Original Article

Young Investigator Challenge: A Novel, Simple Method

for Cell Block Preparation, Implementation, and Use Over
2 Years
Kathryn G. Lindsey, MD; Patricia M. Houser, MHS, CT (ASCP); Wanda Shotsberger-Gray, HT/HTL (ASCP);
Olga S. Chajewski, MD; and Jack Yang, MD

BACKGROUND: The cell block is an essential adjunct to conventional cytopreparatory techniques. The need for molecular
analysis and immunostains will increase the need for successful cell block preparation. Even with this need, to the authors’
knowledge very little has changed regarding the way in which cell blocks are produced. METHODS: The authors devel-
oped A Formalin-Fixed, paraffin Embedded Cytology cell block Technique (AFFECT) that uses a cytospin centrifuge and
funnel to deposit a cell pellet into a well on a piece of open-cell, absorbent foam. The foam and the pellet are then sent
through normal processing. Herein, the authors present the implementation of this method and some of their experience
with its performance over the course of 2 years. RESULTS: Although a comparison of the methods indicated good corre-
lation for the production of a cell block between AFFECT and the agarose method, the AFFECT blocks demonstrated
markedly improved cellular morphology. Over the first 6 months of use, AFFECT produced a successful cell block in 74%
of cases overall, and in 65% of cases with a cell pellet measuring 0.1 mL. The year preceding the implementation of
AFFECT and its first year of use were compared for endoscopic and bronchoscopic ultrasound-guided fine-needle aspira-
tion specimens, and demonstrated an improved success rate. CONCLUSIONS: The authors developed a novel method of
cell block preparation that demonstrates improved histology and has increased the success rate of cell block production
compared with the agarose method. Cancer Cytopathol 2016;124:885-92. V
C 2016 American Cancer Society.

KEY WORDS: agarose method; cell block; fine-needle aspiration; immunostains; molecular analysis.

INTRODUCTION
The cell block is an essential adjunct to conventional cytopreparatory techniques, and when successful, yields mul-
tiple histologic sections for subsequent special and immunocytochemical staining to classify tumors. In addition,
as treatments have become more tumor-directed, cell blocks can provide tumor cells for molecular diagnostics to
determine treatment regimens.
The history of the cell block begins in 1896, when the first cell block was formed from the sediment of asci-
tes fluid that was embedded in celloidin and sectioned on a microtome.1 Celloidin is a highly flammable nitrocel-
lulose compound,2 similar to collodion, which still is used in the collodion bag method of cell preparation.3 The
next major advance was the use of a centrifuge in 1901.4 Shackelford and Jones reported the use of agar in 1959,5
and Kaltenbach et al introduced the plasma-thrombin method in 1973.6 Hologic Inc (Marlborough, Mass) intro-
duced what to our knowledge was the first automated cell block preparation system in 2007 with the Cellient
Automated Cell Block System. Amazingly, the processes and technology used, until introduction of the Cellient
system, had changed relatively little.
Corresponding author: Kathryn G. Lindsey, MD, Department of Pathology and Laboratory Medicine, Medical University of South Carolina, 165 Ash-
ley Ave, Ste 309, MSC 908, Charleston, SC, 29425; Fax: (843) 792-8974; lindseyK@musc.edu

Department of Pathology and Laboratory Medicine, Medical University of South Carolina, Charleston, South Carolina

Received: August 1, 2016; Revised: September 23, 2016; Accepted: October 14, 2016

Published online November 9, 2016 in Wiley Online Library (wileyonlinelibrary.com)

