Hypermetaphase Fluorescence In Situ Hybridization

Document technical information

Format pdf
Size 1.4 MB
First found Nov 13, 2015

Document content analysis

not defined

Leukemia wikipedia, lookup

Chromosome wikipedia, lookup

Karyotype wikipedia, lookup

Chronic myelogenous leukemia wikipedia, lookup

Chromosomal translocation wikipedia, lookup

Cytogenetics wikipedia, lookup

Fluorescent in situ hybridization wikipedia, lookup


Edgar R. Champlin
Edgar R. Champlin

wikipedia, lookup

J. Craig Ruby
J. Craig Ruby

wikipedia, lookup

J Roberto Trujillo
J Roberto Trujillo

wikipedia, lookup




From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
Hypermetaphase Fluorescence In Situ Hybridization for Quantitative
Monitoring of Philadelphia Chromosome-Positive Cells in Patients With
Chronic Myelogenous Leukemia During Treatment
By David C. Seong, Hagop M. Kantarjian, Jae Y. Ro, Moshe Talpaz, Jingping Xu, Johnnie R. Robinson,
Albert B. Deisseroth, Richard E. Champlin, and Michael J. Siciliano
Using Philadelphia chromosome-positive (Ph') chronic myelogenous leukemia (CML) as a model, our aim has been t o
develop a molecular cytogenetic method of high resolution
analysis for monitoring thefrequency of cells with nonrandom chromosome rearrangements in the bone marrow of
patients receiving treatment for hematologic malignancies.
Long-term exposure (24 hours) of bone marrow cultures t o
colcemid (0.1 pg/mL) maximized a high frequency of metaphase collection. Such preparations were subjected t o fluorescence in situ hybridization(FISH) using a 5 Mb probe that
overlapped the region of the translocation at chromosome
9q34. This detected the Ph translocation in the resultant
large number of overly contracted chromosome
The procedure was validated and verified by studying 70
double-blindmarrow samples frompatients in different
stages of Ph' CML and from patientswith Ph- hematologic
malignancies (controls). This hypermetaphase FISH (HMF)
method clearly identified Ph' metaphases andallowed
the analysis of 500 hypermetaphases per sample in less than
1 hour after FISH. HMF (1) identified statistically significant
differences between thefrequencies of Ph' cells in samples
that differed by less than 4%; (2) resolved such differences
among patient samples that were all judged 100% Ph' by
standard G-band cytogenetics (CGI; (3) resulted in the reclassification of response status in 23% of the patients initially
classified by CG; (4) recognized Ph' cells in 16% of patients
characterized as having a complete cytogenetic response
and in one patient with an original diagnosis of Ph- CML;
and (5) was informativewhere insufficient metaphases were
obtainable for analysis by CG. HMF appears t o be uniquely
suitable for monitoring the status of patients with CML receiving treatment. It should also be applicable for patients
with any hematologic diseases where chromosomal alterations are known andappropriate FISH probes are available.
0 7995 b y The American Society of Hematology.
meration and analysis of Ph+ cells in marrow preparations
would be available to monitor MRD and to observe subtle
alterations of the frequency of Ph' cells in patients in various
stages of therapy. In this study, we report the results of
experiments to determine the effectiveness of HMF. Blinded
marrow samples from a large series of patients with CML
in different stages of disease and treatment as well as from
control patients were studied. We report the power of the
procedure to detect MRD and to identify statistically significant alterations in the frequency of Ph+ cells in response to
IEMSA-BAND chromosome analysis (CG) has been
the standard for determination of the presence of the
Philadelphia chromosome (Ph) in the bone marrow of patients with chronic myelogenous leukemia (CML).' While
extremely reliable when a good harvest of metaphases is
obtained from a marrow culture, more than 20 to 25 metaphases are rarely recovered from patients in therapy. Consequently, in patients with CML receiving treatment, monitoring the effect of therapy by CG is limited with respect to
detecting minimal residual disease ( M m ) and identifying
significant changes in Ph+ cell frequency. Analyzing greater
numbers of metaphases per sample would provide higher
resolution and greater efficiency in addressing those issues.
If a significant proportion of the cells cycling in the marrow could be arrested at the metaphase stage, hundreds, or
even thousands, could be collected onto a slide for analysis.
