Hematopathology / QUANTITATIVE BCR-ABL ASSAY
Comprehensive Validation of a Real-Time Quantitative
bcr-abl Assay for Clinical Laboratory Use
Carol D. Jones, Cecilia Yeung, and James L. Zehnder, MD
Key Words: bcr-abl; Quantitative RT-PCR; real-time RT-PCR; Reverse transcriptase–polymerase chain reaction
DOI: 10.1309/60A9C8WGEGHRNXEE
Abstract
We developed and extensively validated a real-time
PCR assay for the quantitation of bcr-abl to determine
residual disease in patients with chronic myelogenous
leukemia. This method quantitates the p210 and the
p190 bcr-abl RNA fusion transcripts with results
normalized to a housekeeping gene, using the 5'-
exonuclease technique and the ABI PRISM 7700
Sequence Detection System (Applied Biosystems, Foster
City, CA). We parallel tested 372 clinical specimens and
50 peripheral blood samples from patients not known to
have any myeloproliferative disorders. The results were
100% specific. Sensitivity studies showed that this
method can detect bcr-abl in cell lines diluted to
0.0001% and can detect a single bcr-abl plasmid spiked
into negative RNA. The between-run reproducibility
showed a coefficient of variance (CV) of 12.3%, and
within-run reproducibility showed a CV of 13.8%. This
method can be used to reliably monitor the disease load
in patients with bcr-abl–positive diseases.
42 Am J Clin Pathol 2003;120:42-48
42 DOI: 10.1309/60A9C8WGEGHRNXEE
The reciprocal translocation of the abl (Abelson murine
leukemia) proto-oncogene on chromosome 9 to the bcr
(breakpoint cluster region) gene on chromosome 22 creates a
transcriptionally active, chimeric bcr-abl gene and gives rise
to the Philadelphia chromosome. This gene encodes a 190- or
210-kd protein product, which is an active, deregulated, intra-
cellular tyrosine kinase.1 By this mechanism, the fusion
protein is thought to induce malignant transformation of
hematopoietic stem cells.2,3
The bcr-abl messenger RNA (mRNA) transcript is
detectable by molecular methods in approximately 95% of
cases of chronic myelogenous leukemia (CML), as well as
some cases of acute lymphocytic leukemia (ALL; approxi-
mately 5% for children, 20% for adults). It also has been
reported as present in some other hematologic disorders,
albeit rarely.4-6 The bcr-abl gene and its mRNA transcript are
tumor-specific markers. Before the availability of reliable and
accurate quantitative polymerase chain reaction (PCR)
methods, the clinical value of using bcr-abl mRNA PCR to
monitor residual disease in CML was not established clearly.
While a negative PCR result was consistently associated with
remission, a positive result did not necessarily indicate
progression or relapse. However, several reports more
recently have shown that quantitative PCR may be more
informative for minimal residual disease monitoring in these
patients. These reports have shown that increasing values
over time are indicative of progression or relapse.7,8
This report describes the comprehensive validation for
clinical use of a real time, 5'-exonuclease method for the
quantitative detection of the p210 and p190 bcr-abl fusion
transcripts, both normalized to the housekeeping gene glycer-
aldehyde-3-phosphate dehydrogenase (GAPD).
© American Society for Clinical Pathology

