DOI:10.1111/j.1365-2303.2011.00885.

Immunocytochemistry in Europe: results of the European

Federation of Cytology Societies (EFCS) inquiry

F. Schmitt*, B. Cochand-Priollet , M. Toetschà, B. Davidson§, A. Bondi– and P. Vielh**

*Medical Faculty of Porto, University and Institute of Molecular Pathology and Immunology, University of Porto, Porto,
Portugal, Department of Pathology and Cytopathology, Lariboisiere Hospital, University Paris 7, Paris, France, àInstitut für
Pathologie und Neuropathologie, University of Essen, Essen, Germany, §Department of Pathology, Norwegian Radium Hospital
and Faculty of Medicine, University of Oslo, Oslo, Norway, –Department of Pathology, Ospedale Maggiore, Bologna, Italy and
**Department of Pathology, Institut de Cancerologie Gustave Roussy, Villejuif, France

Accepted for publication 30 May 2011

F. Schmitt, B. Cochand-Priollet, M. Toetsch, B. Davidson, A. Bondi and P. Vielh

Immunocytochemistry in Europe: results of the European Federation of Cytology Societies (EFCS) inquiry
Objective: This report describes the results of the first chosen topic of the European Federation of Cytology
Societies (EFCS) scientific committee, which concerns the application of immunocytochemistry (ICC) to
cytological material. While ICC has become an important ancillary method, not only for diagnosis but also to
assess prognostic and predictive factors on cytological material, there are many different methodologies used and
the lack of standardization has been criticized. This inquiry aimed first to obtain an overview of the techniques
used and second to suggest mechanisms for standardization and quality control.
Methods: We report the results of 28 replies from 13 countries to a web-based inquiry into types of specimens,
preparations and technical methods.
Results: Conventional smears were the preparations most commonly used, followed by cytospins, cell blocks
and liquid-based preparations, in that order. Avidin-biotin complex, labelled streptavidin-biotin or peroxidase
anti-peroxidase were used in 61% and enhanced-polymer technology in 39%. Automated staining techniques
were used by 64%. More than half used the same antibody dilutions as histopathology although only 20% used
cell blocks.
Conclusion: Reduction of variability by using automation, appropriate controls and customized dilution of
antibodies according to different samples could improve the quality of ICC and standardize the techniques, along
with quality control and quality assurance measures.

Keywords: immunocytochemistry, European inquiry, European Federation of Cytology Societies, cytological

preparations, quality control

increasingly important role in the diagnosis of various

Introduction
The European Federation of Cytology Societies (EFCS)
Scientific Committee was created in June 2006 and
one of its main tasks was to foster some collaborative
European scientific studies. The application of immu-
nocytochemistry (ICC) to cytological material was the
first chosen scientific topic. Cytology has played an
Correspondence: F. Schmitt, Medical Faculty of Porto,
University and Institute of Molecular Pathology and Immu-
nology, University of Porto, Porto, Portugal
Tel.: +351 225570700; Fax: +351 225570799;
E-mail: fernando.schmitt@ipatimut.pt
disease processes, particularly those of neoplastic
origin. In fact, it is not unusual for cytological
specimens to be the only diagnostic sample available
from patients with cancer. We are now expected to
perform on cytological specimens many ancillary tests
that are traditionally performed on histological mate-
rial. ICC has become an important ancillary method
not only for diagnosis but also to assess prognostic and
predictive factors on cytological material.1,2 However,
in contrast to immunohistochemistry (IHC), published
papers on the use of ICC have shown that there are
different methodologies used and have questioned the
lack of standardization in the application of this