Short-term treatment (30 to 60 minutes) of cultures with a
mitotic arresting agent, colcemid, is the procedure presently
used to provide metaphases for analysis.* It has been optimized to collect metaphases in which the chromosomes are
sufficiently elongated and separated for G-band analysis. Unfortunately, longer colcemid treatment, for greater numbers
of collected metaphases, results in chromosomes becoming
overly contracted, with a general deterioration in morphology and distribution. Such preparations are inappropriate for
CG analysis, which requires well-spread metaphases in
which the chromosomes retain good morphology.
We have shown that using a probe that will detect the Ph
translocation by fluorescence in situ hybridization (FISH), a
clear, rapid, and unambiguous signal detected Ph' cells in
metaphases that would, in general, be considered unacceptable for CG a n a l ~ s i s . ~
. ~ we have conducted experiments to maximize the frequency of cells arrested in metaphase (hypermetaphases). By performing FISH on these
preparations, in a procedure we will refer to as hypemetaphase/FTSH (HMF), we expected that high resolution enuBlood, Vol 86,No 6 (September 15), 1995: pp 2343-2349
Hypermetaphuse studies. Bone marrow cells from fivenormal
individuals (prospective donors for allogeneic bone marrow transplantation) and seven patients with CML were studied to optimize
conditions for hypermetaphase preparations. All patient samples in
this report were obtained as part of a program sponsored by the
National Cancer Institute, and appropriate informed consent was
obtained from all patients. Cells were cultured in T25 tissue culture
From the Departments of Hematology, Pathology, Clinical Immunology, and Molecular Genetics, University of Texas M.D. Anderson
Cancer Center, Houston, TX.
Submitted March 14, 1995; accepted May 11, 1995.
Supported in part by National Institutes of Health Grants No.
CA34936 CA49639, and CA55164, and a gift from Kenneth D.
Address reprint requests to Michael J. Siciliano, PhD, Department
of Molecular Genetics, University of Texas M.D. Anderson Cancer
Center, 1515 Holcombe, Houston, TX 77030.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1995 by The American Society of Hematology.
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
Table 1. Patient PoDulation
No. of Patients
Control group
ALL patients
AML patients
CLL patients
Ph- CML patients
Ph' CML patients
Cytogenetic response status
None b 9 5 % Ph' cells by CG)
Minor (35% to 95% Ph' cells by CG)
Partial (1% to 34% Ph' cells by CG)
Complete (0% Ph+ cells by CG)
Unknown; insufficient metaphases
flasks containing 10 mL of RPMI 1640 medium (GIBCO, Gaithersburg, MD) with 10% serum (GIBCO; newborn bovine serum: fetal
bovine serum, 1 :1). Cell densities were adjusted to not exceed 5 X
106/mL, and incubation at 37°C with 5% CO, for 24 hours was
conducted. Cultures were then subdivided and exposed to varying
concentrations (0.01 to 0.4 pglmL) of colcemid (N-methyl N deacetyl-colchicine in phosphate-buffered saline [PBS]; Boehringer Mannheim) over varying time periods (0.5 to 48 hours). Efficiency of
each procedure in collecting metaphases was determined by calculating the mitotic index (total number of mitotic figures seen on a slide
onto which a fixed preparation hadbeen dropped per 1,000 for
the total number of cells counted) for the preparation. Conditions
providing the highest mitotic indices were the conditions applied to
the coded samples.
Study population. From March 1993 through May 1994, 70 patients with hematologic malignancies were included in this investigation after informed consent was obtained according to institutional
guidelines. Bone marrow aspirations were obtained by four investigators (H.M.K., M.T., R.E.C. and J.R.R.), coded, and sent as blinded,
coded samples for both CG (to the Clinical Cytogenetics Laboratory,
University of Texas M.D. Anderson Cancer Center) and HMF (to
D.C.S., J.X., and M.J.S.). HMF preparations were made by J.X. All
HMF slides were read by D.C.S. Those making the HMF preparations had no knowledge of the patient's identity, diagnosis, or cytologic response status for any of the samples studied. After all the
data were tabulated, the code was broken, and it was then determined
what the correlations were between the HMF results and the type
of patient sample analyzed.
Fifty-eight patients with Ph+ CML in various states of responsiveness to therapy as well as one patient with a diagnosis of Phnegative CML were entered into the study. An additional 11 patients