Our method uses a 5'-exonuclease technique to generate
fluorescence during the PCR process that is proportional to
the starting quantity of template, which is monitored in real-
time by using an ABI PRISM 7700 Sequence Detection
System (Applied Biosystems, Foster City, CA).9,10 Briefly,
the principle of this method, called TaqMan (Applied
Biosystems), is as follows: A sequence-specific probe is
included in the PCR reaction, which is dual labeled with a
reporter fluorophore at the 5'-end and a quencher fluo-
rophore at the 3'-end. The probe anneals to the region of
interest internal to the forward and reverse PCR primers.
During the extension phase, the 5'→3' exonuclease activity
of the Taq polymerase will cleave the reporter molecule. Its
release is accompanied by a fluorescent signal, which the
ABI 7700 instrument measures and records in real time.
The assay described herein quantitates the bcr-abl p210
and p190 transcripts, as well as a housekeeping gene, GAPD.
We demonstrated that this assay is highly specific and that its
sensitivity is such that it can reliably detect and quantitate 10
pg of bcr-abl RNA (0.0001%) in a background of negative
RNA.
Materials and Methods
Samples
We performed quantitative analyses on all specimens
submitted to the Clinical Molecular Diagnosis Laboratory at
Stanford University Medical Center, Stanford, CA, from
September 1996 through May 2001 on which our standard
qualitative bcr-abl reverse transcriptase (RT)-PCR was
requested and performed. There were 372 bone marrow,
blood, and pheresis product specimens from 145 patients for
which sufficient RNA and/or complementary DNA (cDNA)
remained in our archives. In addition, we tested peripheral
blood samples from 50 patients not known to have any
myeloproliferative disorders.
Quantitative RT-PCR
RNA was isolated from peripheral blood or bone
marrow leukocytes by a modified guanidinium-thiocyanate
method according to the manufacturer’s instructions (RNA-
STAT, Tel-Test, Friendswood, TX, or TRIzol, Invitrogen,
Carlsbad, CA). The reverse transcription reaction contained
the following in a 40-µL volume: 300 U of M-MLV enzyme,
a 2.5-mmol/L concentration of dithiothreitol, 1× RT buffer
(all from Invitrogen); 40 U of RNasin (Promega, Madison,
WI); 5-µg random hexamers (Amersham Biosciences,
Piscataway, NJ); a 1-mmol/L concentration of deoxynucleo-
side triphosphates (Amersham Biosciences); and 10 µg of
total RNA. The reaction was incubated at 37°C for 2 hours.
© American Society for Clinical Pathology
Hematopathology / ORIGINAL ARTICLE
The quantitative real-time PCR was performed in the
ABI 7700 with cycling as follows: an initial cycle for 10
minutes at 95°C, followed by 40 biphasic cycles of 15
seconds at 95°C and 1 minute at 60°C. Each 96-well plate
included standard curves for p210, p190, and GAPD, as well
as sensitivity controls, no-template controls, and no-RT
controls. Patient samples were run in triplicate for each target
(p210, p190, and GAPD), and there was sufficient space for
4 patient samples on each plate. Each well contained 2.5 U
of TaqGold, 1× PCR buffer, a 5.5-mmol/L concentration of
magnesium chloride, and a 250-µmol/L concentration of
deoxynucleoside triphosphates (all from Applied
Biosystems) and 2 µL of cDNA to constitute a final volume
of 25 µL. The GAPD wells included 50 pmol each of GAPD
forward and reverse primers and 10 pmol of GAPD probe.
The p210 wells included 10 pmol each of the p210 and abl
primers and 10 pmol of abl probe. The p190 wells included
10 pmol each of the p190 and abl primers and 10 pmol of abl
probe. Primers and probes were manufactured by Operon
Technologies, Alameda, CA.
The p210 and p190 reaction used the same reverse
primer and probe (for abl), while the forward primer was
specific for the bcr breakpoint region. Each probe included
FAM (6-carboxy-fluorescein; emission, 518 nm) at the 5'-
end as the reporter and TAMRA (6-carboxy-tetramethyl-
rhodamine; emission, 582 nm) at the 3'-end as the quencher.
The bcr-abl primer and probe sequences were designed using
Primer Express software (Applied Biosystems). The GAPD
primer and probe sequences were adapted from the
sequences used in Applied Biosystems’ GAPD Control
Reagents Kit. However, we used FAM as the reporting fluo-
rophore, rather than VIC. Primer and probe sequences are
listed in ❚Table 1❚.
Qualitative RT-PCR
Our qualitative RT-PCR method used for routine clinical
laboratory testing was adapted from Cross.11 It is a multiplex
RT-PCR assay using only a single round of amplification.
❚Table 1❚❚
Primer and Probe Sequences for Quantitative bcr-abl
Determination
Primer Sequence
GAPD forward 5’-GAAGGTGAAGGTCGGAGTC-3’
GAPD reverse 5’-GAAGATGGTGATGGGATTTC-3’
GAPD probe 5’-FAM-CAAGCTTCCCGTTCTCAGCC-TAMRA-3’
p210 forward 5’-TGCTGACCAACTCGTGTGTG-3’
p190 forward 5’-AACTCGCAACAGTCCTTCGAC-3’
abl reverse 5’-CCATTCCCCATTGTGATTATAGC-3’
abl probe 5’-FAM-AAGACCCGGAGCTTTTCACCTTTAGTT
ATGC-TAMRA-3’
FAM, 6-carboxy-fluorescein; GAPD, glyceraldehyde-3-phosphate dehydrogenase;
TAMRA, 6-carboxy-tetramethyl-rhodamine.
