Role of Fine-Needle AspirationCytology in Evaluation ofCutaneous MetastasesSonal Sharma, M.D.,* Mrinalini Kotru, M.D., D.N.B., Amit Yadav, M.D.,Manish Chugh, M.D., Anu Chawla, M.D., and Mani Makhija, M.D., D.N.B.

Skin is an uncommon site for metastasis. This study was done toevaluate the role of FNAC as an important tool for investigatingcutaneous and subcutaneous nodules in patients with knownmalignancy or as a primary manifestation of an unknown malig-nancy.

All the FNAC done from January 2003 to August 2008 werereviewed (n ¼ 55,556). Ninty-five patients (49 males and 46females with age range of 4–96 years) with cutaneous/subcuta-neous nodules which were diagnosed as metastasis were ana-lyzed. Primary tumors of skin/subcutis were excluded from thestudy.

In our study, 63 out of 95 cases had a known primary malig-nancy. Of these, five had underlying hematological malignancyand 58 patients had solid organ tumors. Lung carcinoma wasseen to metastasize most commonly to skin in males and breastcarcinoma in females. The most common site for a cutaneous/subcutaneous metastasis was chest wall [40 followed by abdomi-nal wall (14) and scalp (9)]. Multiple site involvement was alsoobserved (8). In 32 cases primary site was not known. Theywere most commonly diagnosed as poorly differentiated carci-noma followed by adenocarcinoma.

FNAC can diagnose a variety of tumors in the skin and supportthe diagnosis of a metastasis in case of a known primary andoffer a clue to underlying malignancy in case of an occult pri-mary. Diagn. Cytopathol. 2009;37:876–880. ' 2009 Wiley-Liss, Inc.

Key Words: cytology; cutaneous metastases

Cutaneous metastases are uncommon presentations of an

underlying malignancy. Data from autopsy series indicate

an incidence of 0.8%–4%.1 In most of the cases they

occur in patients known to have some malignancy. But

rarely, they may be the first manifestation leading to rec-

ognition of underlying condition on detailed investigation.

Most of these cases represent terminal phase of the dis-

ease. Therefore, it is appropriate to utilize fine-needle

aspiration cytology (FNAC) as a minimally invasive tool

for diagnosis in these cases, thereby, obviating the need

for a biopsy. Although, there are case reports in literature

on cutaneous metastases,2,3 the role of FNAC as an im-

portant tool for investigating these patients has been

reported in only a few studies.4–6

Therefore, this study was aimed to investigate cases

which presented with a clinical impression of cutaneous/

subcutaneous metastases from pre-existing tumors or as

the first manifestation of an unknown primary. The role

of FNAC in evaluating these cases was assessed.

Materials and Methods

This study is a retrospective study in which patients who

presented with palpable cutaneous or subcutaneous nod-

ules suspicious of being metastases were included.

Records of all the patients subjected to FNAC in the Uni-

versity College of Medical Sciences & GTB Hospital

between January 2003 and August 2008 were reviewed.

Of 55,556 cases reviewed, 95 cases diagnosed as cutane-

ous/subcutaneous metastasis were included in the study.

Cases with cutaneous nodules occurring at the site of pre-

vious surgery were excluded from the study. Cases sus-

pected as primary tumors of skin/subcutaneous tissue

were also excluded from the study.

FNAC was performed using a 22-gauge needle and 10-ml

disposable plastic syringe. Multiple passes were made in

the mass while maintaining negative pressure. Aspirated

material was used to prepare air-dried as well as alcohol-

fixed smears. Air-dried smears were stained by May–

Grunwald–Giemsa method. Alcohol-fixed smears were

stained using Papanicoloau method. Special stains like

Periodic acid Schiff (PAS) were performed if required.

Department of Pathology, University College of Medical Sciences,Delhi

*Correspondence to: Sonal Sharma, M.D., Reader, Department of Pa-thology, University College of medical Sciences, Shahdara, Delhi-110095. E-mail: sonald76@rediffmail.com

Received 17 December 2008; Accepted 27 April 2009DOI 10.1002/dc.21119Published online 15 June 2009 in Wiley InterScience (www.interscience.

wiley.com).

