J Cutan Pathol 2009: 36: 859–864
John Wiley & Sons. Printed in Singapore
Copyright # 2008 John Wiley & Sons A/S
Continuing Medical Education Article
Diagnostic value of CD163 in cutaneous
spindle cell lesions
Background: The histologic diagnosis of atypical fibroxanthoma
(AFX) can sometimes be challenging. No specific marker exists to
confirm the diagnosis other than excluding other entities. CD163 has
been shown to have great specificity for tumors of monocyte/histiocyte
lineage. In this study, we evaluated the diagnostic utility of CD163 in
diagnosing AFX and in identifying skin lesions with histiocytic/
Methods: A total of 157 cases, including 14 AFXs, 5 spindle cell
squamous cell carcinomas (SCCs), and 7 spindle cell/desmoplastic
melanomas, along with other cutaneous spindle cell and histiocytic/
fibrohistiocytic lesions, were stained with CD163.
Results: CD163 was expressed in 11 of 14 (79%) AFXs, with
moderate to strong intensity. No staining was observed in cases of
spindle cell SCC (0/5) and dermatofibrosarcoma protuberans (0/10).
Rare spindle cell/desmoplastic melanomas (2/7) and cutaneous
leiomyosarcomas (1/5) demonstrated positive staining. CD163
reactivity was seen in 24 of 29 of benign fibrous histiocytomas (BFHs),
including 8 of 8 celular fibrous histiocytomas and 6 of 9 epithelioid cell
histiocytomas. The majority of cutaneous histiocytic lesions, including
juvenile xanthogranuloma, Langerhans cell histiocytosis and Rosai–
Dorfman disease, were positive for CD163.
Conclusion: CD163 is a useful adjunct in distinguishing AFX from
other malignant cutaneous spindle cell tumors and offers improved
specificity in identifying cutaneous histiocytic/dendritic lesions.
Pouryazdanparast P, Yu L, Cutlan JE, Olsen SH, Fullen DR, Ma L.
Diagnostic value of CD163 in cutaneous spindle cell lesions.
J Cutan Pathol 2009; 36: 859–864.#2008 John Wiley & Sons A/S.
Limin Yu1, Jonathan E. Cutlan1,
Stephen H. Olsen1,2, Douglas
R. Fullen1,2and Linglei Ma1,2
1Department of Pathology and
2Department of Dermatology, University of
Michigan, Ann Arbor, MI, USA
Linglei Ma, MD, PhD, Department of Pathology,
Dermatopathology Division, University of Michigan,
M3260, Medical Sciences 1, 1301 Catherine Road,
Ann Arbor, MI 48109-0602, USA
Tel: 11 734 764 4579
Fax: 11 734 764 4690
Accepted for publication September 11, 2008
The morphologic diagnosis of cutaneous malignant
spindle cell tumors, such as atypical fibroxanthoma
(AFX), spindle cell squamous cell carcinoma (SCC),
and spindle cell/desmoplastic melanoma, may some-
ical studies are often required to confirm or exclude
a diagnosis. Occasionally, dermatofibrosarcoma pro-
be included in the differential diagnosis. In general,
AFX is a diagnostic exclusion based on negative
immunoreactivity for cytokeratin and S-100 protein.
been widely used to confirm the diagnosis of AFX.
However, it has low specificity and may stain a wide
variety of neoplasms, including melanomas and
SCCs may show a lack of cytokeratin expression,3,4
and occasional spindle cell/desmoplastic melanomas
may fail to react with S-100 protein.5
As a result, new markers are being sought after to
diagnose AFX with more certainty. S100A6 is
expressed in up to 80% of AFXs as well as in 33%
of melanomas.6,7CD99 immunoreactivity was re-
ported in 73% of AFXs, with rare cases of melanoma
being positive, suggesting its value in diagnosing
AFX.8In a recent study, procollagen-1 was found to
reactive.9CD10 has been recently recommended
in the diagnosis of AFX, with positivity observed in
94–100% of tumors.10,11However, 33% of spindle
cell/desmoplastic melanomas and one half of spindle
cell SCCs also showed weak and patchy staining.11
The hemoglobin–haptoglobin scavenger receptor
restricted transmembrane glycoprotein of the scaven-
function as an endocytic receptor for iron homeosta-
sis, CD163 is also involved in anti-inflammatory
processes.12CD163 antigen is strongly expressed on
tissue macrophages, while weak expression is com-
monly seen on peripheral blood monocytes.1,2
Recently, a CD163 monoclonal antibody (clone
10D6) became commercially available for use in
paraffin-embedded tissue samples. It was shown to
label not only mature monocytes and macrophages
but also neoplastic cells in tumors of monocyte/
histiocyte lineage, such as acute myelomonocytic
leukemia.2When compared to other conventional
histiocytic markers, CD163 demonstrates greater
tissue specificity with no reactivity for carcinomas or
melanomas.1,2In a small series, CD163 expression
was found in one of three cases of AFX.2More
recently, CD163 was shown to be a useful marker in
distinguishing between cellular fibrous histiocytomas
and DFSPs, given its higher expression in the former
To date, the utilization of CD163 in a full range
of cutaneous spindle cell tumors and cutaneous
histiocytic/fibrohistiocytic lesions has not been sys-
temically examined. In this study, we proposed that
CD163 may be a useful marker in the diagnosis of
range of expression and tissue specificity. We also
tested the diagnostic utility of CD163 in identifying
skin lesions with histiocytic/dendritic origin.