Am J Clin Pathol 2003;120:42-48 43
43 DOI: 10.1309/60A9C8WGEGHRNXEE 43
Jones et al / QUANTITATIVE BCR-ABL ASSAY
The multiplex primer set includes primers for the normal bcr
transcript, the p210 fusion transcript, and the p190 fusion
transcript. Primer sequences and primers used are given in
❚Table 2❚ and ❚Table 3❚.
Standards
All standards described were assayed in duplicate and
plotted using the SDS software package for the ABI 7700.
Standard curves were considered valid if they achieved a
correlation coefficient of more than 0.990 with a slope of
–3.1 to –3.5. Standard series for each of the 3 targets
(GAPD, bcr-abl p210, and bcr-abl p190) were run on every
plate in parallel with patient samples. Patient results must
fall within the linear range for each target to be quantitated.
For the quantity to be sufficient for testing, the GAPD must
yield results of at least 0.5 µg of RNA equivalent.
Standards for the bcr-abl p210 and p190 assays were
plasmids prepared from the PCR amplicons generated by the
described primer sets. Fresh PCR products were cloned into
a pCR 2.1 vector using the TOPO-TA cloning kit according
to the manufacturer’s instructions (Invitrogen), and the plas-
mids were purified using the QIAprep mini-prep kit
(QIAgen, Valencia, CA). The identities of the purified plas-
mids were confirmed by sequencing (ABI BigDye kit and
ABI310, Applied Biosystems). The plasmids were linearized
by Hind III digestion (New England Biolabs, Beverly, MA)
before quantitation and use. Each plate included a series of 6
standards: 10-fold serial dilutions starting at 105 targets per
well, down to 100 targets per well, for each of the p210 and
p190 targets. An example of the amplification plots with the
associated standard curve for the p210 fusion transcript is
shown in ❚Figure 1❚.
The GAPD standards were prepared from commercial
human RNA (Stratagene, La Jolla, CA), using the same
reverse transcription protocol as described. The series was
composed of five 5-fold dilutions of cDNA corresponding to
an initial RNA quantity per RT reaction of 18, 3.6, 0.72,
0.144, and 0.03 µg.
Sensitivity
We assessed the sensitivity of the method in 2 ways.
The recombinant bcr-abl plasmids were used to spike
❚Table 2❚❚
Primer Sequences for Qualitative bcr-abl Determination*
Primer Sequence
BCR-e1 5’-ACCGCATGTTCCGGGACAAAAG-3’
BCR-b2 5’-ACAGAATTCCGCTGACCATCAATAAG-3’
BCR-rev 5’-ATAGGATCCTTTGCAACCGGGYCYGAA-3’
ABL-a2 5’-TGTTGACTGGCGTGATGTAGTTGCTTGG-3’
* From Cross.11
44 Am J Clin Pathol 2003;120:42-48
44 DOI: 10.1309/60A9C8WGEGHRNXEE
bcr-abl–negative RNA with known quantities of targets.
Tenfold serial dilutions of plasmid were spiked into 5-µg
aliquots of negative RNA (derived from Stanford lymphoma
cell line OCI-Ly8) and reverse transcribed as described. The
cDNA then was used as a template for the real-time PCR reac-
tion. The 10-fold plasmid dilutions spanned 7 logs (106 to 100).
In addition, we performed the quantitative bcr-abl assay
on serial 10-fold dilutions of RNA from a p210-positive cell
line (Meg-01, American Type Culture Collection, Manassas,
VA) and a p190-positive cell line (JD1, courtesy of Michael
Cleary, MD, Stanford University) using bcr-abl–negative
RNA as the diluent. The undiluted cell line RNA was called
100%, because we assumed that all of the cells in the bcr-abl
cell line (ie, 100%) contained the fusion transcript.
Following this model, the 10-fold dilution was called 10%,
the 100-fold was called 1%, and so on. According to our
procedure, 10 µg of each RNA dilution was used for the
reverse-transcription reaction. The serial 10-fold dilutions
spanned 9 logs, from 100% (containing 10 µg of bcr-
abl–positive cell line RNA) to 0.000001% (containing 0.1 pg
of bcr-abl–positive cell line RNA).
Reproducibility
Between-run reproducibility of this method was
assessed by performing the assay on 5 different patient
samples on a single plate, which was repeated on the same
cDNA samples on 3 different days. Within-run repro-
ducibility was evaluated by performing replicate analyses of
a single specimen in each well of a 96-well plate.
Data Analysis
By using the Sequence Detector System software
package for the ABI 7700, the mean threshold cycles for
GAPD, p210, and p190 were obtained for each sample.
Again using the SDS software, the mean threshold cycle was
applied to the appropriate standard curve (run in parallel
with the sample) to derive the target number (for p210 and
p190) or micrograms of RNA equivalents (for GAPD). The
values derived from the standard curves then were used to
calculate 2 simple ratios: the number of p210 transcripts per
microgram of RNA equivalents and the number of p190
transcripts per microgram of RNA equivalents. The term
❚Table 3❚❚
Expected Amplicon Sizes*
Transcript Primers Used Size (base pairs)
Normal bcr BCR-b2 and BCR-rev 808
e1a2 BCR-e1 and ABL-a2 481
b3a2 BCR-b2 and ABL-a2 385
b2a2 BCR-b2 and ABL-a2 310
* See Table 2 for the primer sequences.
© American Society for Clinical Pathology
 Hematopathology / ORIGINAL ARTICLE