were recruited that had other hematologic malignancies: acute
lymphocytic leukemia (ALL), acute myelogenous leukemia (AML),
and chronic lymphocytic leukemia (CLL). This latter group comprised the presumed Ph- controls. The acute leukemia patients were
in first remission, while the CLL patients had active disease. The
number of patients in each group and the status of the cytogenetic
response of patients with Ph+ CML are listed in Table 1.
G-band cytogenetic analysis (CC). Cytogenetic analysis was
performed on freshly aspirated bone marrow samples. Metaphases
were analyzed for G-bands using trypsin-Giemsa techniques as previously described.* Consistent with the routine analysis, 20 to 25
metaphases were analyzed. Cytogenetic response was defined relative to the percentage of Ph+ cells identified by CG: complete if
0%, partial if from 1% to 34%, minor if from 35% to 95%, and
none if greater than 95% Ph cells?
Slide preparation. Marrow cells were transferred from culture
flasksto 15-mL Falcon tubes andthen spun at 1,200 rpm for 5
minutes. Ten milliliters of hypotonic KC1 (0.075 mow) was added
to the pellet for 20 minutes at 37°C. Then, 1 mL of fixative solution
(methano1:acetic acid, 3:l) was added, and cells were spundown
for 5 minutes at 1,200 rpm. Supernatant was decanted and the pellet
was resuspended in fixative for at least 20 minutes at room temperature and spun as above. This process was repeated two to three times
until the pellet was white. Cells were then left in 2 mL fixative.
FISH. Probe for FISH was prepared from radiation hybrid E6B,
which has been shown to contain only approximately 5 Mb of human
DNA from human chromosome band 9q34 as its only human genomic content. Molecular marker analysis has indicated that the human
region retained extends from A K l to ABO, which includes ABL, and
that approximately one third of the region is translocated to the Ph
chromosome in patients withCML.6 The probe, therefore, allows
the ready detection of three signals (from the normal chromosome
9, the chromosome 9q+, and the Ph chromosome), while only two
signals from the chromosomes 9 are seen in normal metaphases!
Methods for preparation of human specific FISH probe from hybrid
cells by inter-Ah-polymerase chain reaction (PCR), biotin-labeling
of probe, competitive hybridization blocking of repeat sequences,
FISH, avidin-fluorescein detection, ultraviolet microscopy, and photography as performed in this laboratory have been described by Liu
E6B probe was visualized in cells as yellow-green fluorescence
when viewed through a B2A Nikon filter (excitation, 470 to 490;
emission, 515; Nikon Inc, Melville, NY). Photomicrographs were
taken on Kodak 160 Ectachrome (Eastman Kodak, Rochester, NY)
directly without image enhancement or processing.
After 24 hours of culture in RPMI medium and 10%fetal
calf serum, bone marrow cultures were exposed to varying
concentrations of colcemid (0.01 to 0.4 pg/mL) over varying
time periods (0.5 to 48 hours). Maximum metaphase collection was achieved after 24 hours of culture in 0.1 pg/mL.
This resulted in a 3% to 6% mitotic index in preparations
from normal marrows, and 4% to 15% from patients with
CML (with higher levels observed in samples with higher
percentages of Ph+ cells). As shown in Fig 1, multiple metaphase figures were present in a single microscope field. The
chromosomes in most such figures were highly overcontracted and useless for a karyotype analysis. However, when
E6B FISH probe was applied to such metaphases from a
CML marrow culture, Ph+ cells could be readily distinguished from cells not carrying the Ph translocation (Fig 2).
The presence of only two pairs of FISH signals, as shown
in Fig 2A, is interpreted as being typical of a non-Ph+ cell
throughout this procedure. As the E6B probe marks the terminal portion of the q-arm of chromosome 9, and as in
overly contracted chromosomes (a consequence of the 24
hours in colcemid) chromatids of a chromosome tend to fly
apart while being held together at their centromeres, each
chromosome is seen as a pair of signals, with each signal of
the pair representing a chromatid. Figure 2B shows a similar
two-pair of chromosomes with a thirdsmaller pair of signals
on what appears to be a smaller chromosomal element. This
cell is interpreted as a classic Ph+ cell. A similar such cell
is seen in the bottom portion of Fig 2C. The lesser intensity
