General Cytology
illustrated later. This technique is advantageous for FNA speci-
mens because it requires only a few cells. It allows also retention
of cellular morphology, which permits simultaneous evaluation
of morphology and chromosomal alterations. Moreover, recent
studies have demonstrated the feasibility of FISH on Papanico-
laou-stained archival cytology slides, highlighting the good flex-
ibility of such method.63-66 These advantages probably explain
why FISH is becoming more and more popular in cytopathol-
ogy laboratories.
FISH Strategy
Interphase FISH requires simple material such as cytospins from
FNA specimens. Cytospin is an optimal preparation for I-FISH
because the monolayer allows excellent hybridization results.
Cytospin preparation can be made by Ficoll-Hypaque gradient-
separation technique and then fixed in methanol-glacial acetic
acid (3:1) for 20 minutes at -20°C. The slides will be then air-
dried and stored at -20°C prior current FISH procedure. Speci-
men handling is thus very simple, but it is critical to avoid delays
in specimen processing in order to prevent possible degradation
of the target DNA and subsequent poor hybridization results.67
Subsequent FISH steps can be then easily performed without fur-
ther manipulation of the samples, with the use of commercially
available kit sets including the premixed probes, and according
to the protocol recommended by the manufacturer. At least, 200
nonoverlapping and intact nuclei per case and at least two dif-
ferent areas on the same slide should be scored before giving a
result. The great advantage of working on an interphase cell can
nevertheless be a source of interpretative pitfalls in that random
chromosome colocalizations occur not infrequently in normal
nuclei and can mimic chromosomal translocations. Although
most of the commercial probes have been designed to limit the
risk of false-positive profiles, it remains critical to determine the
frequency of such false-positive cells in order to define a cutoff
level. Normal lymphocyte nuclei can be used as negative control
to assess hybridization efficiency, and the cutoff level for positiv-
ity should be set at the mean (%) ± 3 standard deviations. Beside
this pitfall, other good practice recommendations are needed
and must be known by the user. Such guidelines are detailed in
an excellent overview recently published that we highly recom-
mend to the reader.68
The commercial probes are usually several hundred kilobases
in length and yield large, bright and easily detectable signals.
They are currently available to detect many of the relevant chro-
mosomal abnormalities described in the previous section and
are known to be highly sensitive.69 For detection of chromosomal
translocations, three different kinds of probes are available,
including the dual-fusion probes, the single-fusion extra-signal
probes, and the break-apart probes, all being dual-color probes
(Figs 2.17A and 2.17B). Dual-fusion and extra-signal probe sets
are made of two differentially labeled (green and red) DNA seg-
ments, each of these segments identifying one of the chromo-
somal loci involved in the translocation. For the dual-fusion
probes, an abnormal pattern will be represented by one red and
one green signal (representing the normal homolog) and by
two fusion or colocalization signals corresponding to the chro-
mosomal translocation and its reciprocal ("2F,1R,1G" pattern).
Typical examples are probes designed to detect lymphoma-
associated chimeric genes subjacent to translocation such as the
in follicular or mantle cell lymphoma,
respectively (Fig. 2.18A). Such probes make it possible to
Gene A
Gene B
Dual fusion probe
Normal cell
Cell with
Break apart probe
Normal cell
Cell with
Fig. 2.17
Schematic representation of two different types of dual-color
the dual fusion and break-apart probes.
(A) Left: Dual-fusion
probes are composed of two differentially labeled (green and red) DNA
segments, each of these segments identifying one of the genes/loci involved
in the chromosomal translocation. The probes are usually several kilobases
in length and extend largely on both sides of the gene of interest. Right:
a normal pattern will show two red and two green spots, whereas a cell
harboring a chromosomal translocation will demonstrate two fusion or
colocalization signals corresponding to the chromosomal translocation and
its reciprocal; the red and green spots indicate the two remaining normal
chromosomes ("2F,1 R,1G” pattern). (B) Left: break-apart probes are made
of two differentially labeled (green and red) DNA segments flanking the
breakpoint cluster region of a gene involved in chromosomal translocations.
Right: a normal cell will show two yellow fusion signals corresponding to
two copies of a normal gene. The disruption of one of these two copies
subsequently to a chromosomal translocation will lead to split of one yellow
signal into two red and green signals ("1 R,1G,1F” pattern).
significantly reduce the risk of false positives as the possibility
that two overlapping signals are due to random spatial prox-
imity of the participating chromosomal loci remains very low.
The abnormal pattern for extra-signal translocation probes will
be represented by a single fusion (corresponding to one deriv-
ative chromosome) plus a small extra signal representing the
residual portion of one of the loci involved in the translocation.
Again, the probability that such pattern is observed in a normal
nucleus is very low. A well-known example is the probe used to
detect the
chimeric gene in chronic myeloid leukemia.
Such a probe has not been designed for detection of recurrent
chromosomal translocations in lymphoma or sarcoma and will
not be illustrated here. Dual-color break-apart probes are made
of differentially labeled (green and red) DNA segments located
on either side of a breakpoint cluster region within a target
gene. The separation of green and red signals indicates break
between the 5' and 3' regions of the rearranged gene. In normal
cells, the two probes colocalize to produce two yellow fusion
previous page 46 ComprehensiveCytopathology 1104p 2008 read online next page 48 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off