A B

101 55.00
50.00
45.00
100

Threshold Cycle
40.00
∆ Rn

35.00
30.00
10–1
25.00
A B C D E F 20.00
15.00
10–2 10.00
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 5.00
Cycle 100 101 102 103 104 105 106
Starting Quantity

❚Figure 1❚ Example of bcr-abl p210 standards. A, Amplification, bcr032801. Standards were as follows: A, 105; B, 104; C, 103; D,
102; E, 101; and F, 100. ∆Rn, change in fluorescence. B, Standard curve, bcr032801. Slope, –3.499; Y-intercept, 33.670;
correlation coefficient, 0.998. Red, unknown; black, standards.

RNA equivalents rather than RNA was chosen because the
result represents the amount of signal present in the 2 µL of
cDNA rather than some direct measurement of the RNA itself.
Results
Specificity
The results of the parallel testing of 372 patient samples
are given in ❚Table 4❚.
In our parallel-tested specimens, the qualitative results
by breakpoint were as follows: p210-positive, 180 (48.4%);
p190-positive, 22 (5.9%); and negative, 170 (45.7%). The
quantitative results completely correlated by breakpoint.
Moreover, 52 (28.9%) of the 180 p210-positive specimens
showed a small amount of p190 positivity as well, as
reported previously to be a common occurrence.12
We performed our quantitative bcr-abl assay on 50
samples from people not known to have any myeloproliferative
disorders. This entire cohort was negative for both the p210
and p190 fusion transcripts, further demonstrating the speci-
ficity of the assay.
possibly because the amplicon is larger (284 base pairs [bp]
for the b2a2-p210 or 359 bp for the b3a2-p210 vs 381 bp
for p190), which, according to the manufacturer, can
decrease the amplification efficiency in the TaqMan
system. The 10-pg sensitivity limit is equivalent to a 1 in
106 dilution (0.0001%).
Of the 372 parallel-tested samples, 170 (45.7%) were
negative by the qualitative method. Of the 372 samples, 102
(27.4%) also were negative by the quantitative method, but
68 (18.3%) showed some p210 positivity by the quantitative
method, albeit at a low level that cannot be accurately quanti-
tated. In other words, the signal was less than that of our
lowest standard, which necessarily defines the linearity limits
of the assay. This phenomenon is considered further in the
discussion section.
Reproducibility
Between-run reproducibility showed a coefficient of
variance of 12.3%, and the within-run reproducibility
showed a coefficient of variance of 13.8%.
❚Table 4❚❚
Results of 372 Parallel-Tested Clinical Samples*