of the signal on the small element is consistent with the fact
that in the Ph translocation only one third of the signal is
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
Fig l. Single microscope field demonstrating the
yield of metaphases by the hyparmetaphase procedure (original magnification x 200).
moved from the 9q- chromosome to the Ph chromosome.6 less ideal yet still interpretable cells were identified and clasRarely, the paired signals became completely separated, re- sified. In the cell at the top
of Fig 2C, the two chromosomes
sulting infour signals of relatively equal intensity (interpre- 9 are crowded together, and the paired signalis resolved on
ted as a normal cell) six
or signals (two with lesser intensity; only one of them, while the cell on top in Fig 2D exhibits
interpreted as a Ph+ cell). In addition to these ideal situations,
a metaphase of very poor quality in which the two signals
Fig 2. Four dlfferent photomicrographs(A-D) of FISH usingE6B probe on hyperrnataphase culture preparations from patient bone marrow
samples demonstrating the rangeof cell types evaluated[seetext). Arrows identify the Ph chromosome.
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
Table 2. HMF Results in Control Samples
readings in 500 cells of a blinded control sample allows US
to have 99%confidence that when no Ph cells are identified
in a 500-cell sample, the maximum frequency of Phi- cells
present in the sample is less than 1%.
HMF and CG results in patients with CML. Samples of
16 patients classified as having no cytogenetic response (Table 3) anddeterminedto be 100% Ph'by CC hadfrom
91.0% to 99.0% Ph' cells by HMF. Inno case wasthe
frequency of Ph' cells determined by HMF statistically significantly different from the frequency determined by CG.
In the case with the greatest difference (D372), 20 of 20 Ph'
cells by CG compared with 455 of 500 by HMF produced
x' of only 1 .0 ( P < .32). The lack of a statistically significant
difference is attributable to the small number of cells studied
by CG. The inability to identify a small percentage ofPh
cells by that method is attributed to sampling error.
One might ask if the difference between 100% Ph' observed by CC and the90+% observed by HMF in this patient
set might be the result of false-negative readings (scoring a
Ph' cell as Ph- because of the inability to observe the third
signal due to artifact, eg, loss of the Ph chromosome from
the spread, overlap of one of the three signals with another,
poor staining rendering the smaller Ph' signal invisible in
some cells). It is our experience that false-negative cells are
not a problemwiththis
procedure. Metaphases analyzed
have a recognizable cluster of chromosomal elements mitigating the question of chromosome loss. The dual signals
from a chromosome make indistinguishable overlaps unlikely. Even staining and the bright signal seen from the Ph
chromosome due to the large size of the probe and hybridization sequences onthe chromosomes suggest that the elements being targeted will be visible when present. This positionwas
supported by our observations inthe
material where two chromosomal elements were seen in every cell. Therefore, the non-Ph' cells seen in this category
of nonresponders were most likely due to the real presence
and detection of cells without the translocation in the population. Determining whetherthey are non-myeloid dividing
cells or possible normal myeloid progenitors, and whether
there are clinical consequences in patients with different
frequencies of them, must await further analyses. Such analyses should be possible now because when samples of 500
cells each from patients differ by as little as 496, the difference is significant at the 99% confidence level.
Among the 10 patients with minor cytogenetic response
by CC (Table 3), highly statistically significant differences
in the frequencies in the levels ofPh' cells as determined
by CG as opposed to HMF were observed for samples from
two patients (D466 and D662j. The HMF results reclassify
the patients into the no-response category. Follow-up CG
data were available for one of these patients (D466), and it
was determined that the cytogenetic response had been lost.
This indicated thattheHMF result gave a more accurate
picture of patient D466's clinical status than the lower Ph+
frequency observed by CG.
Among the patients that were classified as having a partial
cytogenetic response by CG, there were two samples (D4 12
and D685) with statistically significant differences inthe