DOI: 10.1002/cncy.21795, wileyonlinelibrary.com

Cancer Cytopathology December 2016 885

Original Article

Studies have indicated that many laboratories are

unsatisfied with their technique for creating cell blocks.7
Some of the frequently encountered issues include the
techniques having a high cost, being labor-intensive, and
yielding less than adequate cellular material for diagnosis
or ancillary testing. Some techniques use fixatives other
than formalin, which may interfere with immunohisto-
chemical (IHC) studies.8 The automated method currently
available is expensive and time-consuming, produces lim-
ited numbers of slides, and uses alcohol as the fixative. To
resolve these issues, we developed a formalin-fixed, paraf-
fin-embedded cytology cell block technique, AFFECT, to
make cell blocks. The method is simple and efficient, and
shows improved cytomorphology. In the current study, we
have described this novel method and summarized our
experience with its implementation in the study laboratory
Figure 1. A 5-mm well is made in a circle of foam measuring
over a period of 2 years. 1 cm in diameter. The foam receptacle is placed in line with
the outlet of the funnel. A single piece of BIO-WRAP tissue
MATERIALS AND METHODS paper is placed behind the foam. The complex is held
together with the metal clip.
The equipment and materials used to develop the
AFFECT technique include a regular centrifuge (Centra
CL2; Thermo Electron Corporation, Waltham, Mass), a
allowed to fix for 1 to 2 hours. The tube is centrifuged
cytospin centrifuge (Shandon Cytospin 4; Thermo Fisher
again at 2700 RPM for 5 minutes and the formalin super-
Scientific, Waltham, Mass), a cytospin funnel, and absorb-
natant fluid is again poured off, leaving 1 mL of formalin
ent foam (Infinicel; Procter & Gamble Company, Cincin-
and specimen in the tube. Using the vortex, the cell pellet
nati, Ohio). Before specimen preparation, a cytospin
is reconstituted into 1 mL of solution. Approximately
funnel and metal clip are assembled as follows. A piece of
0.5 mL of the specimen is aspirated and transferred into
Infinicel foam measuring 1 cm in diameter is prepared the cytospin funnel assembly as described above.
using a 1-cm biopsy punch. A well is made in the foam The assembled funnel, specimen, and clip are placed
with a 5-mm biopsy punch. The foam receptacle is placed in the cytospin centrifuge. The specimen is centrifuged for
at the outlet of the cytospin funnel, a single piece of BIO- 5 minutes at 1000 RPM (relative centrifugal force, 113).
WRAP tissue paper (Leica Biosystems Richmond Inc, Upon removal from the centrifuge, the clip is opened and
Richmond, Ill) is placed behind the foam, and a filter card the foam with the cell pellet in the well is wrapped in the
may be placed behind the tissue paper to absorb excess for- tissue paper and submitted for routine histologic process-
malin. The complex is held together with the metal clip ing. After processing, the foam is inverted, embedded in
(Fig. 1). paraffin wax, and sectioned at 4 mm (Fig. 2).
Cytology specimens are transferred to 50-mL plastic Because the method previously used by the study lab-
conical tubes. Fluid specimens are received fresh in sterile oratory was the agarose method of Varsegi and Shidham,9
containers and fine-needle aspiration (FNA) needle rinses this method was used to prepare cell blocks in parallel to
are collected in a buffered saline solution. The tubes then AFFECT for comparison. Cellularity and cytomorphology
are centrifuged at 2700 revolutions per minute (RPM) for on hematoxylin and eosin-stained and, in some cases,
5 minutes (relative centrifugal force, 1133). This produces IHC-stained tissue sections from both methods initially
a small cell pellet at the bottom of the tube. The superna- were evaluated. A pilot study was performed using the
tant fluid is poured off carefully so as not to dislodge the remaining portions from 24 large-volume body fluid speci-
cell pellet. A 10% formalin solution (approximately mens that already had yielded a cytologic diagnosis. The
20 mL) is then poured over the cell pellet. The specimen is specimens consisted of 16 pleural fluids, 4 ascites fluids, 1