876 Diagnostic Cytopathology, Vol 37, No 12 ' 2009 WILEY-LISS, INC.

Results

There were 95 patients with cutaneous/subcutaneous nod-

ules suspected to be metastases in whom FNAC was car-

ried out between January 2003 and August 2008. The

patients were represented by 49 males and 46 females

(M:F ratio ¼ 1.06:1). Age range was 4–96 years (Mean ¼62 years) in males and 16 to 80 years in females (Mean

¼ 51.5 years).

In 63 of these 95 cases there was a known primary

malignancy at the time of aspiration. Fifty-eight patients

had solid organ tumors and five had underlying hematolog-

ical malignancy. Table I shows the sites of metastasis and

FNA diagnosis in cases with known primary and solid

organ malignancy. Table II displays sites of metastasis and

FNA diagnosis in cases with known hematological malig-

nancy. In the rest 32 cases primary site was not known.

These were classified as cutaneous/subcutaneous metasta-

ses from an unknown primary as shown in Table III.

The commonest source of a known primary tumor was

breast (21) followed by lung (13), ovary, larynx, bone,

Table I. Cutaneous Metastases in Cases With Known Primary

S. No.Site ofprimary No. of cases Site of cutaneous metastasis FNA diagnosis

1 Breast 21 Chest wall (18), axilla (1),abdominal wall (2)

Infiltrating duct carcinoma (17), Poorly differentiatedcarcinoma (3), Invasive carcinoma with mucinousbackground (1)

2 Lung 10 Chest wall (4), arm (1), axilla (1),scalp (2), neck (1), multiple sites(1)

Adenocarcinoma (3), squamous cell carcinoma (3),poorly differentiated carcinoma (4)

3 Larynx 3 Neck (2), Back (1) Squamous cell carcinoma (2), poorly differentiatedcarcinoma

4 Bone 3 Leg (1), Shoulder (2) Ewing’s sarcoma/PNET5 Gall bladder 3 Chest wall (2), umbilicus (1) Adenocarcinoma6 Ovary 3 Chest wall (1), Umbilicus (1),

abdominal wall (1)Adenocarcinoma

7 Kidney 2 Back (1), multiple sites (1) Renal cell carcinoma8 Pancreas 2 Abdominal wall (2) Adenocarcinoma9 Stomach 2 Umbilicus (1), chest wall (1) Adenocarcinoma10 Thyroid 2 Scalp (1), chest wall (1) Follicular carcinoma11 Cervix 2 Chest wall (1), abdominal wall (1) Poorly differentiated carcinoma12 Periampullary 1 Abdominal wall Adenocarcinoma13 Oral cavity 1 Neck Poorly differentiated carcinoma14 Esophagus 1 Abdominal wall Adenocarcinoma15 Colon 1 Multiple sites Adenocarcinoma16 Prostate 1 Abdominal wall Adenocarcinoma

Total 58

Table II. Cutaneous Involvement in Cases With Underlying Hematological Malignancies

S. No. Hematological malignancy No. of cases Site of involvement FNA diagnosis

1 Chronic myeloid leukemia 2 Scalp (1), arm (1) Extramedullary hematopoiesis2 Multiple myeloma 2 Scalp (1), chest wall (1) Plasmacytoma3 Non-Hodgkin lymphoma 1 Chest wall Non-Hodgkin lymphoma

Total 5

Table III. Cutaneous Metastases in Cases with Unknown Primary

S. No. FNA diagnosis No. of cases Site of cutaneous metastases

1 Adenocarcinoma 12 Chest wall (5), abdominal wall (4), umbilicus (1),back (1), multiple sites (1)

2 Poorly differentiated carcinoma 15 Chest wall (5), abdominal wall (1), scalp (3), arm (1),back (1), forehead (1), shoulder (1), multiple sites(2)

3 Squamous cell carcinoma 1 Multiple sites4 Small round cell tumor possibly Ewing’s/PNET 2 Forearm (1), multiple sites (1)5 Small round cell tumor possibly neuroblastoma/Wilms 1 Scalp6 Poorly differentiated tumor possibly neuroendocrine 1 Abdominal wall

Total 32

FNAC OF CUTANEOUS METASTASES

Diagnostic Cytopathology, Vol 37, No 12 877

Diagnostic Cytopathology DOI 10.1002/dc

gall bladder, ovary, and kidney (Fig. 1). Overall the most

common site for a cutaneous/subcutaneous metastasis was

chest wall [n ¼ 40 (42%)] followed by abdominal wall

[n ¼ 14 (15%)] and scalp [n ¼ 9 (9%)]. Multiple site

involvement was also observed [8 cases, (9%)]. Chest

wall involvement was most frequently associated with

carcinoma breast (17 cases) and lung cancer (four cases)

due to proximity of these tumors. However, chest wall

involvement was not limited to these two primary sites

only. It was seen in many other malignancies of organs

like carcinoma ovary (Fig. 2), cervix, thyroid (Fig. 3) and

gallbladder.