238 Cytopathology 2011, 22, 238–242 ª 2011 Blackwell Publishing Ltd


technique to cytological material.2–4 There are many
potential reasons for the variability of results, such as
the use of improper controls or the absence of
controls, the use of non-customized antibody concen-
trations, and different methodologies of fixation and
preparation of the material, among others. Cytopa-
thologists should avoid automatically applying stain-
ing conditions used in histological material to
cytological material. Unfortunately, this quick jump
from a histological section to a cytological specimen
using ICC is frequently done without consideration of
the major differences in specimen preparation that
can affect the final interpretation of the reaction, with
potential clinical consequences for the patient.3
After several discussions in scientific sessions of
European congresses of pathology and cytology, it
became clear that there was a broad and different use
of ICC in different laboratories across Europe. So, the
goal of our scientific committee was first to obtain an
overview about the ICC techniques used in Europe
and as a second step to suggest mechanisms of
standardization and quality control.
Methods
An inquiry, including questions about the ICC tech-
nical steps as well as the cytological material subjected
to this technique, was discussed during the EFCS
working group party in 2007 and broadcast on the
EFCS website in 2008. All the representatives of the
EFCS were informed of this inquiry during the 34th
European Congress of Cytology (ECC) in Rovaniemi,
Finland (June 2008). Most of the information has
been obtained from expert cytopathologists informed
about this survey by their national societies. We had
more than 2000 accesses to the inquiry at the EFCS
website (http://www.efcs.eu). We obtained 42 at least
partially complete answers, 28 of which applied to
named countries: Belgium (3), Czech Republic (1),
France (6), Germany (2), Greece (2), Italy (3), Norway
(2), Portugal (1), Russia (1), Slovenia (1), Sweden (2),
Switzerland (1) and United Kingdom (3); 14 have
been excluded either because the answers were very
incomplete, or because they did not identify the
country where the answer was coming from.
Results
Of 42 total responses to the questionnaire shown in
Table 1, 22 (52%) said ICC was used routinely in the
laboratories, and applied to 6–30% of cytological
Cytopathology 2011, 22, 238–242 ª 2011 Blackwell Publishing Ltd
Immunocytochemistry in Europe 239
material available in these laboratories. The most
frequent types of cytology specimens to which ICC
was applied were: lung cytology (including material
from bronchoalveolar lavage), fine needle aspiration
(FNA) of the breast, thyroid or deep organs, serous
fluids, and urines. The lack of results concerning
paediatric and neuropathological samples may be due
either to the fact that these may be dealt with in
specialist departments not reached by the inquiry and
many general laboratories may not handle these
specimens or to the fact that IHC is more likely to be
applied on the accompanying biopsy. The six main
antibodies routinely used were: cytokeratins (CK7
was specifically mentioned in 21 inquiry answers),
oestrogen and progesterone receptors (ER, PR), leuko-
cyte common antigen (CD45), CD 20 and CD3,
reflecting common diagnostic dilemmas. Diam-
inobenzidine substrate (DAB) as chromogen (95%)
was used in the majority of laboratories for ICC on
cytological material. In contrast, the results for the
other items were more variable: these results are
summarized in Table 2. Finally, 52% of the laborato-
ries used the same antibody dilution for IHC and
immunocytochemistry, and 48% did not.
Discussion
Apart from discussions at the EFCS scientific commit-
tee at several European congresses, we have presented
28 of 42 responses from 13 countries to an inquiry on
the EFCS website. We have discussed the possible
reasons for the low number of answers: the fact that
2000 visitors accessed the survey may be explained by
a high number of accidental visits and ⁄ or repeated
accesses and ⁄ or visitors interested in the survey but
having no time to answer.
This survey, as expected, showed wide variation in
the use of ICC among different laboratories across
Europe. As known, ICC can be performed on most
cytology samples, including FNA, serous fluid, wash
and brush specimens and even on Pap smears.3,4
These different types of cytology specimens may be
received in the laboratory for morphological evalua-
tion as smears, or centrifuged or liquid-based prepa-
rations, or sometimes can be prepared as cell blocks.
All these different preparations may be processed
using different fixatives, such as alcohol-based fixa-
tive, methanol-based fixative or formalin. So, the
variability in slide preparation is reflected in the ICC
technique and this variability may also exist within
the same laboratory. In contrast to histology, where
240 F. Schmitt et al.

