Role of FNAC in Extramammary Tumors Metastatic to the Breast.

BACKGROUND: Metastasis to the breast of an extra-mammary origin is very rare. FNAC plays an important role in differentiating non-mammary breast metastasis from primary malignancy. This study aimed to analyze the cytomorphological criteria and its pitfalls in differentiating metastatic lesion of the breast from primary malignancy.
METHODOLOGY: Retrospective analysis of 891 FNACs of the breast was performed for a time span of 3 years. A total of 12 cases were diagnosed on FNAC as secondary neoplasms to the breast. Clinical and radiological data, along with Pap and MGG stained smears of each case were examined and correlated with the histopathology of the primary tumor. Statistical analysis was carried out. All cases of primary breast malignancies were excluded from our study.
RESULTS: In 10 out of 12 cases, primary malignancies were identified as Plasma cell myeloma (one case), B-acute lymphoblastic leukemia (two cases), acute myeloid leukemia (one case); chronic myeloid leukemia (one case), Burkitt's lymphoma of the ovary (one case), Diffuse large B-cell lymphoma (one case), esophageal squamous cell carcinoma (one case), spindle cell sarcoma (one case) and squamous cell carcinoma of the cervix (one case). The remaining two cases in our study were misdiagnosed on cytology as metastasis and turned out to be breast primaries on histopathology.
CONCLUSION: Our case series highlights the importance of FNAC to differentiate secondary lesions from primary breast malignancy and thus helps to avoid unnecessary surgery to the patient. It emphasizes on the need to keep in mind the possibility of metastatic breast neoplasms in the presence of unusual cytological features on FNAC. Copyright:

INTRODUCTION

According to Globocon India 2018, a world health organization (WHO) initiative, breast cancer is now considered the most common malignancy affecting Indian women accounting for about 25% of the malignancies seen.[1] The most common primary breast carcinoma is invasive ductal carcinoma (IDC) which accounts for 80% of the cases.[2] Breast as a site of metastasis is an uncommon entity. The majority of the cases metastasizing to the breast are primaries from the contralateral breast.

Fine needle aspiration cytology (FNAC) is a valuable tool in the evaluation of breast lesions as the first line of investigation and also for suspected recurrence in breast malignancy. It is simple, safe, cost-effective, and relatively less invasive with reported specificity of 89.8% to 100% and sensitivity of 43.8% to 95%.[3]

Metastatic involvement by extramammary primary is extremely rare and accounts for less than 2% of breast malignancies.[4] Among the non-mammary *primaries, hematological malignancies and melanomas are the most common cancers followed by lung, ovary, soft tissue sarcomas, and prostate carcinomas showing metastasis to the breast.[5] Metastasis from gastrointestinal tumors like the esophagus, stomach, and colo-rectum is very rare.[6] Clinically, these lesions present as firm, rounded, freely mobile masses without any fixation to surrounding parenchyma. On mammography, they show single or multiple well-circumscribed nodules with irregular margins, in the absence of microcalcification and spiculations.[7]

Accurate diagnosis is mandatory to differentiate breast metastasis from primary carcinoma as they have different treatment protocols. Failure to make this differentiation can result in unnecessary mastectomy and sentinel lymph node dissection. Extramammary metastasis is often a marker for disseminated malignancy and it indicates a poor prognosis.

This study aims to analyze the cytomorphological criteria and its pitfalls in differentiating metastatic lesion of the breast from primary malignancy.

MATERIALS AND METHODS

An observational and retrospective cross-sectional study was conducted on metastatic malignancies to the breast, diagnosed on FNAC. These cases were obtained from the files of the Department of Cytopathology from January 2016 to December 2018. The study was conducted with approval from the ethics committee of the institute.

Clinical data including age and sex of the patient along with features of the breast lump such as laterality and number were noted. Findings were correlated with the radiological data of each case. History of known primary malignancy and presence of systemic metastasis in sites other than breast was also evaluated.

FNAC procedure was performed by direct palpation or under ultrasound guidance using 22G to 24G needle. Five to six smears were prepared with the aspirated material in each case. Air-dried and alcohol fixed smears were stained with May-Grunwald-Giemsa (MGG) and Papanicolaou stain, respectively. Smears were evaluated for cellularity, patterns, cellular details, mitosis, necrosis, and presence of benign breast tissue.

In cases with a known primary, the biopsy, surgical specimens, and immunohistochemistry (IHC) of the primary malignant mass were examined. The histopathology slides of breast biopsy (available in 7 cases) were reviewed and compared with the cytological material. In the case of hematolymphoid malignancies, immunophenotyping (IPT) of the bone marrow was performed for further characterization.