frequencies of Ph' cells as determined by the two different
Patient No.
ALL patients
AML patients
CLL patients
96 Ph' Cells (no. of cells analyzed)
0 (500)
0 (500)
of one of the pairs (on the right) appears stretched apart.
These cells were scored as not being Ph', because the third
paired signal regions were not observed. In the cell at the
bottom of Fig 2D, the signal from the Ph chromosome does
not appeared paired. This was a frequent observation, as that
signal is closer to the centromere than the signal at the end
of the 9q chromosome, and, therefore, the chromatids often
are not sufficiently separated to resolve the signal as paired.
However, such cells were scored as Ph', as the unpaired
signal was clearly in addition to clear signals from the chromosomes 9 and 9q-. The following rare types of cells were
not counted: cells poorly stained because of being under a
bubble or in an area of the slide with poor staining and cells
in which the nucleus was clearly not intact. These represent
the spectrum of types of signals in cells and how they were
interpreted in the analysis of patient samples.
Scoring 500 or more such cells per sample for presence
or absence of the Ph chromosome by the criteria indicated
could be performed in less than 1 hour. To determine the
quantitative limits of this procedure in detecting the frequency of Ph+ cells in a heterogeneous population, doubleblind experiments were conducted as described in Materials
and Methods on samples from patients with CML in different
clinical states of disease and control samples from putative
non-Ph' patients. The results are reported in Tables 2
through 4 and are described below.
Control population. In 10 of the l 1 blinded control samples, no Ph' cells were detected by HMF out of at least
500 metaphases analyzed per sample. All 10 had a diploid
karyotype as determined by CG. One of the blind samples
provided as a control, D052 (Table Z), turned out to be a
Ph' ALL. We scored 505 cells and determined that 76.2%
were Ph+. When notified of our result, co-investigators who
randomly inserted control samples among the CML samples
and who had access to the CG results verified that CG on
patient D052 conducted at the same time indicated 15 of 20
cells (75%) to be Ph+. These data clearly verify the effectiveness of the probe on these preparations to specifically identify the presence of the Ph chromosome in such samples.
Applying a Clopper-Pearson confidence interval for a binomial proportion,' we conclude that getting no false-positive
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
Table 3. Ph' Cells by CG and HMF in CML Patients Demonstrating
Different Levels of Cytogenetic Response (as Determined by CG)
% Ph+ Cells
(no. of calls analyzed)
Patient No.
No cytogenetic response
Minor cytogenetic response
Partial cytogenetic response
93.1 (484)
98.1 (515)
98.9 (500)
95.7 (548)
97.2 (500)
96.1 (481)
94.7 (421)
91.O (500)
97.8 (501)
91.5 (5251
99.2 (500)
99.0 (500)
99.2 (400)
93.1 (500)
96.4 (500)
96.5 (500)
70 (20)
95 (19)
38 (20)
70 (20)
92 (25)
55 (20)
89 (9)
80 (30)
45 (20)
67.2 (119)
93.8 (500)
42.0 (400)
68.2 (500)
91.7 (151)
95.3 (400)'
59.0 (563)
97.8 (500)"
49.4 (500)
5 (20)
5 (20)
1.6 (500)
18.2 (500)
31.4 (525)t
65.6 (500)
12.2 (500)
59.8 (500)t
11 (18)
Complete cytogenetic response
D l 40
0 (19)
0 (20)
0 (20)
10.2 (507)
0 (20)
0 (25)
0 (25)
0 (20)
0 (20)
0 (20)
0 (8)
0 (12)
0 (20)
0 (20)
0 (20)
0 (20)
0 (20)
0 (20)
Diagnosed as Ph- CML
0 (20)
0 (20)
0 (20)
% Phi Cells (no. of cells analyzed)
100 (20)
100 (20)
100 (14)
100 (20)
100 (20)
100 (20)
100 (20)
100 (20)
100 (20)
100 (20)
100 (20)
100 (35)
100 (20)
100 (20)
100 (20)
100 (20)
Table 4. CG and HMF Results in CML Patients With an Insufficient
Number of Scorable Metaphases to Analyze Percentage
of Ph+ Cellsby CG
0 (500)
0 (500)
0 (500)
0 (500)
0 (500)
0 (500)
0 (500)
0 (500)
0 (500)
0 (500)
6.1 (165)
Boldface indicates samples that were reclassified as to cytogenetic
response as a result of HMF result.
* Significantly different from CG result at P i ,001.
t Significantly different from CG result at P i .05.
Patient No.
NA (2)
NA (0)
NA (1)
NA (5)
NA (3)
NA (0)
NA (1)
NA (2)
39.4 (510)
0 (500)
6.5 (428)
9.7 (4211
0 (500)
0 (500)
46.3 (393)
0 (500)
Abbreviation: NA, not applicable.
methods (Table 3). Patient D685 was reclassified as a minor
responder. The verymuch higher frequency of Ph+ cells
seen in patient D495 using HMF,which also resulted in