Materials and methods
Case selection and tissue microarray construction
The study was approved by the Institutional Review
Board at the University of Michigan Health System
for human subject research. As listed in Table 1,
a spectrum of cutaneous spindle cell tumors as well as
benign histiocytic/fibrohistiocytic lesions and their
mimickers was identified through a retrospective
search of the pathology database from the Depart-
ment of Pathology at the University of Michigan.
Hematoxylin and eosin-stained sections, together
with relevant immunohistochemical stains, were
Table 1. CD163 expression in cutaneous spindle cell tumors and fibrohistiocytic lesions
Diagnosis (TMA and non-TMA)n
Percentage of CD163-positive cells
,10 (negative) 10–50
Spindle cell squamous cell carcinoma
Dermatofibroma (see variants below)
Epithelioid cell histiocytoma
Superficial acral fibromyxoma
Langerhans cell histiocytosis
Mast cell disease
Malignant fibrous histiocytoma
TMA, tissue microarray.
Pouryazdanparast et al.
reviewed by two dermatopathologists (L. M. and
D. R. F.) to confirm the original diagnoses. In
particular, all AFX cases included in the study were
diagnosed based on the histologic features and on the
exclusion of other entities through immunohisto-
were retrieved, and the diagnostic area of interest was
selected for the cylinder core. The tissue microarrays
(TMAs) were constructed using an ATA100 Tissue
described.16A total of 157 cases were included in this
study. Most of cutaneous spindle cell tumors and
histiocytic/fibrohistiocytic lesions (130 cases) were
sufficient tissue for evaluation were excluded from the
sarcoidosis and granuloma annulare, conventional
unstained slides were used.
paraffin blocks were stained with CD163 antibody.
with Ventana CC1 Solution (pH 8.0) for antigen
retrieval and were incubated with NCL-CD163
monoclonal antibody (clone 10D6, 1:50 dilution;
Vision Biosystems, Norwell, MA, USA) using a
Ventana Benchmark XT system (Ventana Medical
and a 3,3#-diaminobenzidine chromagen were used
Cytoplasmic and membranous staining was con-
sidered positive for CD163. According to the
percentage of positive lesional cells, the staining
was semiquantitatively recorded as ,10% (negative),
as weak, moderate or strong. Two pathologists (L. M.
and P. P.) independently scored the CD163 staining
for all cases.
Statistical analysis was carried out using SAS 8.2
software (SAS Institute Inc., Cary, NC, USA). The
Fisher’s exact test was used to assess the difference in
CD163 expression between AFX and other malig-
nant cutaneous spindle cell tumors. A p-value ,0.05
was considered statistically significant.
In normal skin, CD163 labeled the dermal dendritic
cells (data not shown). A cytoplasmic and membra-
nous staining in at least 10% lesional cells was
considered positive for CD163. When entrapped
intralesional dendritic cells or macrophages were
stained, it was not considered as true positivity for
lesional cells. The results of CD163 expression in
a number of skin lesions are summarized in Table 1.