Sensitivity Qualitative Assay Results

Sensitivity with plasmid-spiked RNA showed that we Positive Negative
could detect 1 (plasmid) bcr-abl target spiked into a
reverse-transcription reaction in a background of negative Quantitative assay results
Positive 202 (54.3) 0 (0)
RNA. By using dilutions of RNA from cell lines, we were Weak positive† 0 (0) 68 (18.3)
able to reliably detect and quantitate 1 pg of p210-positive Negative 0 (0) 102 (27.4)
RNA (0.00001%) or 10 pg of p190-positive RNA * Data are given as number (percentage).
†
(0.0001%) in a background of negative RNA. The p190- We detected bcr-abl by the quantitative method but at a level below the linear range
of the standard curve, which precludes accurate quantitation. See text for further
positive assay is 1 log less sensitive than the p210 assay, discussion.

© American Society for Clinical Pathology Am J Clin Pathol 2003;120:42-48 45

45 DOI: 10.1309/60A9C8WGEGHRNXEE 45
Jones et al / QUANTITATIVE BCR-ABL ASSAY

Discussion These results generated a reproducible, exponential signal,

The clinical usefulness of molecular detection of the
bcr-abl transcript in a posttreatment CML setting is contro-
versial.7 By qualitative methods, a negative result has been
shown to be consistent with ongoing remission, but the
meaning of a positive result was unclear.13 Indeed, most
patients with CML show qualitative PCR positivity in the
immediate peritransplant period, but the results for approxi-
mately two thirds become negative within 6 to 12 months.
Continuing PCR negativity after 1 year seems to be a good
prognostic indicator, while patients showing positivity after 6
months seem to be at increased risk of relapse. Despite this
statistically increased risk of relapse, qualitative PCR posi-
tivity does not predict relapse in individual patients.7
However, with the advent of reliable quantitative tech-
niques, it has been shown that increasing transcript numbers
precede relapse or progression.14-18 Quantitative bcr-abl
detection also has been shown to be reliable for evaluating
mobilized peripheral blood stem cells in patients with
CML.19 ❚Figure 2❚, ❚Figure 3❚, and ❚Figure 4❚ are representa-
tive examples of longitudinal follow-up results for selected
patients from the 145 whose samples were used in the
present study. The clinical usefulness of the assay has been
evaluated extensively by other investigators.14,16,20
The real-time quantitative method described in this
report is more sensitive than the nonnested, qualitative RT-
PCR routinely performed in our clinical laboratory and with
which this method was parallel tested. Of the 372 parallel-
tested samples, 68 samples (18.3%) showed some p210 posi-
tivity by the quantitative method but were negative by the
qualitative method, as mentioned in the Results section.
but the signal was less than that of our lowest standard,
which necessarily defines the linearity limits of the assay.
Hence, these results cannot be quantitated accurately, and we
refer to such results as weak positive. While the results were
not confirmed by sequencing, the TaqMan probe added
specificity to the reaction, similar to the specificity gained
from using probe hybridization techniques following elec-
trophoresis of PCR products. Furthermore, the signals gener-
ated were exponential and seen in all 3 of the triplicate wells.
❚Figure 5❚ is an example of one such weak-positive result,
shown alongside the standard series and a negative result
from the same plate. Finally, these weak-positive results
were seen only in specimens from patients with a history of
bcr-abl–positive CML. While there have been reports of
detecting bcr-abl transcripts in people without myeloprolifer-
ative disorders, 21 they occur in approximately 1 in 10 8
WBCs, a sensitivity several logs in excess of the sensitivity
of the assay described herein. Accordingly, we do not believe
our assay will detect such low circulating levels of bcr-abl.
We conclude, therefore, that these results are due to an
increased sensitivity in this method over the nonnested, quali-
tative RT-PCR used for parallel testing. This conclusion is
further supported by the higher sensitivity also seen in our
sensitivity studies. The nonnested, qualitative RT-PCR
showed a sensitivity of 1 in 105, while the quantitative method
was 10-fold more sensitive at 1 in 106. Accordingly, we would
expect to detect more positive results with the latter.
Nested methods for qualitative RT-PCR assays of bcr-
abl, using 2 rounds of PCR amplification, have been reported
by others to show a sensitivity of 0.0001% (1 in 106 dilution),
which is equivalent to that of this quantitative method.19