the patient's reclassification to minor cytogenetic response,
lacked statistical significance because of the small number
of cells available for CG analysis.
Three (patients D032, D135, and D533), or 16%, of the
20 patients classified as having a complete cytogenetic response as determined by finding no Ph+ cells by CG were
shown by HMF to have persistent disease (Table 3). All
three patients have been recategorized as partial cytogenetic
responders. Follow-up data on these patients are notyet
available. Once again, the small number of cells analyzed
by CG in each of those three cases did not allow us to attach
statistical significance to the different outcome achieved by
HMF. However, comparison between the patient with the
lowest frequency ofPh' cells (yet still greater than 0) as
determined by HMF (patient D533) with any other patient
showing 0% Ph+ cells by HMF revealed highly statistically
significant differences (all with P < .001). While not central
to the study, an interesting observation (Table 3) was made
in a patient with a morphologic diagnosis of CML who was
diagnosed as Ph-negative by CG (0 Ph+ metaphases of 20
analyzed). Reverse transcriptase-PCR (RT-PCR) studies indicated the presence of the BCWABL chimeric transcript
associated with that disease (data not shown). HMF identified 25 Ph' cells of 475 analyzed (5.3%).
HMF results in patients with insuflcient metaphases by
CG. A proportion of patients receiving alpha interferon or
chemotherapy often have insufficient metaphases for analysis by CG. Among all eight patients investigated with insufficient metaphases for CG analysis in our study, all were
analyzable by HMF. Four (50%) had 0% Ph' cells by HMF,
two had a partial cytogenetic response (6.5% and 9.7%Ph+,
respectively), and two had a minor cytogenetic response
(46.3% and 39.4% Ph+, respectively) by HMF (Table 4).
We conclude that the HMF procedure is a valid method
of determining the frequency of Ph+ cells in the bone marrow
of patients with CML undergoing therapy. The procedure
identifies the frequency of Phi cells at a much higher level
of resolution than standard methodology, is often capable of
identifying low frequencies of Ph+ cells not identifiably by
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
Table 5. Numbers of Patients in Cytogenetic Response Categories
as Determined by CG and Their Classification
as Determined by HMF
more, in patients shown to be 0% Ph+ by HMF on 500 cells,
an argument could be made to reduce or stop a interferon
therapy, while continuing close monitoring by HMF. Studies
Cytogenetic Response by HMF
on 500 cells have very narrow confidence intervals ( ? O s % )
and minor changes (less than 4%) observed at periodic interby CG
vals (eg, every 3 to 6 months) can be reliable measurements
None, 16
of the efficiency of a particular therapy.
Minor, 9
The result finding 5.3% Ph' cells by HMF in the sample
Partial, 6
the Ph- patient foundto have BCWABL chimeric
Complete, 19
by RT-PCR is of particular interest. It offers an
Unknown', 8
alternative to the suggestion that subtle or masked transloca* Insufficient number of metaphases to make determination.
tions, undetectable by CG, may be the basis of such observations.'" Our finding suggests that small proportions of Phi
cells, which may not get sampled in the 20 or so cells studied
CC, andcan generate such data in samples that are not
by CG, could provide transcript for such molecular detection.
analyzable by CC.
Clearly, more patients with this syndrome need to be studied
by HMF before such an explanation for the basis of Ph-,
RT-PCR' CML can be generally applied.
From the results reported here, HMF is clearly a highly
In patients with Ph+ CML, a interferon therapy produces
method for quantitating the level of cancer burden,
hematologic responses in 70% to 80% and cytogenetic reand possibly also MRD, in hematologic malignancies. While
sponses in 30% to 50%. Cytogenetic responses are major
RT-PCR may be more sensitive" than HMF in identifying
and durable in 15% to 30% of patients who continue therapy
and are associated with significant survival pr~longation.'.~ MRD, it is not enumerative or quantitative and cannot provide information on the constitutive level of Ph' cells, as it
Thus, monitoring cytogenetic response at frequent intervals
depends on the expression of chimeric transcript, levels of
is essential for patient care delivery.
which may vary in cells carrying the translocation. However,