886 Cancer Cytopathology December 2016


pelvic wash, 1 pericardial fluid, and 2 cyst fluids. To ensure
that this method would consistently achieve the expected
results, we followed the guidelines provided by the College
Cancer Cytopathology December 2016
New Method for Cell Block Preparation/Lindsey et al
of American Pathologists for the analytic validation of
immunohistochemical assays.10 Although these recommen-
dations are for IHC specimens, the intent of the recom-
mendation is to ensure consistency of a procedure or
method that does not produce a numeric output. After this
initial assessment, we extended the use of AFFECT to all
specimens, including the needle rinses of FNA cases.
For the next 6 months, every specimen that was sub-
mitted for cell block preparation was evaluated in 2 ways.
First, after the first centrifugation step, the size of the cell
pellet was evaluated by the cytopreparation technologist.
Each pellet was assigned a score based on size (see Fig. 3
for reference images of cell pellet size). Second, after the
specimen was signed out, the slides and the paraffin block
were evaluated by the cytotechnologist laboratory manager.
AFFECT has been implemented and used in the
study laboratory since 2014. We performed a natural lan-
guage search of our laboratory information system to find
specimens for the data sets for 2 time frames. For the year
preceding the implementation of AFFECT and the year
after, we evaluated cases from endoscopic ultrasound-
guided (EUS)-FNA with a cytologic diagnosis of malig-
nancy or neoplasm and cases from endobronchial
ultrasound-guided (EBUS)-FNA with a malignant diagno-
sis. We considered a successful cell block to be one that
yielded sections sufficient for the performance and evalua-
tion of immunostains or one for which the material in the
cell block matched the diagnosis. A failure was a case in
which the cell block yielded too few cells of interest to
characterize or material that was incongruent or inconse-
quential to the diagnosis (eg, yielding only blood, merely
Figure 2. (A) Specimen preparation and formalin fixation. The
specimen is added to a conical vial and centrifuged, produc-
ing a cell button at the base of the vial. The supernatant fluid
is poured off and formalin is added. The button is resus-
pended and allowed to fix. The vial is again centrifuged, pro-
ducing a formalin-fixed cell button. The supernatant fluid is
again poured off, leaving approximately 1 mL of formalin. The
button is resuspended in the formalin. A volume of the resus-
pended cell pellet is aspirated with a plastic transfer pipette.
(B) Funnel assembly and addition of the specimen. The speci-
men is added to the funnel with the prepared foam recepta-
cle at the outlet. The specimen volume in the funnel does not
exceed the horizontal reservoir. The cell pellet collects in the
well of the foam receptacle during centrifugation. (C) Sub-
mission of the specimen for routine processing. The funnel
and metal clip are removed, and the foam and cell pellet are
wrapped in the BIO-WRAP tissue paper and submitted for
regular processing.
887
Original Article
Figure 3. Reference images for cell pellet size and results of
the first 372 specimens prepared with AFFECT.
atypical cells, benign lymphocytes, or cartilage). The suc-
cess rate, preservation of cytomorphology, and use for
ancillary tests were assessed and discussed.
For the pilot study, there was no interaction or inter-