Table 1. Inquiry questionnaire

Questions Available answers

Do you use immunocytochemistry in your laboratory? u Never u On some selected cases exclusively
u Routinely (on all cases of cytology, when necessary
for the diagnosis)
If you apply immunocytochemistry only on some selected u Do you apply immunocytochemistry only on cytological
cases, please answer these questions? cases included in a clinical research programme?
u Do you apply immunocytochemistry only on selected
cases of high interest for you? What is (are) this
(these) topic(s)? Please give your comments below
If you apply immunocytochemistry routinely, please
answer the following questions (if there is more than
one option please mark)
For what types of cytology do you apply u Pap smears u Thyroid FNA u Deep organs FNA
immunocytochemistry? u Urines u BAL
u Breast FNA u Brain smears u Bronchogenic specimens
u CSF u Serous fluids
u Paediatric cytology
u Other
In order to appreciate your activity in cytopathology, u Number of cytological specimens ⁄ histological specimens
please give us some necessary information u Number of people involved in cytopathology in your laboratory
u Cytopathologists
u Cytotechnicians
u Percentage of immunocytochemistry ⁄ cytological specimens
What is (are) the most frequent type of cytological u Conventional smears
preparations used in your laboratory? u Centrifugation or cyto-centrifugation
u Liquid-based cytology (please add the name of the technique)
u Cell block
u Other
What fixative do you recommend for cytological u Methanol u 10% buffer formalin
specimens? u Methanol ⁄ acetone u Paraformaldehyde (specify %)
u Ethanol-based fixative u Other (specify)
What kind of epitope retrieval do you use? u Heat epitope retrieval (specify) u Other (specify)
u Enzymatic epitope retrieval (specify)
What type of immunocytochemistry technique is u Automated technique (please give the name of the device)
yours? u Manual technique
What type of immunocytochemistry detection system is u ABC method u Enhanced polymer technology
yours? u PAP method u Other
(please give the name of the suppliers)
u APAAP
u LSAB
What substrate ⁄ chromogen do you use? u DAB u AEC
u Fast Red u Other
What kind of slides do you use? u SuperFrost + u Self-silanated slides
u Polysin u Other (specify)
Do you use the same antibody u Yes u No
dilutions for immunocytochemistry as for
immunohistochemistry?
What are the six main antibodies you routinely use? Please give the name of the antibodies and the supplier
[for example: HBME1 (Menarini)]

FNA, fine needle aspirate; BAL, bronchoalveolar lavage; CSF, cerebrospinal fluid; ABC, avidin-biotin complex; PAP, peroxidase anti-
peroxidase; APAAP, alkaline phosphatase anti-alkaline phosphatase; LSAB, labelled streptavidin-biotin; DAB, diaminobenzidine; AEC,
aminoethylcarbazole.

almost all the IHC studies are carried out in formalin-
fixed paraffin-embedded specimens, protocols for ICC
largely depend on specific procedures in each labora-
tory, trying to adapt and optimize techniques locally.
It is well known that slide preparation and fixation
technique has a great impact on the quality of ICC
reactions. However, equally good ICC results can be
achieved in cytological samples fixed in 95% alcohol,
buffered formalin, methanol or ethanol-based fixa-
tives.2–6 Brief fixation in any of the above fixatives is
adequate for most cytological preparations.
The variability in material and fixatives is, in our
opinion, a major factor preventing standardization of
some procedures using the technique. The recent
publications from the American Society of Clinical
Oncology and College of American Pathologists con-