In addition, cases diagnosed as extra-mammary metastasis to the breast on histopathological samples were also examined. These were cross-referenced with the cytology records to identify cases diagnosed as extramammary metastasis on histopathology and reported as otherwise on cytology. No such cases were identified.

For statistical analysis, a two by two contingency table was prepared [Table 1] and the sensitivity, specificity, positive, and negative predictive value of FNAC in extra-mammary metastasis to the breast was calculated.

Table 1

FNAC and Histopathological correlation of 891 cases

FNAC outcome	Histopathological diagnosis
	Extramammary metastasis	Non-Metastatic lesion
Extramammary metastasis	10 (True positives)	2 (False positive)
Non-Metastatic lesion	0 (False negatives)	879 (True negatives)

All cases with primary breast cancer were excluded. Patients were followed up for a period of 6 months to 2 years.

OBSERVATION AND RESULTS

A total of 891 FNAC from breast masses were performed over a period of three years and twelve cases were diagnosed cytologically as extramammary tumors metastasizing to the breast.

There were ten female and two male patients in the age group of 20-70 years. Clinically, 7 cases showed solitary breast lesions, 1 case showed multiple lesions, and the remaining 4 cases were found to be bilateral tumors. Eight out of twelve cases had a known primary at the time of diagnosis. Two of them had existing systemic metastasis and the remaining six cases had breast as their only site of metastasis. In four cases the lesion in the breast was the initial presentation of disease which led to the diagnosis of the primary malignancy.

Squamous cell carcinoma- smear shows benign ductal cells of breast with malignant squamous cells having high N: C ratio, round to oval hyperchromatic nucleus. Inset shows malignant squamous cells. Papanicolaou's stain. (×400)

On correlation with histopathology, we found 2 cases in our study that were misdiagnosed as metastasis on cytological examination. A detailed evaluation of the histopathology slides of these cases found them to be primary tumors of the breast.

Key clinico-radiological features and final histopathological diagnosis have been summarized in Table 2. Cytological features of all cases have been described in Table 3.

Table 2

Clinical and radiological details of cases

Case no.	Age/sex	U/L or B/L breast	Radiology	Cytological diagnosis	Known Primary	Primary site	Other site of metastasis	Final histopathological diagnosis
1.	67/M	U/L	-	Malignant recurrent plasma cell myeloma [Figure 1c]	Yes	Bone Marrow	-	Multiple Myeloma
2.	49/F	U/L	Soft tissue density in retro areolar region and extends to involve overlying skin. P/O breast primary	Spindle cell tumor [Figure 2a]	Yes	Soft tissue	Lung	Undifferentiated Spindle cell sarcoma
3.	48/F	B/L	Solitary rounded lesion having benign appearance	Malignant round cell tumor. P/O granulocytic sarcoma	Yes	Bone marrow	-	Acute Myeloid Leukemia-M4
4.	33/F	U/L	No detectable mammographic changes	Involvement by acute lymphoblastic leukemia [Figure 1b]	Yes	Bone marrow	-	B-Acute Lymphoblastic Leukemia
5.	46/F	U/L	-	Squamous cell carcinoma [Figure 3]	Yes	Cervix	Lung	Squamous cell carcinoma
6.	40/F	U/L	Ill defined soft tissue opacities with specks of micro calcification suggestive of metastasis	Malignant- Metastasis from Esophagus [Figure 1a]	Yes	Esophagus	-	Poorly differentiated squamous cell carcinoma
7.	35/F	B/L	Solitary rounded lesion having benign appearance	Malignant round cell tumor P/O Non Hodgkin’s lymphoma	No	B/L Ovary	-	Burkitt’s lymphoma
8.	72/F	U/L	-	P/O - Non Hodgkin’s lymphoma or Poorly differentiated carcinoma	No	Axillary lymph node	-	DLBCL-ABC
9.	22/F	B/L	Mass with infiltrating margins and axillary lymph nodes involved suggesting primary breast carcinoma with nodal metastasis	Malignant round cell tumor	Yes	Bone Marrow	-	B-Acute Lymphoblastic Leukemia
10.	26/M	B/L	-	Granulocytic sarcoma [Figure 1d]	Yes	Bone Marrow	-	Chronic Myeloid Leukemia- myeloblastic crises
11.	56/F	U/L	-	?Metastatic carcinoma?Malignant Phylloides tumor [Figure 2b]	No	Breast	-	(L) MRM- Malignant Phylloides
12.	38/F	U/L	Heterogeneous lesion with few solid and cystic areas ? Fibroadenoma ? Neoplasm	P/O Stromal tumor	No	Breast	-	Giant cell tumour of soft tissue breast