vention with living individuals for the specific purpose of
the study, and no individually identifiable information was
ascertained.11 Institutional Review Board approval was
granted for the collection of data to compare the methods
of cell block preparation.
RESULTS
For all 24 cases in the pilot study comparing AFFECT
with the agarose method, both preparation methods pro-
duced a cell block. All cases were deemed adequate for
evaluation, but the cytoplasmic and nuclear detail preserva-
tion on cell blocks that were prepared with AFFECT were
judged to be superior in all 24 cases (Fig. 4).
There were 372 cell blocks produced over the first 6
months of implementation of AFFECT. The specimen
types are listed in Table 1. A malignant diagnosis was ren-
dered in 132 cases (35%), neoplasm in 17 cases (5%), sus-
picious in 9 cases (2%), atypical in 19 cases (5%), negative
in 124 cases (33%), and unsatisfactory in 24 cases (6%).
888
The remaining 47 cases (13%) were descriptive of benign
or reactive findings. Twenty of the specimens had no data
recorded regarding the cell pellet size. The numbers of
specimens for each score are shown in Figure 3 along with
the reference images for cell pellet size. Overall, 276 of the
372 cell blocks (74%) correlated with the final diagnosis.
In 87% of the cases, pellets were scored as <0.2 (very small
samples), and 72% of these tiny pellets both produced a
cell block and correlated with the final diagnosis. With the
agarose method, these samples would have been considered
by our cytopreparatory technicians to be too small to pro-
cess as cell blocks. There were 180 immunostains ordered
for 62 of the cases. Seven of the immunostains were con-
sidered noncontributory due to insufficient pertinent cells
on sections.
Cytologic examination of 356 EUS-FNA specimens
in the year before the implementation of AFFECT
revealed 34 cases with a cytologic diagnosis of a neoplasm
and 112 cases with a diagnosis of malignancy. The success
rates of cell block preparation in these 2 subsets were each
47%. During the year after AFFECT was implemented,
27 of 373 specimens were diagnosed as neoplasms and 89
were diagnosed as a malignancy. The success rate was 74%
and 62%, respectively. It is interesting to note that cases in
which a neoplasm is the preliminary on-site diagnosis fre-
quently have extra passes taken specifically for cell block;
conversely, malignant cases do not. The practice of obtain-
ing more material for cell block in the case of a presumed
neoplasm did not change during the 2-year timeframe.
EBUS-FNA specimens were evaluated for the same time
periods. In the year before AFFECT was implemented,
there were 231 specimens with 78 malignancies, and in the
year after AFFECT was implemented, there were 230
specimens with 90 malignancies. The success rate of cell
block production was 88% and 92%, respectively. Immu-
nostains were performed on 38% and 54%, respectively,
of these specimens. Examples of comparable adenocarcino-
mas are shown in Figure 5.
DISCUSSION
The cell block plays an increasingly important role in accu-
rate diagnosis and ancillary testing. Various issues involving
cost, adequacy, morphology, and fixation have been raised
with the commonly used methods for cell block.7 For
example, the plasma thrombin method, which uses dyed
blood products to coagulate the specimen, may introduce
Cancer Cytopathology December 2016
 New Method for Cell Block Preparation/Lindsey et al