Abdominal wall and umbilical involvement was almost

exclusively observed in intra-abdominal malignancies,

except for two cases of carcinoma breast. Another inter-

esting observation was that both cases in which thyroid

was the primary site, were follicular carcinoma.

In males the most common primary was from lung

while in females it was breast. The lung carcinomas

included adenocarcinoma (3), squamous cell carcinoma

(3) and poorly differentiated carcinoma (4). The breast

cancers comprised of infiltrating duct carcinoma (18)

(Fig. 4) and three cases of poorly differentiated carci-

noma. We also encountered 5 cases of hematological

Fig. 1. FNA smears of cutaneous deposits of renal cell carcinoma. MGG3200. [Color figure can be viewed in the online issue, which is availableat www.interscience.wiley.com.]

Fig. 2. Aspirate of subcutaneous nodule ovarian papillary carcinomafrom ascitic fluid tap site. MGG 3100. [Color figure can be viewed inthe online issue, which is available at www.interscience.wiley.com.]

Fig. 3. FNA smears from scalp swelling showing metastatic follicularcarcinoma thyroid. MGG 3200. [Color figure can be viewed in theonline issue, which is available at www.interscience.wiley.com.]

Fig. 4. Aspirate from chest nodule showing infiltratng ductal carcinomabreast. MGG 3400. [Color figure can be viewed in the online issue,which is available at www.interscience.wiley.com.]

SHARMA ET AL.

878 Diagnostic Cytopathology, Vol 37, No 12

Diagnostic Cytopathology DOI 10.1002/dc

malignancies who presented with cutaneous/subcutaneous

nodules. These included previously diagnosed cases of

chronic myeloid leukemia (2 cases) (Fig. 5), multiple my-

eloma (2 cases) and non Hodgkin lymphoma (1 case).

The cytomorphology of aspirates in all these cases corre-

lated with the morphology of underlying hematological

malignancy. Most of these cases involved either scalp or

chest wall except one case which involved the arm.

There were 32 cases with unknown primary where cu-

taneous metastases were the first symptoms of disease.

They comprised of 19 males and 13 females (M:F ratio ¼1.5:1). The most common site for a cutaneous/subcutane-

ous metastasis in this group was also chest wall. The next

most common site was abdominal wall. Multiple site

involvement was also noted. The most common FNA di-

agnosis in this group was poorly differentiated carcinoma

(no further cytological diagnosis could be offered) fol-

lowed by adenocarcinoma and small round cell tumors

(Fig. 5). Immunochemistry was not available with us at

that time which could have helped to confirm the primary

tumor of origin.

Discussion

Skin is an uncommon site for development of metastases.

In a series reported by Saikia et al., out of a total 1022

metastatic malignancies at various sites, skin involvement

was observed in 58(5%) cases.7 In our series the incidence

of cutaneous metastasis was 0.17% (95/55556). In a similar

series by Gupta et al., an incidence of 0.5% was reported.

They had however, excluded pediatric age group in their

study.8 Furthermore, development of metastatic deposits in

skin indicates a very bad prognosis with rapid death. Cuta-

neous metastases from malignancy can arise either follow-

ing surgery or simultaneously with the tumor or first mani-

festation of unknown primary.6 They are mostly multiple

and rarely solitary. However in the present study, multiple

site involvement was seen in only 8 cases (9%). When soli-

tary, they may be confused with primary skin tumors.

They present as firm, nonulcerated skin nodules.9 The skin

metastases usually occur close to the site of primary tumor

for example chest in lung carcinoma, abdominal wall in

gastrointestinal tumors and lower back in renal cell carci-

noma.10 Tumor spread to regional skin is thought to be via

the lymphatics; subsequent spread to distant sites is due to

hematogenous spread.

