markers used for postoperative monitoring of breast cancer are CEA and CA15-3 (21, 22). The European Group on Tumor Markers suggests that CEA and CA15-3 testing should be performed even in asymptomatic women despite the impacts of the lead time on patients' survival not being clear (23). Nobels and colleagues analyzed 211 patients with neuroendocrine tumors and 180 controls with nonendocrine tumors, and stated that chromogranin is the best general neuroendocrine serum marker available. Unfortunately, it is not a very sensitive marker (24). Until now, the association between PNCB and tumor markers has not been well investigated because of its rarity. In this case, because of the significantly elevated CEA value in serum before surgery and into gradually reduction to normal within one month after surgery, we propose that the change of CEA levels may be a potential marker in the follow-up of treatment of patients and their prognosis.
In our present case, the specimen of core biopsy was initially misdiagnosed as a mucinous adenocarcinoma due to florid extracellular mucin production over the stroma. But in considering the whole specimen of this tumor, the extracellular mucinous stroma did not fit the criterion of mucinous adenocarcinoma that more than 50% of the tumor should contain extracellular mucinous pools (9). In a previous report, about 26% of neuroendocrine carcinomas exhibited either intracellular or extracellular mucin, or both (13). Therefore, the amount of extracellular mucin production may reveal a limitation for making a correct diagnosis of neuroendocrine carcinoma of the breast based on small core biopsies.
Ultrasonography showing a hypoechoic lesion with slightly irregular contours and heterogenous content.
Mammography revealing a hyperdense focal lesion with circumscribed margins and without microcalcification: medial-lateral oblique view (A), caudal-coronal view (B).
The clinical and imaging features of neuroendocrine carcinoma of the breast mimic other types of breast carcinoma in many ways without any specificity (9). Neuroendocrine tumors often present as dense round or irregular masses with a spiculated or lobulated margin in mammography (3, 25, 26). On ultrasonography, neuroendocrine tumors are hypoechoic with cystic or no cystic component (26, 27). Hence it is difficult to diagnose neuroendocrine tumors based on these image findings alone. The ultrasonographic and mammo-graphic examinations in our present case revealed a focal lesion with indistinctive margin in the left breast. The image studies and clinical features were not specific for diagnosis of PNCB.
Histopathology showing ribbons and nests of tumor cells within the mucinous stroma. Haematoxylin-eosin stain (×200).
Immunohistochemistry showing diffusely synaptophsin-positive staining (brown color, ×200).
The best treatment for patients with PNCB is still unknown (28). In general, surgery is regarded as the major treatment among the strategy choices (6). The surgical choice includes lumpectomy and modified radical mastectomy with axillary lymph node dissection, or sentinel lymph node biopsy based on the tumor size, location and stage. Chemotherapy or hormone therapy may be given according to clinical tumor status and biomarkers (6, 28, 29). Our case surgery operation and then received adjuvant hormone therapy with letrozole thereafter. There was no evidence of recurrence or distant metastasis at 20 months after operation.
In conclusion, PNCB is a very rare incidence. Tumor histological grade is a known prognostic factor. Long-term prognosis and tumor behavior are not well known because of its rarity. The only index in the present case is the significantly altered CEA level before and after surgery. We propose that the change of CEA levels may be a potential marker for clinical application in follow-up, in addition to its was in detection. Further studies recruiting larger patient populations with long-term follow-up may lead us to clearly understand this new concept.
Acknowledgements
This study was supported by research grants from the Terry Fox Cancer Research Foundation, and China Medical University and Hospital (DMR-100-179).
- Received October 7, 2011.
- Revision received December 6, 2011.
- Accepted December 7, 2011.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