1,000,000
1,200,000 BMT #2
900,000
p210/µg RNA Equivalents

BMT #1
p210/µg RNA Equivalents

8,000 800,000
DLI
7,000 700,000
6,000 600,000
5,000 500,000
4,000 400,000
3,000 300,000
2,000 200,000
1,000 100,000
0
0
30

0
30
60

d+ 0
0

d+ 0
d+ 0
d+ 0
d+ 0
d+ 0
0

d+ 0
0
21
d+

9
12

15
18

24
27

33
36
30

30

60

90

0
d–

d+
d+
d+

21
d–

12

15

18

24

27

30
d+

d+

d+

d+

d+

d+
d+
d+

d+

d+

d+

d+

Time From BMT Time From BMT

❚Figure 2❚ Longitudinal quantitative bcr-abl results. Patient
with no relapse following allogeneic bone marrow
transplantation (BMT).
❚Figure 3❚ Longitudinal quantitative bcr-abl results. Patient with
relapse at day +100 after allogeneic bone marrow transplan-
tation (BMT), who subsequently received donor lymphocyte
infusion (DLI) but died at day +996 following the second BMT.
Increased p210 values correlated with clinical relapse.

46 Am J Clin Pathol 2003;120:42-48 © American Society for Clinical Pathology

46 DOI: 10.1309/60A9C8WGEGHRNXEE
 Hematopathology / ORIGINAL ARTICLE

A number of quantitative bcr-abl methods have been STI

2,200,000
published, although in our opinion, none fully met our require-
ments for a clinical assay. The techniques included competi-

p210/µg RNA Equivalents

450,000 BMT
tive PCR,16 which is more cumbersome to perform and only
semiquantitative, and nested methods.19,20 Furthermore, the
60,000 DLI
validity of quantitation in nested PCR has not been established
50,000
adequately.13 Among the real-time PCR methods in the litera-
ture, no single offering met all of our criteria, which include 40,000

quantitation of both the major (p210) and the minor (p190) 30,000

breakpoint transcripts, using absolute quantitation in the data 20,000

analysis, and using a rigorous validation process examining a 10,000
sufficiently numerous and varied group of specimens. 0

d+150
d–30
d+30
d+90

d+210
d+270
d+330
d+395
d+455
d+515
d+575

Only Kreuzer et al14 included a measurement of the
p190 transcript in their method. We included the p190 tran-
script because a previous report indicated possible prognostic
significance of the p190 transcript in p210-positive patients.22
However, to our knowledge, this result has not been reported
by others, and we did not see significant p190 expression in
our patients with p210. Therefore, at present, there seems to
be little usefulness in measuring the p190 transcript in
patients for whom p210 is the dominant transcript.
However, a second reason we chose to include the p190
transcript was so that p190-positive patients could be moni-
tored in our laboratory. Although uncommon in CML, the
p190 transcript is the dominant transcript in approximately
two thirds of Ph-positive ALL. 23 Hence, a substantial
number of patients monitored at this institution would be
excluded from the benefits of quantitative bcr-abl moni-
toring were we to eliminate this breakpoint altogether.
However, it is not necessary to run the transcripts in tandem
for all patients to provide that service.
To our knowledge, this is the first report of a p210 and
p190 TaqMan method that used extensive parallel testing
with the existing method. Saffroy et al17 examined 186 spec-
d+695
d+760
d+820
d+880
d+940
d+635
Time From BMT
❚Figure 4❚ Longitudinal quantitative bcr-abl results. Patient
with relapse at day +377 after allogeneic bone marrow
transplantation (BMT), who subsequently received donor
lymphocyte infusion (DLI) and imatinib mesylate (STI).
Increased p210 values correlated with clinical relapse.
products from multiple clinical sites and patient groups.
We established and validated a quantitative bcr-abl
method that is rapid and easy to perform, reproducible,
sensitive, and specific and includes the p210 and the p190
transcripts. The sensitivity, specificity, and reproducibility
compare favorably with other real-time methods described in
the literature. There are 2 important differences between our
method and previously described methods. One is the detec-
tion of the p190 transcript, which permits us to offer
101