Figure 4. Comparative images of the histology of cell blocks prepared from the same specimen in parallel using (Left) the aga-
rose method and (Right) AFFECT. (A and B) Pleural effusion demonstrating metastatic lung adenocarcinoma. (C and D) Ovarian
cyst fluid. (E and F) Ascites fluid with metastatic breast adenocarcinoma. (G and H) Pericardial effusion (H & E,
magnification 3 400).

Cancer Cytopathology December 2016 889

Original Article

TABLE 1. Specimen Types formalin as the fixative, decrease the diameter of the cell
Specimen Source No. of Specimens pellet (thereby increasing the thickness), and make the
media that holds the cell pellet less likely to obscure the tis-
Lymph nodes 127 sue from the view of the histotechnologist.
Pancreas FNA 46
Pleural fluids 44 The method that we invented is a formalin-fixed,
Soft tissue 36
Lung FNA 22
paraffin-embedded cytology cell block technique that uses
Liver FNA 21 a cytospin centrifuge and funnel to deposit a cell pellet
Bone FNA 18
Ascites 12
into a well on a piece of open-cell, absorbent foam. The
Pelvic wash 12 foam and the pellet are then sent through normal process-
Stomach 4
Peritoneal wash 3 ing. This method allows for the production of cell blocks
Breast FNA 3 quickly and inexpensively, and can be readily performed in
Retroperitoneum 3
Bile duct FNA 3 the majority of cytology laboratories.
Pericardial fluid 2 We closely monitored the performance of AFFECT
Mediastinum 2
Rectum 2 for the initial 6 months of the study, collecting data regard-
Parotid 2
Adrenal 2
ing 372 specimens and cell blocks. Including the smallest
Duodenal mass 1 of cell pellet sizes, the method allowed for an adequate cell
Omentum 1
Vulvar cyst 1
block in 72% of cases. With the agarose method, we would
Fluid small bowel 1 not even have attempted to create a cell block with these
Abdominal wall FNA 1
BAL 1 tiny samples. We then compared the preceding and subse-
Knee 1 quent years of cell block specimens from FNA specimens
Kidney 1
from endoscopy and bronchoscopy. With AFFECT, we
Abbreviations: BAL, bronchoalveolar lavage; FNA, fine-needle aspiration.
nearly doubled our success rate for specimens from EUS-
FNA cases and increased our success rate of cell block
biologic markers; the heating process in the agarose preparation from EBUS-FNA cases. It is important to note
method may affect cell morphology; and some other tech- that the specimens from EBUS-FNA with cell blocks pro-
niques use fixatives other than formalin, which can inter- duced with AFFECT had a higher rate of sufficient mate-
fere with IHC. rial for immunostains. It would not be fair to suggest that
The poor cellular yields and poorer histology of the this is a frankly increased rate of material sufficient for the
cells due to heat artifact from the molten agarose prompted performance of immunostains in the cell block because
us to seek a new method of cell block preparation. We many factors determine whether immunostains are consid-
hypothesized that the number of specimen transfers was ered necessary, but tumor-specific treatments are driving
likely the cause of the loss of material. The inability of the the need for very specific diagnoses. Having material for
histotechnologist to view the cell pellet at the time of sec- immunostains is helpful and sometimes critical.
tioning also contributed to losses. We had been using a The benefit of well-preserved histology and cytologic
block marker to indicate the position of the cell pellet in details in a cell block specimen is obvious but less easily
the block, but the paraffinized agarose had a tendency to quantifiable. In the current study, the only method we
buckle, thereby moving the marker out of the plane of the compared in a head-to-head study with AFFECT was the
tissue. In addition, it had been noted that the paraffinized agarose method, and the only specimen type was large-
agarose was friable and likely to crumble during the volume body fluid specimens. This is interesting to note
embedding process. Finally, the surface area on which the because studies have demonstrated that laboratories gener-
cell pellet collects is inversely proportional to the thickness ally are less satisfied with the agarose method.7
of the pellet. If the surface could be decreased, the thick- The problem of the cell block appears simple. The
ness of the pellet would increase, possibly allowing for objective is only to keep single cells and blood clots bound
increased numbers of histologic sections. together long enough to make it through histologic proc-
In developing a new method, we attempted to essing without loss of the cellular material. One con-
decrease the number of steps in the process, decrease the founder is that the ideal fixation in formalin also renders
time expenditure of the cytopreparatory technician, use cells less likely to stick together. In addition, any material

890 Cancer Cytopathology December 2016

 New Method for Cell Block Preparation/Lindsey et al

Figure 5. Comparative images of the histology of cell blocks prepared from different fine-needle aspiration (FNA) specimens
using (Right) the agarose method and (Left) AFFECT. (A and B) Endoscopic ultrasound-guided FNA of pancreatic adenocarcino-
mas. (C and D) Bronchoscopic ultrasound-guided FNA of lung adenocarcinomas (H & E, magnification 3 400).