FNAC=Fine-Needle Aspiration Cytology, HPE=Histopathological examination, U/L=Unilateral, B/L=Bilateral, P/O=Possibility of, NHL=Non-Hodgkins Lymphoma, IHC=Immunohistochemistry, MRM=Modified Radical Mastectomy

Table 3

Cytomorphological Features

Case	Case 1 [Multiple Myeloma]	Case 2 [Undifferentiated Spindle cell sarcoma]	Case 3 [AML-M4]	Case 4&9 [B-ALL]	Case 5 [Squamous cell carcinoma]	Case 6 [Squamous cell carcinoma]	Case 7&8 [NHL]	Case 10 [CML-myeloblastic crises]	Case 11 [Malignant Phyllodes]	Case 12 [Giant cell tumor soft tissue breast]
Arrangement of cell	Relatively monotonous, discohesive cells	Loose cohesive clusters	Dispersed	Dispersed population	Small, cohesive clusters	Cohesive cluster	Discohesive cells	Discohesive, polymorphic population	Cohesive clusters and sheets	Singly scattered with few fragments
Size and shape of cells	Medium to large cells plasmacytoid cells	Spindle to cigar shaped cells with high N: C ratio and	Medium to large cells	Medium sized atypical, monotonous population of cells	Medium to large cells having high N: C ratio	Polygonal cells with high N: C ratio	Small to medium sized cells	Predominantly large cells	Proliferation of spindle shaped stromal cells with aggregates of epithelial cells	Spindle cells along with multinucleate giant cells
Cytoplasm	Moderate amount of pale cytoplasm and paranuclear halo	Scant cytoplasm	Moderate amount of pale cytoplasm	Scant rim of pale cytoplasm	Moderate amount of eosinophilic cytoplasm	Moderate amount of refractive, orangish cytoplasm	Scant amount of cytoplasm	Moderate amount of pale cytoplasm	Pale eosinophilic	Moderate pale cytoplasm
Nuclear features	Eccentrically situated, round to oval nucleus with coarse nuclear chromatin	Moderately pleomorphic, hyperchromatic	Irregular nuclear membrane, fine nuclear chromatin with prominent nucleoli	Irregular nuclear membrane, coarse chromatin and inconspicuous nucleoli (+/-)	Pleomorphic cells, round to oval nucleus with open nuclear chromatin, prominent nucleoli	Pleomorphic, hyperchromatic nucleus	Monomorphic roundnucleus, coarse nuclear chromatin, and prominent nucleoli	Pleomorphic, irregular nuclear membrane, fine chromatin with 1 or more prominent nucleoli	Stromal cells with mild atypia. Epithelial cells- marked anisonucleosis, irregular nuclear membrane, coarse chromatin	Bland spindle cells with elongated nucleus and incospicious nucleoli
Binucleation/Multinucleation	Binucleation +	Multinucleation +	-	-	-	-	-	-	Multinucleate tumor giant cells	Multinucleation +
Mitosis	-	+	+	+	-	+		+	+	+
Necrosis	-	-	-	-	+	+	-	-	+	-
Other	Plasmablasts +	-		Apoptotic bodies seen		-		Few myeloid precursors, neutrophils and eosinophils +	-	-
Normal breast tissue	-	-	+	-	+	+	-	-	+	+

AML-M4=Acute Myelomonocytic Leukemia, B-ALL=B-cell Acute Lymphoblastic Leukemia, NHL=Non-Hodgkin’s Lymphoma, CML=Chronic Myelogenous Leukemia, N: C ratio=nuclear-cytoplasmic ratio

Of the two cases misdiagnosed in our study, one case showed malignant spindle and pleomorphic cells admixed with normal ductal cells. The differential diagnosis of malignant phyllodes along with the possibility of metastatic carcinoma was given on cytology [Figure 2b]. On the final histopathological examination, it was found to be malignant phyllodes. The second case was reported as a stromal tumor on cytology which showed mononuclear spindle cells admixed with osteoclast giant cells, with no evidence of epithelial component. On histopathology, it was reported as a primary giant cell tumor of soft tissue breast.