Cytopathology 2011, 22, 238–242 ª 2011 Blackwell Publishing Ltd

 Immunocytochemistry in Europe 241

Table 2. Results of the inquiry (28 responses from 13 concern is the fact that still many laboratories are using
countries) methods of detection that are considered to have low
Cytological preparation (more than one option allowed)
sensitivity, such as avidin-biotin complex (ABC) and
Conventional smears 23 peroxidase anti-peroxidase (PAP) methods. There is
Cytospins 19 ample evidence from practical observations in different
Cell blocks 13 laboratories that the enhanced polymer technology
Liquid-based preparations 11 should be adopted as standard in ICC reactions.
Fixative Characterization of lymphoid cells (CD45, CD20
Ethanol or ethanol based 12 and CD3), study of markers of therapeutic response
10% buffered formalin 7 (ER and PR) and characterization of epithelial cells
Methanol based 5 (CK profiles) are common situations in which we use
Slides
ICC, as reflected by the antibodies used most often in
Super-frost 15
cytological material in this inquiry. A reasonable
Silane-coated 7
Polylysine-coated 5
differential diagnosis is usually based on the cytomor-
Retrieval phology, clinical information and the probability of
Heat-epitope 20 certain disease occurring according to the age group of
Enzymatic-epitope 7 the patient and the anatomical location sampled.
No retrieval 1 Another important factor is the availability of markers
Detection for the entity within the differential diagnosis. Perhaps
Avidin-biotin complex 7 the most essential factor in determining the clinical
Labelled streptavidin-biotin 6 value of ICC is the experience of the cytopathologist.
Peroxidase anti-peroxidase 4 This has a major impact on several steps, from the
Enhanced polymer technology 11
selection of the markers to the evaluation of results
Staining method
and rendering of a diagnosis. Because most diagnostic
Manual 10
Automated 18
problems in cytology can be narrowed down to a few
possibilities, the choice of antibodies can also be
restricted to two or three. This approach requires the
input of the cytopathologist and necessitates active
cerning guidelines for interpretation of HER2 and participation in the formulation of a working diagno-
hormonal receptors in histological samples are good sis. Considering the limited amount of tissue in
examples of how it is possible to have a framework cytological specimens, large pre-determined panels
from where we can adopt some procedures that of antibodies should be avoided and a judicious choice
enhance the reproducibility of the technique.7,8 Cur- of antibodies based on morphological differential
rently, ICC methods have been refined and high- diagnosis, combined with the pertinent clinical infor-
quality reagents and automation are more widely mation, should be encouraged. This last issue is crucial
available, so it is expected that technical problems will when dealing with cytological samples.
be reduced in the near future. In fact, our inquiry One surprising and unexpected answer that we
demonstrates that more than half of the laboratories received during the inquiry was that half of the
already use automated techniques. The need for laboratories used antibodies at the same dilution for
automation in IHC was raised many years ago.9 cytology and histology. These results may be partially
Automated ICC offers the opportunity to reach new explained by the use of cell blocks and therefore this
levels of quality, reproducibility and consistency while may be a slight over-estimation of this apparent poor
reducing labour and reagent costs. In addition, the practice. Nevertheless, considering that conventional
controlled environment of automated ICC may smears and cytospins are mostly used, it appears
increase the speed and timeliness of results. Automa- obvious that this practice is applied to cytology as well.
tion contributes decisively to the standardization of As pointed out in the introduction, cytopathologists
ICC techniques and improves the confidence of the should avoid the assumption that conditions applied
pathologist in the ICC reactions. So, the fact that the to histological material are appropriate for cytological
majority of laboratories are using automated ICC on material.3 When applying ICC to a non-paraffin
cytological samples is good news and a decisive step formalin-fixed specimen, the application of non-cus-
towards standardization. However, one point of tomized antibody concentrations and the use of