Several problems arise inthe monitoring of patient reit should be realized that, unlike RT-PCR, HMF monitors
sponse to alpha interferon therapy. The time-consuming CG
only cycling cells. interphase FISH,'2"4 on the other hand,
procedures and the antiproliferative effect of alpha interferon
looks at the constitutive level of cancer cells and does not
reduce the number of metaphases analyzed to 25 or less;
depend on the cycling status of cells under study. However.
thus, confidence in the measurement is low. For example,
substantial levels of false-positive and false-negative readin a patient with a complete cytogenetic response (0% Ph+)
ings make quantitative assessment of interphase FISH results
based on 20-metaphase analysis, the confidence interval in
difficult. Genomic PCR, which also does notdepend on
the cytogenetic response is 0% to 11%. Therefore, many
expression of the chimeric locus to be identified, has not yet
patients may have only a partial cytogenetic response, with
been shown to be practical in identifying translocations in
the consequent therapeutic and prognostic implications.
patients because of thewiderange
of positions
More significantly, 10% to20% of samples analyzed on
where translocation breakpoint sites may occur within the
alpha interferon therapy have no metaphases for analysis,
large introns that are interrupted by the event. However,
requiring interruption of interferon for short periods (3 to 7
new developments in tailoring genomic PCR for individual
days) before CC analysis, repetition of tests, and delays of
may soon makethattechnology available tothis
treatment decisions.
field,'','' Further work in which these various technologies
With respect to the assessment of minimal residual disease
are performed on the same MRD samples for which HMF
during therapy, we have shown that complete cytogenetic
been performed may help define the relevance of data
response by CC may in fact be only a partial cytogenetic
all these procedures to clinical status and patient outresponse by HMF, or it may be a complete cytogenetic recome.
sponse of high quality (0 Ph+ cells in 500 cells analyzed).
The major limitation of HMF is the inability of the proceThis should help in decisions regarding continuation of alpha
to detect cytogenetic clonal evolution, an indicator of
interferon therapy andor addition of other agents (eg, lowworse
prognosis," as itis focussed on scoring the single
dose cytosine arabinoside [ara-C]), dose schedule adjustcytogenetic event of interest (in this case, the Ph chromoment, or even discontinuation of therapy.
some). For patients with CML undergoing alpha interferon
HMF addresses these concerns. In this analysis, 24% of
this limitation may be without serious clinical conpatients were reclassified into a cytogenetic response catesequences, asof the 51 CML samples for which CC data
gory that is more appropriate than the one assigned by CC
were available, none displayed clonal evolution. There is
(Table 5). It was also successful in providing response data
also recent evidence that clonal evolution may not carry the
in patients with insufficient metaphases for regular CC studsame poor prognostic impact in such cases as when occumng
ies. Of significance was the ability of HMF to detect MRD
with other features of accelerated phase CML.I8
in 3 of 19 patients in complete cytogenetic response by
While HMF canbe applied to any hematologic maligroutine CC,thus providing insight into their true response
particular nonrandom chromosomal restatus, which may influence treatment decisions (dose schedarrangement in a patient is known and for which appropriate
ule, additional agents, need to continue therapy). Further-
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
FISH probes for its identification have been identified, monitoring for other cytogenetic alterations by CG remains a
significant procedure in the management of such disease in
patients. CG also has an irreplaceable role in diagnosis where
the particular nonrandom chromosomal alteration associated
with a patient’s disease needs to be initidly identified before
any follow-up with HMF can be considered. We conclude,
however, that for observing the course of patients having
known chromosomal alterations associated with their disease, HMF appears to be most efficient in monitoring the
percentage of cells with that alteration in the management
of such malignancies. Application of the procedure to other
hematologic malignancies where the nature of the chromosomal event associated with the disease has been identified
