Guidelines for the measurement of BCR-ABL1 transcripts in chronic myeloid leukaemia.

Abstract

Molecular testing for the BCR-ABL1 fusion gene by real time quantitative polymerase chain reaction (RT-qPCR) is the most sensitive routine approach for monitoring the response to therapy of patients with chronic myeloid leukaemia. In the context of tyrosine kinase inhibitor (TKI) therapy, the technique is most appropriate for patients who have achieved complete cytogenetic remission and can be used to define specific therapeutic milestones. To achieve this effectively, standardization of the laboratory procedures and the interpretation of results are essential. We present here consensus best practice guidelines for RT-qPCR testing, data interpretation and reporting that have been drawn up and agreed by a consortium of 21 testing laboratories in the United Kingdom and Ireland in accordance with the procedures of the UK Clinical Molecular Genetics Society.

Full text

Page 1

Guidelines for the measurement of BCR-ABL1 transcripts in
chronic myeloid leukaemia
Letizia Foroni,1Gill Wilson,2Gareth Gerrard,1Joanne Mason,3David Grimwade,4Helen E. White,5David Gonzalez de Castro,6
Stephen Austin,7Abida Awan,8Emma Burt,9Tim Clench,10Joanna Farruggia,11Jeremy Hancock,12Alexandra E. Irvine,13Aytug
Kizilors,14Stephen Langabeer,15Benedict J. Milner,16Guillermina Nickless,17Anna Schuh,18Anne Sproul,19Lihui Wang,20
Caroline Wickham21and Nicholas C. P. Cross5,22
1Imperial College Academic Health Science Centre, London,2Sheffield Children’s NHS Foundation Trust, Sheffield,3West Midlands
Regional Genetics Laboratory, Birmingham Women’s Hospital, Birmingham,4King’s College London School of Medicine, London,
5National Genetics Reference Laboratory (Wessex), Salisbury,6Royal Marsden NHS Foundation Trust, Sutton,7Cardiff and Vale
University Health Board,8Manchester Royal Infirmary, Manchester,9Barts and the London NHS Trust, London,10Bristol Royal
Infirmary, Bristol,11Derriford Hospital, Plymouth,12Bristol Genetics Laboratory, Southmead Hospital, Bristol,13Centre for Cancer
Research and Cell Biology, Belfast,14Laboratory for Molecular Haemato-Oncology, Kings College Hospital, London,15St. James’s
Hospital, Dublin, Ireland,16Aberdeen Royal Infirmary, Foresterhill, Aberdeen,17Guy’s Hospital, London,18John Radcliffe Hospital,
Oxford,19Western General Hospital, Edinburgh,20The Royal Liverpool University Hospital, Liverpool,21Royal Devon and Exeter
NHS Foundation Trust and Peninsula College of Medicine and Dentistry, and22University of Southampton School of Medicine,
Southampton
Summary
Molecular testing for the BCR-ABL1 fusion gene by real time
quantitative polymerase chain reaction (RT-qPCR) is the most
sensitive routine approach for monitoring the response to
therapy of patients with chronic myeloid leukaemia. In the
context of tyrosine kinase inhibitor (TKI) therapy, the
technique is most appropriate for patients who have achieved
complete cytogenetic remission and can be used to define
specific therapeutic milestones. To achieve this effectively,
standardization of the laboratory procedures and the inter-
pretation of results are essential. We present here consensus
best practice guidelines for RT-qPCR testing, data interpreta-
tion and reporting that have been drawn up and agreed by a
consortium of 21 testing laboratories in the United Kingdom
and Ireland in accordance with the procedures of the UK
Clinical Molecular Genetics Society.
Keywords: BCR, ABL1, real time PCR, guidelines, chronic
myeloid leukaemia.
Chronic myeloid leukaemia (CML), characterized by the
translocation between chromosome 9 (9q34.1) and 22
(22q11.2), is a triphasic disease typically presenting with a
prolonged and generally indolent chronic phase (CP-CML).
This subsequently transforms into an aggressive and almost
invariably fatal acute leukaemia, termed blast crisis (BC), often
via an intermediate accelerated phase (AP). Thirteen years after
the first description of the Philadelphia (Ph) chromosome in
1960 (Nowell & Hungerford, 1960) it was demonstrated that
chromosome 9 was also involved in the pathogenesis of CML
(Rowley,1973),andwasfollowed12 yearslaterbythemolecular
characterization of the BCR-ABL1 (previously termed BCR-
ABL)fusiongene(Shtivelmanet al,1985;Melo&Chuah,2007).
The t(9;22) (q34.1;q11.2) is detected cytogenetically in more
than 95% of CML patients, whilst in the remaining 5% the
fusion gene is ‘cryptic’ and located on an apparently normal
chromosome 22 or, more rarely, on chromosome 9. The
presence of BCR-ABL1 can be demonstrated in these cases by
fluorescence in situ hybridization (FISH) or reverse transcrip-
tase polymerase chain reaction (RT-PCR). BCR-ABL1 also
occurs in approximately 25% of adults and 5% of children
with acute lymphoblastic leukaemia (ALL) (Westbrook et al,
1992) with a translocation which is molecularly indistinguish-
able from the CML cases except for the frequency of one of its
subtype, e1a2, common in ALL (75%) but rarely detected in
CML (?1%), as discussed below.
The most common breakpoint in the ABL1 gene is upstream
of exon 2 (ABL1 a2 fusion type) or, more rarely, downstream
of exon2 (ABL1 a3 fusion type). In the BCR gene, the most
frequent breakpoint is downstream of either exon 13 or exon
14 (e13, e14 previously referred to as exons b2 and b3), leading
to e13a2 or e14a2 mRNA fusion subtypes, respectively. These
are found in approximately 97–98% of CML patients, result in
Correspondence: Letizia Foroni, Department of Haematology,
Imperial Molecular Pathology Laboratory, G Block 2nd floor, room
313; Haematology Department Hammersmith Hospital Du Cane Rd
London W12 OHS, UK.
E-mail: l.foroni@imperial.ac.uk
guideline
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology
doi:10.1111/j.1365-2141.2011.08603.x