Figure 2

(a) Spindle cell sarcoma: smear showing spindle to cigar-shaped cells arranged in loose clusters with cells having high N: C ratio, moderate nuclear pleomorphism, and scant cytoplasm. (Papanicolaou's, ×400). (b) Malignant Phyllodes Tumor: smear shows sheets of benign ductal epithelial with bizarre multinucleated giant cell. (Papanicolaou's, ×400)

The treatment plan for 9 out of 10 cases was available. Since the involvement of the breast indicates Stage IV malignancy, most of the patients received chemo-radiotherapy according to NCCN guidelines in each case according to its primary malignancy. Follow up of 7 patients was available for up to 2 years. These patients are alive and well. The remaining 5 cases were lost to follow up.

Statistical analysis

Out of the total 891 breast FNACs performed, 10 cases showed extramammary metastasis to the breast, whereas 2 cases were misdiagnosed as metastasis on cytology and turned out to be primary tumors of the breast on histopathology. The remaining 879 did not show evidence of extramammary metastasis on both cytology and histopathology. There were no cases that were missed on cytology and diagnosed as extramammary metastasis on histopathology.

To evaluate the diagnostic accuracy of FNAC in non-mammary metastasis to the breast, we calculated the sensitivity and specificity, which was 100% and 99.7%, respectively. A positive predictive value of 83.3% and a negative predictive value of 100% was also determined for the same.

DISCUSSION

Metastasis to the breast from an extra-mammary neoplasm is an extremely rare phenomenon with the reported incidence varying from 0.5 to 6.6% in various studies.[789] Our series also showed an incidence of 1.2% highlighting the infrequent occurrence. The rarity of breast as a site of metastasis in comparison to other body organs is believed to be due to large areas of fibrous tissue and relatively less vasculature.[1011] Increased reports of metastasis at unusual sites including breast in recent times are due to multiple factors such as better radiological modalities for early detection of asymptomatic lesions, increased overall survival of cancer patients owing to the multidisciplinary approach, targeted, and specific treatment which acts predominantly on the primary tumor but may not be as effective on the metastatic lesions.[12]

In our study, 33% of cases presented with the breast as the initial site of malignant disease, which was in concordance with other studies reporting a range of 11-33%.[71314] According to the literature, common malignancies metastasizing to the breast are of hematolymphoid origin, malignant melanoma, rhabdomyosarcoma, ovarian, and bronchogenic carcinoma.[56713] 58% of our cases were metastasis from hematolymphoid malignancies [Figure 1]. Shukla et al. and Mandal et al. had comparative findings with our study, having hematolymphoid origin tumors as the commonest metastasis to breast.[513] Metastasis to the male breast is an uncommon entity, reported most commonly from prostatic adenocarcinoma.[714] We had two extra-mammary metastasis in the male breast. Both showed hematolymphoid [Figure 1c] type malignancy- metastasis from plasma cell myeloma in one and chronic myeloid leukemia in another case was suggested [Figure 1d].

Figure 1

(a) Burkitt's Lymphoma- Medium-sized cells with a round nucleus, coarse nuclear chromatin, and prominent nucleoli. (Papanicolaou's, ×400). (b) Acute lymphoblastic leukemia – Monomorphic round cells with scant cytoplasm, fine nuclear chromatin, and inconspicuous nucleoli. (Papanicolaou's, ×400) (c) Plasma cell myeloma- Medium to large plasmacytoid cells with coarse nuclear chromatin. Binucleation is seen. (Papanicolaou's, ×400) (d) CML-blast crisis- Predominant blast population having high N: C ratio with pleomorphic nucleus, fine chromatin with 1 or more prominent nucleoli. (Papanicolaou's, ×400)

Clinically, metastasis to the breast from extra-mammary primary usually presents as palpable, rapidly growing, well-circumscribed rounded mass without overlying skin involvement or nipple retraction.[5] On mammography, metastatic breast lesions have non-specific and deceptive features which makes it difficult to distinguish it from primary breast lesions. In our case series, 3 cases were reported as benign, 2 cases as primary breast malignancy, and only one case as metastatic lesion on mammography.

Hence, FNAC plays a key role as a first-line diagnostic procedure with a reported sensitivity of 43.8% to 95%, but the cytomorphological diagnosis of an extramammary tumor is not always straightforward. The reasons may be a confusing clinical picture, non-specific mammographic findings, and/or tumors involving the breast in the clinical setting of genetically determined multiple primary malignancies. Hence, along with cytomorphological features, it may be necessary to use ancillary studies to arrive at the correct diagnosis.