imens and compared the results with RT-PCR and cytoge-

netics results. Elmaagacli et al18 examined 343 specimens 100

and compared their results with the results of cytogenetics

and fluorescence in situ hybridization analysis. However,
10–1
neither measured nor validated the p190 transcript, which
A B C D E F
we also wanted to measure in our laboratory. Others studied
only selected groups of patients.24-27
10–2 Weakly
For the purposes of validation, we tested in parallel an positive
unselected series of specimens rather than evaluating 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44
samples from only a selected group of patients. Our series Cycle
included everything submitted to the clinical molecular diag-
❚Figure 5❚ Example of a weak-positive result. Amplification
nosis laboratory for which the request included the currently
curves of the standards are shown in blue, of the weak-
available bcr-abl test. In this way, our validation study
positive sample in red, and of a negative sample in green.
included specimens across the spectrum of specimen types, Samples were run in triplicate. Note that the weak-positive
collection variables, patient variables, quality, and quantity result shows exponential amplification (red), while the
similar to what we could expect to be evaluating by the new negative does not (green). Standards were as follows:
method. The specimen archive yielded sufficient material on A, 105; B, 104; C, 103; D, 102; E, 101; and F, 100. ∆Rn, change
372 specimens including blood, bone marrow, and pheresis in fluorescence.