Among malignant cutaneous spindle cell tumors,
79%), with all positive cases demonstrating moderate
to strong intensity (Fig. 1A). In contrast, none of the
spindle cell SCC (0/5) and DFSP (0/10) reacted with
CD163 (Fig. 1B, C). Most spindle cell/desmoplastic
melanomas were negative for CD163 (Fig. 1D), with
the exception of two cases (2/7; 29%) showing
moderate staining. One leiomyosarcoma (1/5) dem-
onstrated focal positivity with moderate intensity,
pattern in AFX was significantly different from other
that in malignant cutaneous spindle cell tumors,
including spindle cell SCC, spindle cell/desmoplastic
melanoma, DFSP and leiomyosarcoma (p , 0.05
for each). Interestingly, most cases (6/8; 75%) of
malignant fibrous histiocytoma (MFH) demonstrated
seen in AFXs.
As shown in Table 1, CD163 expression was
histiocytoma (BFH) regardless of the tumor subtypes
(Fig. 2A). Among 9 epithelioid cell histiocytomas
tested, the majority (6/9; 67%) reacted positively. Of
all positive BFHs, 20 of 24 displayed moderate to
strong staining (Fig. 2B). We noted that diffuse and
hemosiderotic BFH, lipidized BFH and epithelioid
cell histiocytoma, whereas sclerotic variant or pauci-
cellular areas in conventional BFH often had focal
in dermatomyofibromas (0/5), leiomyomas (0/8),
cellular neurothekeomas(0/2) orsuperficial acralfib-
10%), neurofibromas were not stained with CD163.
Most cases of fibrous papule (3/4) were positive, with
two showing weak staining intensity.
CD163 positivity was observed in all cases of juvenile
xanthogranuloma (9/9), reticulohistiocytoma (2/2),
Langerhans cell histiocytosis (3/3), Rosai–Dorfman
The majority of sarcoidosis (6/7; 86%) had strong
staining. All xanthomas (4/4) were positive for
CD163, with most (3/4) showing weak intensity.
No CD163 reactivity was observed in mimicking
lesions, including epithelioid sarcoma (0/3), Kaposi’s
sarcoma (0/5) and mastocytosis (0/6).
CD163 is a hemoglobin scavenger receptor and
functionsas anendocytic receptor-scavenging
Diagnostic value of CD163
differentiation. In skin, CD163 is typically expressed
in dermal macrophages and dendritic cells.1,13,14
Recently, CD163 expression has been found in
tumors with monocytic and histiocytic derivation,
such as acute myeloid leukemia with monocytic
differentiation and Rosai–Dorfman disease. In the
same study, none of the carcinomas or melanomas
were reactive.1,2Of interest, Nguyen et al. showed
while all melanomas (0/22) were non-reactive.2
cutaneous spindle cell tumors, such as spindle cell
SCC, were not investigated. Although their 22
melanomas were reported to be negative for
CD163, there was no detailed information regarding
the morphologic subtypes of these melanomas. In the
current study, we examined the expression of CD163
in more cases of AFX in comparison to its histologic
CD163 positivity with moderate to strong intensity,
while no cases of spindle cell SCC or DFSP showed
CD163 expression. Although a few cases of spindle
cell/desmoplastic melanoma (2/7) and cutaneous
leiomyosarcoma (1/5) showed some staining, the dif-
ference in CD163 expression between AFX and other
malignant cutaneous spindle cell tumors was statisti-
cally significant (p , 0.05). Our findings suggest that
CD163 is a useful marker in separating AFX from
immunohistochemical panel for malignant cutaneous
spindle cell tumors. Regarding the occasional cases of
is probably because of antigen cross-reactivity.
The diagnosis of AFX is, by definition, dependent
on the use of a wide panel of immunohistochemical
markers. It is invariably negative for keratins, S-100
protein and desmin. In the past, there have been
efforts toward establishing a confirmatory stain for
Fig. 1. CD163 staining in cutaneous malignant spindle cell tumors. A) Atypical fibroxanthoma with diffuse and strong CD163 staining in
tumor cells. There is a lack of CD163 reactivity in spindle cell squamous cell carcinoma (B), dermatofibrosarcoma protuberans (C) and spindle
cell/desmoplastic melanoma (D) (original magnification, 3200).
Pouryazdanparast et al.