that is used to hold the cell block together must be able to
be paraffinized and cut on a microtome designed to cut
paraffinized tissues.
Producing a successful cell block is a multivariable
process. The preanalytic variables, such as the type of
tumor and the skill of the operator obtaining the sample,
are likely the most important. The success or failure of any
method of cell block preparation also is difficult to judge
because, to the best of our knowledge, there is no ready
method with which to ascertain whether useful lesional
cells are within the needle rinses. Cellular smears are an
indication that the needle was once in the lesion but dem-
onstrate nothing of the fidelity of each and every aspiration
attempt. In addition, visible material within the needle
rinses does not necessarily represent lesional material.
To the best of our knowledge, there are relatively few
innovations and technologies that address this cytology
adjunct, especially when compared with the medical
advances made over the lifespan of the field of cytology.
Indeed, there are numerous emerging small biopsy devices
threatening to outperform cytology in the acquisition of
sufficient material. The cell block is central to the future of
cytology, allowing us, with high sensitivity and specificity,
to do more with less.
FUNDING SUPPORT
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
There is a provisional patent on the technology described
herein (Method and Apparatus for Cell Block Preparation
of Cytology Specimens) that is owned by the Medical Uni-
versity of South Carolina.

Cancer Cytopathology December 2016 891

Original Article
AUTHOR CONTRIBUTIONS
Kathryn G. Lindsey: Conceptualization, methodology,
validation, formal analysis, investigation, resources, data
curation, writing–original draft, writing–review and edit-
ing, visualization, supervision, and project administration.
Patricia M. Houser: Conceptualization, validation, formal
analysis, investigation, resources, data curation, writing–
original draft, writing–review and editing, supervision, and
project administration. Wanda Shotsberger-Gray: Con-
ceptualization, methodology, validation, formal analysis,
resources, and data curation. Olga S. Chajewski: Concep-
tualization, validation, formal analysis, writing–original
draft, writing–review and editing, supervision, and project
administration. Jack Yang: Conceptualization, validation,
formal analysis, writing–original draft, writing–review and
editing, supervision, and project administration.
REFERENCES
1. Bahrenberg L. On the diagnostic results of the microscopical exam-
ination of the ascitic fluid in two cases of carcinoma involving the
peritoneum. Cleveland Med Gaz. 1896;11:274-278.
892
2. Sigma-Aldrich. Material safety data sheet: cellulose nitrate: MSDS
No. 435066. http://www.orcbs.msu.edu/msds/082905_LLQ_022_
CELLNIT11POINT5.PDF. Accessed August 25, 2016.
3. Fahey C, Bedrossian UK. Collodion bag: a cell block technique for
enhanced cell collection. Lab Med. 1993;24:94-96.
4. Josefson A. Primur lungkancer med svulstceller i pleuraexudat och
sputum. Hygiea. 1901;63:435-445.
5. Shackelford RI, Jones JL. An embedding medium for permanent
sections of exudative material and fragments of tissue removed for
biopsy. Tech Bull Regist Med Technol. 1959;29:155-156.
6. Kaltenbach FJ, Hillemanns HG, Fettig O, Hilgarth M. Thrombin cell
block technic in gynecologic cytodiagnosis. Acta Cytol. 1973;17:128-130.
7. Crapanzano JP, Heymann JJ, Monaco S, Nassar A, Saqi A. The
state of cell block variation and satisfaction in the era of molecular
diagnostics and personalized medicine. Cytojournal. 2014;11:7.
8. Benerini Gatta L, Cadei M, Balzarini P, et al. Application of alternative fix-
atives to formalin in diagnostic pathology. Eur J Histochem. 2012;56:e12.
9. Varsegi GM, Shidham V. Cell block preparation from cytology
specimen with predominance of individually scattered cells. J Vis
Exp. 2009;(29). pii: 1316.
10. Fitzgibbons PL, Bradley LA, Fatheree LA, et al; College of American
Pathologists Pathology and Laboratory Quality Center. Principles of
analytic validation of immunohistochemical assays: guideline from
the College of American Pathologists Pathology and Laboratory
Quality Center. Arch Pathol Lab Med. 2014;138:1432-1443.
11. Office for Human Research Protections, US Department of Health
and Human Services. Human subject regulations decision charts.
http://www.hhs.gov/ohrp/regulations-and-policy/decision-charts/index.
html. Accessed May 10, 2016.
Cancer Cytopathology December 2016