The most common sites of metastases reported are

chest and abdomen, followed by head and neck.5,6 How-

ever in the present study the most common sites were

chest wall followed by abdominal wall and scalp. This

difference is probably due to larger number of cases from

lung and breast which are known to present with local

metastases via lymphatic involvement. The common

source of primary differs between the two sexes. In males

the most common sources are lung (25%), the large intes-

tine, melanoma, renal cell carcinoma and carcinoma of

the oral cavity.1,10 In our series, the most common pri-

mary in males was lung (10 cases) constituting 45% of

male patients with known primary. In females breast

accounts for large majority of cases (69%) followed by

lung, melanoma, kidney and ovary.1,10 In the present

study the most common source of primary in females was

also breast (21 cases), comprising 63% of female patients

with a known primary.

Cutaneous involvement in hematological malignancies

is a rare event. In the present study there were 5 such

cases. These included cases of chronic myeloid leukemia

(2 cases), multiple myeloma (2 cases) and non Hodgkin

lymphoma (1 case). FNA diagnosis was consistent with

the underlying hematological malignancy in all these

cases.

There were 32 cases in which there was no known pri-

mary at the time of FNAC. The overlying skin & underly-

ing muscle was normal and not fixed to the tumor.

Though it is difficult to differentiate metastatic adenocar-

cinoma from primary cutaneous adenocarcinomas, pres-

ence of pools of extracellular mucin, signet cells and

three dimensional papillae indicate metastasis rather than

primary lesion. On basis of these clinical features, multi-

plicity in few cases and morphology we suggested a diag-

nosis of metastasis. A definite primary could not be estab-

lished in these cases as they were lost to follow up.

Immunochemistry is known to be helpful in these cases in

identifying the primary tumor; however, it was not avail-

able to us at that time. Didolkar et al.11 and Osteen et

al.12 have reported that in many cases of cutaneous metas-

tasis, in spite of thorough clinical investigation and even

an autopsy, a primary may not be found.

Fig. 5. FNA smears from scalp swelling showing deposits of chronicmyeloid leukemia. MGG 3400. [Color figure can be viewed in theonline issue, which is available at www.interscience.wiley.com.]

FNAC OF CUTANEOUS METASTASES

Diagnostic Cytopathology, Vol 37, No 12 879

Diagnostic Cytopathology DOI 10.1002/dc

In conclusion FNAC is an invaluable tool by which

rapid diagnosis can be offered in cases presenting with

suspected cutaneous metastases. Similar views have been

expressed in previous studies also.4–6 If possible an

attempt to make a cell block should be made in these

cases. Immunohistochemistry can be performed on the

cell block to delineate the likely primary in cases with

unknown primary.

References

1. Reingold IM. Cutaneous metastases from internal carcinoma. Can-cer 1966;19:162–168.

2. Gates O. Cutaneous metastasis of malignant disease. Am J Cancer1937;30:718–730.

3. Abrams HL, Spiro N, Goldstein N. Metastasis in carcinoma. Cancer1950;3:76–85.

4. Pak HY, Foster BA, Yokota SB. The significance of cutaneous me-

tastases from visceral tumours diagnosed by fine-needle aspirationbiopsy. Diagn Cytopathol 1987;3:24–29.

5. Reyes CR, Jensen J, Eng AM. Fine needle aspiration cytology of

cutaneous metastases. Acta Cytol 1993;37:142–148.

6. Srinivasan R, Ray R, Nijhawan R. Metastatic cutaneous and subcu-

taneous deposits from internal carcinoma: An analysis of casesdiagnosed by fine needle aspiration. Acta Cytol 1993;37:894–898.

7. Saikia B, Dey P, Saikia UN. Fine needle aspiration cytology of

metastatic scalp nodules. Acta Cytol 2001;45:537–541.

8. Gupta RK, Naran S. Fine needle aspiration cytology of cutaneous

and subcutaneous metastatic deposits from epithelial malignancies.An analysis of 146 cases. Acta Cytol 1999;43:126–130.

9. Mckee PH. Cutaneous metastases. J Cutan Pathol 1985;12:235–250.

10. Brownstein MH, Helwig EB. Patterns of cutaneous metastases.Arch Dermatol 1972;105:862–868.

11. Didolkar MS, Fanous N, Elias EG, Moore RH. Metastatic carcino-mas from occult primary tumours: A study of 254 patients. AnnSurg 1977;186:625–630.

12. Osteen RT, Kopf G, Wilson RE. In pursuit of the unknown primary.Am J Surg 1978;135:494–498.

SHARMA ET AL.

880 Diagnostic Cytopathology, Vol 37, No 12

Diagnostic Cytopathology DOI 10.1002/dc