AFX. Markers such as CD68, lysozyme, alpha-1-
antitrypsin and alpha-1-antichymotrypsin are com-
monlyexpressed in AFX but have low specificity.17,18
A number of recent studies have offered several
potentially useful diagnostic markers, including
CD99, procollagen-1 and CD10.8,10,11The fact that
CD163 expression was nearly restricted in AFX
makes it another useful adjunct in the diagnosis of
AFX. However, because of the presence of CD163
reactivity in a small subset of spindle cell/desmo-
plastic melanomas and cutaneous leiomyosarcomas,
special caution should be taken to avoid diagnosing
AFX based exclusively on CD163 positivity. In add-
ition, it should be made aware that some entrapped
CD163-positive dendritic cells/macrophages may be
seen within the tumor, and these cells should not be
interpreted as positive tumor cells. As the CD163
positivity was observed in both AFX and MFH, it is
not helpful in separating the two entities.
Earlier work has shown that a subset of BFH
reacted with CD163.2More recently, Sachdev and
Sundram15found that CD163 was expressed in the
majority of BFH (89%) and cellular fibrous histiocy-
toma (100%), but only in a small number of DFSP
(17%, 3/18), indicating its potential utility in the
differential diagnosis of cellular BFH and DFSP. Our
data confirm and expand their results. We found that
the cellular variant. None of our DFSPs (0/10) were
labeled with CD163. Both histiocytic and spindle cell
components were labeled positively, similar to that
reported in a previous study.15Intriguingly, we
observed that 67% of epithelioid cell histiocytomas
reacted diffusely and intensely with CD163. Nguyen
stated that, in their experience (unpublished obser-
vation), the main cellular component of epithelioid
cell histiocytoma is negative for CD163. Histologi-
cally, epithelioid cell histiocytoma resembles lesions
comprising epithelioid cells, such as Spitz nevi and
potential pitfalls exist in that occasional cases of
Fig. 2. CD163 expression in benign cutaneous histiocytic/fibrohistiocytic lesions. A) Conventional dermatofibroma with moderate staining
intensity in tumor cells. B) Epithelioid cell histiocytoma showing diffuse and strong CD163 reactivity. C) Dermatomyofibroma displaying
negative staining. D) Juvenile xanthogranuloma with diffuse labeling in lesional cells (original magnification, 3200).
Diagnostic value of CD163
epithelioid cell histiocytoma may have focal S-100
protein positivity and/or lack the expression of factor
XIIIa.20Our data suggest that CD163 can aid with
the diagnosis in this settings.
In this study, we also tested a wide spectrum of
cutaneous histiocytic/fibrohistiocytic lesions. CD163
reactivity was not observed in cases of dermatomyofi-
broma, cellular neurothekeoma or superficial acral
fibromyxoma, a novel finding not previously re-
ported. In concordance with earlier studies,1,2,19we
found CD163 expression in all lesions of juvenile
Dorfman disease and Langerhans cell histiocytosis.
As expected, most histiocytic lesions, such as granu-
CD163. In contrast, epithelioid sarcoma, mastocy-
tosis and Kaposi’s sarcoma were CD163 negative.
Our work further supports the notion that CD163 is
a relatively specific marker for lesions of histiocyte/
dendrocyte derivation and expands the utility of
CD163 in diagnosing skin lesions.1,2
The potential histogenesis of AFX has been
controversial. Some investigators have proposed that
AFX may represent a reactive process, while others
contend that it is a fibrohistiocytic neoplasm, closely
related to MFH. It is variously regarded as being of
myofibroblastic, fibrohistiocytic and fibroblastic dif-
ferentiation.22,23The expression of CD163 in AFX
suggests that it may, in part, be related to histiocytic/
dendritic lineage. Alternatively, AFX lesions may
dendrocyte through partial differentiation. Nonethe-
less, continued investigation is necessary to unravel
the true histogenesis of AFX.
In summary, CD163 appears to be a useful marker
in identifying cutaneous lesions of histiocytic/den-
dritic lineage. Our data suggest that CD163 may be
AFX. It appears helpful in separating BFH from
DFSP and particularly in assisting with the diagnosis
of epithelioid cell histiocytoma. Future research with
larger study populations is necessary to establish the
utility of CD163 in diagnosing malignant cutaneous
spindle cell tumors.
1. Lau SK, Chu PG, Weiss LM. CD163: a specific marker of
macrophages in paraffin-embedded tissue samples. Am J Clin
Pathol 2004; 122: 794.
2. Nguyen TT, Schwartz EJ, West RB, Warnke RA, Arber DA,
Natkunam Y. Expression of CD163 (hemoglobin scavenger
receptor) in normal tissues, lymphomas, carcinomas, and
sarcomas is largely restricted to the monocyte/macrophage
lineage. Am J Surg Pathol 2005; 29: 617.