and for which FISH probes are avaialable is in progress.
1. Rowley JD: Anew consistent chromosomal abnormality in
chronic myelogenous leukemia identified by quinacrine fluorescence
and Giemsa staining. Nature 243:290, 1973
2. Trujillo JM, Cork A, Ahearn MJ, Youness EL, McCredie KB:
Hematologic and cytologic characterization of 8/21 translocation in
acute granulocytic leukemia. Blood 53:695, 1979
3. Liu P, Siciliano J, Seong D, Craig J, Zhao Y, de Song PI,
Siciliano MJ: Dual Alu PCR primers and conditions for isolation of
human chromosome painting probes from hybrid cells. Cancer Genet
Cytogenet 65:93, 1993
4. Seong DC, Liu P, Siciliano J, Zhao Y, Cork A, Henske E,
Warburton D, Champlin R, Trujillo JM, Deisseroth A, Siciliano
MJ: Detection of variant Ph’-positive chronic myelogenous leukemia
(CML) involving chromosomes #l, #9,and #22 by fluorescence in
situ hybridization. Cancer Genet Cytogenet 65:100, 1993
5. Talpaz M, Kantarjian HM, Keating MJ, Talpaz M, McCredie
KB, Freireich J: Clinical investigation of human alpha interferon in
chronic myeloid leukemia. Blood 69:1280, 1987
6. Henske EP, Ozelius L, Anderson MA, Kwiatkowski DJ: A
radiation-reduced hybrid cell line containing 5 Mb/17 CM of human
DNA from 9q34. Genomics 132341, 1992
7. Bickel PI, Dobsum KA: Mathematical Statistics: Basic Ideas
and Selected Topics. San Francisco, CA, Holden-Day, 1977
8. The Italian Cooperative Study Group on Chronic Myeloid Leukemia: Interferon alpha-2A as compared with conventional chemotherapy for the treatment of chronic myeloid leukemia. N Engl J
Med 330820, 1994
9. Kantarjian HM, Smith TL, O’Brien S, Beran M, Robertson
LE, Koller C, Pierce S, Estey E, Keating M, Talpaz M: Prolonged
survival in chronic myelogenous leukemia following cytogenetic
response to alpha interferon therapy. Ann Intern Med 122:254, 1995
10. Bernstein R, Pinto MR, Rosendorff J, Kramer S, Mendelow
B: “Masked” Ph’ chromosome abnormalities in CML: A report of
two unique cases. Blood 63:399, 1984
11. Lee MS, LeMaistre A, Kantarjian H, Talpaz M, Freireich E,
Trujillo J, Stass S: Detection of two alternative bcr/abl MRNA junctions and minimal residual disease in Philadelphia chromosome positive chronic myelogenous leukemia by polymerase chain reaction.
Blood 73:2165, 1989
12. Tkachuk DC, Westbrook CA, Andreeff M, Donlon TA,
Cleary ML, Suryanarayan K, Homge M, Redner A, Gray J, Pinkel
D: Detection of bcr-abl fusion in chronic myelogenous leukemia by
in situ hybridization. Science 250:559, 1990
13. Bentz M, Cabot G , Moos M, Speicher MR, Ganser A, Lichter
P, Dohner H: Detection of chimeric BCR-ABL genes on bone marrow
samples and blood smears in chronic myeloid and acute lymphoblastic leukemia by in situ hybridization. Blood 83:1922, 1994
14. Seong DC, Song MY, Henske E, Zimmerman SO, Champlin
RE, Deisseroth AB, Siciliano MJ: Analysis of interphase cells for
the Philadelphia translocation using painting probe made by interalu PCR from a radiation hybrid. Blood 83:2268, 1994
15. Miller W, Blechert G: Inverse PCR across bcr/abl genomic
breakpoint allows sensitive detection of Ph-containing cells. Blood
84:494a, 1994 (suppl 1)
16. Zhang JG, Lin F, Cross NCP, Goldman JM: Detection of
residual disease after BMT for CML by amplification of the bcr-ab1
breakpoint from genomic DNA: Comparison with RT-PCR. Blood
84:496a, 1994 (suppl 1)
17. Kantarjian HM, Deisseroth A, Kurzrock R, Estrov Z, Talpaz
M: Chronic myelogenous leukemia: A concise update. Blood 82:691,
18. Majlis A, Kantarjian H, Smith T, Talpaz M, O’Brien S, Keating M, Freireich E: What is the significance of cytogenetic clonal
evolution (CE) in patients (PTS) with Philadelphia chromosome
(Ph)-positive chromonic myelogenous leukemia (CML)? Blood
84:150a, 1994 (suppl 1)
From www.bloodjournal.org by guest on January 20, 2015. For personal use only.
1995 86: 2343-2349
Hypermetaphase fluorescence in situ hybridization for quantitative
monitoring of Philadelphia chromosome-positive cells in patients
with chronic myelogenous leukemia during treatment
DC Seong, HM Kantarjian, JY Ro, M Talpaz, J Xu, JR Robinson, AB Deisseroth, RE Champlin
and MJ Siciliano
Updated information and services can be found at:
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
Information about ordering reprints may be found online at:
Information about subscriptions and ASH membership may be found online at:
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.

Similar documents


Report this document