© American Society for Clinical Pathology Am J Clin Pathol 2003;120:42-48 47

47 DOI: 10.1309/60A9C8WGEGHRNXEE 47
Jones et al / QUANTITATIVE BCR-ABL ASSAY
quantitative bcr-abl monitoring to patients with Ph-positive
ALL served by this institution. The second is the extensive
parallel testing against an unsorted series of clinical speci-
mens. Although not established in the present study, it may
be possible in future studies to establish a cutoff value above
which a CML relapse is imminent, permitting earlier inter-
vention and, perhaps, better patient outcomes.
From the Department of Pathology, Stanford University School of
Medicine, Stanford, CA.
Supported by grant 2PO1CA49605 from the National
Institutes of Health, Bethesda, MD.
Address reprint requests to Dr Zehnder: Dept of Pathology,
Stanford University, 300 Pasteur Dr, Mail Code 5324, Stanford,
CA 94305.
Acknowledgment: We are grateful to Daniel Arber, MD, for
critical reading of the manuscript and helpful suggestions.
References
1. Sawyers CL. Chronic myeloid leukemia. N Engl J Med.
1999;340:1330-1340.
2. Daley GQ, van Etten RA, Baltimore D. Induction of chronic
myeloid leukemia by the p210 bcr/abl gene of the Philadelphia
chromosome. Science. 1990;247:824-830.
3. Muller AJ, Young JC, Pendergast AM, et al. BCR first exon
sequences specifically activate the bcr/abl tyrosine kinase
oncogene of Philadelphia chromosome positive human
leukemia. Mol Cell Biol. 1991;11:1785-1792.
4. Svarch E, de la Torre E. Myelomonocytic leukaemia with a
preleukaemic syndrome and Ph1 chromosome in monozygotic
twins. Arch Dis Child. 1977;52:72-74.
5. Hoppin EC, Lewis JP. Polycythemia rubra vera progressing in
Ph1-positive chronic myelogenous leukemia. Ann Intern Med.
1975;83:820-823.
6. Price CM, Rasool FS, Shivji MK, et al. Rearrangement of the
breakpoint cluster region and expression of p210 bcr-abl in a
“masked” Philadelphia chromosome–positive acute myeloid
leukemia. Blood. 1988;72:1829-1832.
7. Hochhaus A, Weisser A, La Rosee P, et al. Detection and
quantification of residual disease in chronic myelogenous
leukemia. Leukemia. 2000;14:998-1005.
8. Lin F, van Rhee F, Goldman JM, et al. Kinetics of increasing
bcr-abl transcript numbers in chronic myeloid leukemia
patients who relapse after bone marrow transplantation.
Blood. 1996;87:4473-4478.
9. Heid CA, Stevens J, Livak KJ, et al. Real-time quantitative
PCR. Genome Res. 1996;6:986-994.
10. Holland PM, Abramson RD, Watson R, et al. Detection of
specific polymerase chain reaction product by utilizing the 5'
to 3' exonuclease activity of Thermus aquaticus DNA
polymerase. Proc Natl Acad Sci U S A. 1991;88:7276-7280.
11. Cross NCP. Detection of bcr-abl in hematological
malignancies by RT-PCR. In: Cotter FE, ed. Molecular
Diagnosis of Cancer. Totowa, NJ: Humana Press; 1996.
12. van Rhee F, Hochhaus A, Lin F, et al. p190 bcr-abl mRNA is
expressed at low levels in p210-positive chronic myeloid and
acute lymphoblastic leukemias. Blood. 1996;87:5213-5217.
48 Am J Clin Pathol 2003;120:42-48
48 DOI: 10.1309/60A9C8WGEGHRNXEE
13. Faderl S, Talpaz M, Kantarjian H, et al. Should polymerase
chain reaction analysis to detect minimal residual disease in
patients with chronic myelogenous leukemia be used in
clinical decision making? Blood. 1999;93:2755-2759.
14. Kreuzer KA, Lass U, Nagel S, et al. Applicability of an absolute
quantitative procedure to monitor intra-individual bcr/abl
transcript kinetics in clinical samples from chronic myelog-
enous leukemia patients. Int J Cancer. 2000;86:741-746.
15. Mensink E, van de Locht A, Schattenberg A, et al.
Quantitation of minimal residual disease in Philadelphia
chromosome positive chronic myeloid leukaemia patients
using real-time quantitative RT-PCR. Br J Haematol.
1998;102:768-774.
16. Olavarria E, Kanfer E, Szydlo R, et al. Early detection of bcr-
abl transcripts by quantitative reverse transcriptase–
polymerase chain reaction predicts outcome after allogeneic
stem cell transplantation for chronic myeloid leukemia.
Blood. 2001;97:1560-1565.
17. Saffroy R, Lemoine A, Brezillon P, et al. Real-time
quantitation of bcr-abl transcripts in haematological
malignancies. Eur J Haematol. 2000;65:258-266.
18. Elmaagacli A, Freist A, Hahn M, et al. Estimating the relapse
stage in chronic myeloid leukaemia patients after allogeneic
stem cell transplantation by the amount of bcr-abl fusion
transcripts detected using a new real-time polymerase chain
reaction method. Br J Haematol. 2001;113:1072-1075.
19. Preudhomme C, Revillion F, Merlat A, et al. Detection of
bcr-abl transcripts in chronic myeloid leukemia (CML) using
a “real time” quantitative RT-PCR assay. Leukemia.
1999;13:957-964.
20. Elmaagacli AH, Beelen DW, Opalka B, et al. The amount of
bcr-abl fusion transcripts detected by the real-time
quantitative polymerase chain reaction method in patients
with Philadelphia chromosome positive chronic myeloid
leukemia correlates with the disease stage. Ann Hematol.
2000;79:424-431.
21. Biernaux C, Loos M, Sels A, et al. Detection of major bcr-abl
gene expression at a very low level in blood cells of some
healthy individuals. Blood. 1995;86:3118-3122.
22. Serrano J, Roman J, Sanchez J, et al. Molecular analysis of
lineage-specific chimerism and minimal residual disease by RT-
PCR of p210 bcr-abl and p190 bcr-abl after allogeneic bone
marrow transplantation for chronic myeloid leukemia:
increasing mixed myeloid chimerism and p190 bcr-abl detection
precede cytogenetic relapse. Blood. 2000;95:2659-2665.
23. Butturini A, Arlinghaus RB, Gale RP. bcr/abl acute leukemia.
Leuk Res. 1996;20:523-529.
24. Emig M, Saussele S, Wittor H, et al. Accurate and rapid
analysis of residual disease in patients with CML using
specific fluorescent hybridization probes for real time
quantitative RT-PCR. Leukemia. 1999;13:1825-1832.
25. Bolufer P, Sanz GF, Barragan E, et al. Rapid quantitative
detection of bcr-abl transcripts in chronic myeloid leukemia
patients by real-time reverse transcriptase polymerase-chain
reaction using fluorescently labeled probes. Haematologica.
2000;85:1248-1254.
26. Amabile M, Giannini B, Testoni N, et al. Real-time
quantification of different types of bcr-abl transcript in
chronic myeloid leukemia. Haematologica. 2001;86:252-259.
27. Eder M, Battmer K, Kafert S, et al. Monitoring of bcr-abl
expression using real-time RT-PCR in CML after bone
marrow or peripheral blood stem cell transplantation.
Leukemia. 1999;13:1383-1389.
© American Society for Clinical Pathology