While evaluating cytology smears in such cases, our first goal is to exclude any primary breast lesion such as pre-malignant carcinoma-in-situ or invasive carcinomas along with its variants. Presence of any atypical ductal epithelial cells or foci of ductal carcinoma in situ raises the possibility of a primary lesion. However, the presence of a dispersed population of undifferentiated small cells, malignant cells of hematopoietic origin, dysplastic keratinized cells, pleomorphic spindle cells, and cells with neuroendocrine morphology should alert one to rule out metastasis before labeling lesion as breast primary.[7]

In our series, eight cases had a known primary at the time of breast FNAC and this information played a significant role in reaching a diagnosis. Along with the clinical history, cytological features like presence of a dispersed population of malignant round cells suggestive of hematopoietic origin, premature myeloid series cells suggestive of granulocytic sarcoma, atypical bizarre and immature plasma cells suggestive of metastasis from myeloma, dysplastic and malignant squamous cells suggestive of metastasis from known primary squamous cell carcinoma and pleomorphic spindle cells without malignant breast epithelial cells suggestive of a metastatic lesion from sarcoma, played an important role in the diagnosis of metastatic lesion over breast primary.

The remaining 4 cases in our study, presented with breast as the initial site of malignant disease. On cytology, two of the four cases had a classical dispersed, monotonous population of tumor cells without the formation of true tissue aggregates, suggestive of malignant round cell tumor- Non-Hodgkin's lymphoma. On further workup, they were confirmed to be an advanced stage primary nodal lymphomas.

The other two cases were misdiagnosed in our study, one case showed even distribution of multinucleated osteoclast-like giant cells with stromal cells that resembled the giant cell tumor of bone along with the presence of sheets of benign ductal cells and myoepithelial cells. This case was misinterpreted as metastasis, as on further histopathology it was proven to be primary giant cell tumor of the breast. When dealing with such cases it is important to consider the following differentials- primary breast malignancy with giant cells, giant cell-rich variants of osteosarcoma/leiomyosarcoma, and metastasis from a giant cell tumor of bone.[1516] With respect to our case, the lack of malignant epithelial component (for carcinoma), absence of high-grade nuclear atypia (as seen in osteosarcoma/leiomyosarcoma), and no evidence of any primary bone malignancy helped us to make the diagnosis of giant cell tumor of the breast on histopathology.[15] This entity although rare must be borne in mind when encountered with giant cells on FNAC of the breast.

The second case showed atypical cytological features including high cellularity, presence of malignant spindle cells, focal crowding, nuclear abnormalities, bizarre multinucleated giant cells along with proliferative ductal cells. We misdiagnosed this case because of minimal stromal component which later turned out to be malignant phyllodes on histopathology.

A variety of metastatic and primary lesions to the breast can present with spindle cell morphology. These lesions may range from inflammatory, benign proliferative to high-grade malignancies making their diagnosis very challenging. Phyllodes tumors (PT) with stromal overgrowth and metaplastic spindle-cell carcinoma (MC) represent the most common malignant spindle-cell lesions of the breast. The presence of a conspicuous, benign, non-neoplastic epithelial component would favor a phyllodes tumor, while a biphasic epithelial/mesenchymal phenotype would favor a metaplastic carcinoma.[17] Primary breast sarcomas and sarcomas metastatic to the breast are exceedingly rare and difficult to distinguish histologically from phyllodes tumor or metaplastic breast carcinoma. A history of previous sarcomas along with imaging and clinical correlation may help in making the diagnosis.[17]

Ancillary studies on cell bock material and immunocytochemistry can play an important role to avoid such misdiagnosis. In most circumstances, sufficient material for these ancillary studies is not obtained. Working in close coordination with the clinician and rapid onsite evaluation can help to obtain adequate material for cell block, immunocytochemistry, flow cytometry, or other ancillary studies. Estrogen receptor, gross cystic disease fluid protein 15 (GCDFP-15), mammaglobin-A, and GATA3 are the most commonly used breast-specific immunomarkers but have variable sensitivities and specificities.[18] Hence, careful examination of cytomorphology and clinical history is key to an accurate diagnosis.

In conclusion, achieving optimal diagnosis in breast lesions requires a comprehensive approach that includes clinical history, radiographic findings, and careful assessment of cytological features. When unusual cytological features like malignant round cells, squamous cells- dysplastic or malignant, melanin pigment, or psammoma bodies are seen, we need to consider the possibility of metastatic lesions. The cytology of the metastasis depends upon the nature of the primary tumor and comparing it's cytomorphology with the morphology of a known primary tumor can help to confirm or exclude the diagnosis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.