Cytopathology 2011, 22, 238–242 ª 2011 Blackwell Publishing Ltd

242 F. Schmitt et al.
improper control specimens are the most common
mistakes in cytology laboratories.3 The antibody con-
centration used on histological sections may be much
stronger than needed for a cytological sample (either
formalin-fixed or paraffin-embedded), resulting in
potential false-positive results due to excess anti-
body.2–4 The same is true for retrieval procedures. In
general, retrieval may not be necessary for specimens
that are neither paraffin embedded nor formalin fixed,
or may be less aggressive.5 In view of the fact that cell
blocks are the cytological samples that are closest to
histological sections, some authors consider them as
ideal specimens for ICC, requiring fewer needs for
adaptation. Finally, another important point that was
not addressed in the inquiry is the use of controls. For
quality assurance, controls must be of similarly
prepared material. This is usually difficult for the
laboratory, with respect to obtaining sufficient variety
and number of controls, as well as meeting storage
requirements of cytology specimens.3,10 Although
preparation and storage of various cytological samples
to be used exclusively for ICC may not be practical,
the use of cell lines prepared in the same way as the
sample being tested can be used to overcome this
obstacle.11 The use of histological sections as positive
controls for cell blocks may be acceptable.
In conclusion, the results of our inquiry provide an
overview of the diversity of the use of ICC in 13
different countries across Europe. Reduction of this
variability by using automation, appropriate controls
and customized dilution of antibodies according to
different samples is likely to improve the quality of ICC
and will lead to standardization for the technique.
However, it is equally or even more important to
standardize criteria for high-quality ICC reactions and
to encourage all laboratories performing this technique
to use proper quality control and quality assurance
measures. The aim of this special issue of Cytopathology is
to support the EFCS project and to motivate cytopa-
Cytopathology 2011, 22, 238–242 ª 2011 Blackwell Publishing Ltd
thologists to carry out protocols for standardization and
quality control in their laboratories.
References
1. Schmitt FC. Cells carry the clue for targeted treatment: a
new horizon for cytopathology. Cytopathology 2007;18:
275–7.
2. Akerman M, Alves VA, Bubendorf L et al. How technol-
ogy is reshaping the practice of nongynecologic cytology.
Acta Cytol 2007;51:123–52.
3. Fowler LJ, Lachar W. Application of immunohistochem-
istry to cytology. Arch Pathol Lab Med 2008;132:373–83.
4. Leong AS. Immunostaining of cytologic specimens. Am J
Clin Pathol 1996;105:139–40.
5. Cano G, Milanezi F, Leitão D et al. Estimation of
hormone receptor status in fine-needle aspirates and
paraffin-embedded sections from breast cancer using the
novel rabbit monoclonal antibodies SP1 and SP2. Diagn
Cytopathol 2003;29:207–11.
6. Freitas C, Milanezi F, Dias AJ, Bento MJ, Schmitt FC. Use
of cell block preparation for morphological, immunocy-
tochemistry, and ploidy analysis in Thinprep monolayer
preparations. Diagn Cytopathol 2001;25:415–6.
7. Wolff AC, Hammond ME, Schwartz JN et al. American
Society of Clinical Oncology ⁄ College of American
Pathologists guideline recommendations for human epi-
dermal growth factor receptor 2 testing in breast cancer.
Arch Pathol Lab Med 2007;13:18–43.
8. Hammond ME, Hayes DF, Dowsett M et al. American
Society of Clinical Oncology ⁄ College of American
Pathologists guideline recommendations for immunohis-
tochemical testing of estrogen and progesterone recep-
tors in breast cancer (unabridged version). Arch Pathol
Lab Med 2010;134:48–72.
9. Grogan TM. Automated immunohistochemical analysis.
Am J Clin Pathol 1992;98(4 Suppl 1):S35–8.
10. Wick MR, Swanson PE. Targeted controls in clinical
immunohistochemistry. Am J Clin Pathol 2002;117:7–8.
11. Schmitt FC, Dias AJ, Wagner R. Use of Thinprep
processing for studying breast cancer cell lines. Diagn
Cytopathol 2000;23:27–8.