3. Sigel JE, Skacel M, Bergfeld WF, House NS, Rabkin MS,
Goldblum JR. The utility of cytokeratin 5/6 in the recognition
of cutaneous spindle cell squamous cell carcinoma. J Cutan
Pathol 2001; 28: 520.
4. Prieto VG, Lugo J, McNutt NS. Intermediate- and low-
molecular-weight keratin detection with the monoclonal
antibody MNF116. An immunohistochemical study on 232
paraffin-embedded cutaneous lesions. J Cutan Pathol 1996;
5. Blessing K, Sanders DS, Grant JJ. Comparison of immunohis-
tochemical staining of the novel antibody melan-A with S100
protein and HMB-45 in malignant melanoma and melanoma
variants. Histopathology 1998; 32: 139.
6. Ribe A, McNutt NS. S100A6 protein expression is different in
Spitz nevi and melanomas. Mod Pathol 2003; 16: 505.
7. Fullen DR, Reed JA, Finnerty B, McNutt NS. S100A6 expres-
sion in fibrohistiocytic lesions. J Cutan Pathol 2001; 28: 229.
8. Monteagudo C, Calduch L, Navarro S, Joan-Figueroa A,
Llombart-Bosch A. CD99 immunoreactivity in atypical fibrox-
anthoma: a common feature of diagnostic value. Am J Clin
Pathol 2002; 117: 126.
9. Jensen K, Wilkinson B, Wines N, Kossard S. Procollagen 1
expression in atypical fibroxanthoma and other tumors.
J Cutan Pathol 2004; 31: 57.
10. Weedon D, Williamson R, Mirza B. CD10, a useful marker for
atypical fibroxanthomas. Am J Dermatopathol 2005; 27: 181.
11. Hultgren TL, DiMaio DJ. Immunohistochemical staining of
CD10 in atypical fibroxanthomas. J Cutan Pathol 2007; 34: 415.
12. Kristiansen M, Graversen JH, Jacobsen C, et al. Identification
of the haemoglobin scavenger receptor. Nature 2001; 409: 198.
13. Ochoa MT, Loncaric A, Krutzik SR, Becker TC, Modlin RL.
‘‘Dermal dendritic cells’’ comprise two distinct populations:
CD1(1) dendritic cells and CD209(1) macrophages. J Invest
Dermatol 2008; 128: 2225.
14. Zaba LC, Fuentes-Duculan J, Steinman RM, Krueger JG,
Lowes MA. Normal human dermis contains distinct populations
of CD11c1BDCA-11 dendritic cells and CD1631FXIIIA1
macrophages. J Clin Invest 2007; 117: 2517.
15. Sachdev R, Sundram U. Expression of CD163 in dermatofi-
broma, cellular fibrous histiocytoma, and dermatofibrosarcoma
protuberans: comparison with CD68, CD34, and Factor XIIIa.
J Cutan Pathol 2006; 33: 353.
16. Nocito A, Kononen J, Kallioniemi OP, Sauter G. Tissue
microarrays (TMAs) for high-throughput molecular pathology
research. Int J Cancer 2001; 94: 1.
17. Pernick NL, DaSilva M, Gangi MD, Crissman J, Adsay V.
18. McHugh M, Miettinen M. KP1 (CD68): its limited speci-
ficity for histiocytic tumors. Appl Immunohistochem 1994;
19. Miettinen M, Fetsch JF. Reticulohistiocytoma (solitary epithe-
lioid histiocytoma): a clinicopathologic and immunohistochem-
ical study of 44 cases. Am J Surg Pathol 2006; 30: 521.
20. Glusac EJ, McNiff JM. Epithelioid cell histiocytoma: a simulant
of vascular and melanocytic neoplasms. Am J Dermatopathol
1999; 21: 1.
21. Zelger BG, Steiner H, Kutzner H, Maier H, Zelger B. Cellular
?neurothekeoma’: an epithelioid variant of dermatofibroma?
Histopathology 1998; 32: 414.
22. Longacre TA, Smoller BR, Rouse RV. Atypical fibroxanthoma.
Multiple immunohistologic profiles. Am J Surg Pathol 1993;
23. Ma CK, Zarbo RJ, Gown AM. Immunohistochemical charac-
terization of atypical fibroxanthoma and dermatofibrosarcoma
protuberans. Am J Clin Pathol 1992; 97: 478.
Pouryazdanparast et al.