Page 2

a p210 oncoprotein, and are referred to as ‘major BCR-ABL1
fusion subtype’. In up to 75% of cases of BCR-ABL1 positive
ALL, but <1% of CML, a breakpoint downstream of BCR exon
1 results in an e1a2 mRNA fusion encoding a p190 oncopro-
tein that is referred to as ‘minor fusion subtype’. The
remaining 2–3% of CML patients are accounted for by a
variety of fusions which involve BCR exons 6, 8 or 19 (e6a2,
e8a2, e19a2) or, as mentioned above, ABL1 exon 3 (most
commonly e13a3 or e14a3 fusions (Hochhaus et al, 1996;
Melo, 1996).
Becauseof thevariety
pre-treatment samples in newly diagnosed CML patients
should always be tested so that MRD monitoring can target
the correct fusion subtype (Cross et al, 1994; Burmeister &
Reinhardt, 2008; Foroni et al, 2009) and false negative results
safely excluded. If pre-treatment fusion subtype has not been
identified, any final test report on follow up samples may need
to be appropriately qualified and a comment added in the
report (For example: ‘Because the fusion subtype prior to
treatment is not known, false negative results cannot be excluded.
Please provide this information if available’).
ofBCR-ABL1transcripts,
Tyrosine kinase inhibitors and CML
The first clinically available tyrosine kinase inhibitor (TKI),
imatinib, was introduced in 1998 (Druker et al, 1996, 2001)
and is now the front-line therapy for CML in chronic phase.
The drug exerts its effect by competitively binding to the ATP
binding pocket of the ABL1 tyrosine kinase domain (TKD) and
thus preventing access to ATP. The second generation TKIs,
dasatinib and nilotinib, have shown very promising results as
first line therapy (Kantarjian et al, 2010; Saglio et al, 2010), but
to date their use in the UK is limited mainly to imatinib-
intolerant and -resistant patients. The most common identified
cause of resistance in TKI-treated cases is the acquisition of
point mutations in the tyrosine kinase domain of BCR-ABL1
that interfere with binding of imatinib and other TKIs but not
with access of ATP to the ATP pocket (O’Hare et al, 2005).
Molecular monitoring
In CML patients, remission is conventionally measured by the
normalization of the total white cell count and the achieve-
ment of complete cytogenetic response (i.e. no identifiable Ph-
positive marrow metaphases in at least 20 bone marrow cells).
Nevertheless, it has been estimated that as many as 107
leukaemic cells may still be present in the body of a patient in
complete cytogenetic remission and prompted the interest in
more sensitive PCR-based techniques. Molecular monitoring
was pioneered in the late 1980s but not widely adopted as a
monitoring tool until the introduction of real time PCR (RT-
qPCR) in the 1990s. However, differences in protocols,
primers, plasmids used for the generation of standard curves,
interpretation of data and the choice of reference gene can
result in widely varying end-results, even for identical samples
tested in different laboratories (Branford & Hughes, 2006;
Cross, 2009). This can lead to lack of confidence in the
reliability of molecular tests by clinicians and a continued
reliance on the less sensitive cytogenetic results.
In recent years there has been an increased effort regarding
the standardization of molecular monitoring and provision of
guidelines in the interpretation of results. (Bustin, 2010) An
International Scale for BCR-ABL1 RT-qPCR measurement was
proposed that is essentially identical to that used in the
International Randomized Study of Interferon versus STI571
(IRIS) trial, i.e. measurements are related to a standardized
baseline (Hughes et al, 2006). The scale is beginning to be
widely adopted by centres that have derived laboratory-specific
conversion factors (Branford et al, 2008) and may be further
facilitated by the recent development of primary, accredited
BCR-ABL1 reference reagents (White et al, 2010). In addition,
the need for robust internal quality control processes has been
emphasized and described in detail elsewhere (Branford et al,
2006; Bustin, 2010).
Whilst we believe that these initiatives are important, they
do not address all aspects of the testing process that can lead
to variable results. As part of a survey of practices in the UK
and Ireland we identified variation in laboratory protocols,
but, more surprisingly, we identified significant differences in
the way in which raw data are analysed. Importantly, these
differences produced different results even when identical
datasetsweredistributedfor
prompted us to develop consensus best practice guidelines
for the UK and Ireland that cover technical aspects of the
test including sample collection and processing, reagents,
RT-qPCR, data analysis and interpretation. While reviewing
the whole process from sample collection to reporting,
particular attention was given to experimental set-up and the
assessment of data quality. The procedure for each step
leading to real time quantification is briefly commented
below with detailed protocols provided in Appendices 1–4.
analysis. Thesefindings
Recommendations for sample processing
Total white blood cells from peripheral blood (PB) should be
collected, preferably in EDTA. 5 mL of PB is adequate at
diagnosis but a minimum of 10–20 mL is recommended for
follow-up samples to achieve adequate sensitivity. Although
there is no rigorous published data to indicate the acceptable
time interval between taking the sample and laboratory
processing, it is generally accepted that samples should be
processed as soon as possible (van der Velden et al 2004;
Hughes et al, 2006). In our experience, samples must be
processed by the laboratory within, and not later than, 72 h
from collection. Processing samples older than 72 h should
only be considered for pre-treatment samples to establish
BCR-ABL1 transcript type. Reliable quantification is not
possible from a follow-up sample older than 72 h. Even
pre-treatment samples that are more than 5 days old should be
discarded and a repeat sample requested as even qualitative
Guideline
2
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology

Page 3

tests may fail. However, some laboratories are processing
samples as old as 5–7 d with some degree of qualitative success
(Appendix 1).
Red cells should be removed by two consecutive treatments
of the blood sample with ammonium chloride lysis (or other
equivalent red cell lysis method) followed by one to two
washes in phosphate buffered saline (PBS). Mononuclear cell
separation is not recommended. The use of lymphoprep
separation will remove the granulocytes, which are the
affected cells, and enrich for lymphocytes (most of which
are BCR-ABL1 negative) thus skewing MRD results. The
entire white cell pellet collected should be lysed in either
guanidinium thiocyanate (GTC) solution, TRIzol (Invitro-
gen), RLT buffer (Qiagen) or equivalents from other
manufacturers. Lysates can be stored overnight at )20?C or
indefinitely (at )80?C).
It should be noted that in contrast to CML, molecular
monitoring of BCR-ABL1 positive ALL patients should be
performed on mononuclear cells derived from bone marrow
(BM) aspirates and not PB samples.
Sample frequency
Follow up monitoring, after complete cytogenetic remission
has been achieved, can be performed using PB samples
collected at three-monthly intervals. This is adequate and in
line with European LeukaemiaNet guidelines (Baccarani et al,
2009) as well as those for most clinical trials in the UK and
Ireland (e.g. SPIRIT1 and SPIRIT2).
Recommendations for RNA extraction
A minimum of 1/3 of the original lysate is required per
extraction but some laboratories extract the whole lysate and
use 1–2 lg of RNA for cDNA synthesis. Optical density (OD)
measurement to estimate RNA concentrations should always
be interpreted with caution due to variable amounts of residual
DNA. This may be negated by incorporating a DNase
treatment step in the extraction protocol (e.g. using the
RNase-Free DNase Set from Qiagen; Appendix 2).
TRIzol extraction achieves a purer RNA yield and OD
readings are, therefore, more reliable.
Poor sample quality may be due to a variety of causes
during the initial steps of sample processing, including (i)
low total white cell count in leucopenic patients, especially
during/after TKI therapy; (ii) samples <100 d post-stem cell
transplantation; (iii) samples more than 72 h old; (iv) poor
quality reagents; (v) poor sample handling procedures, such
as: incubating samples or spinning them at incorrect
temperature or speed; cross-contamination and using solu-
tions prepared with incorrect concentrations of reagents or
pH, can all result in poor RNA preparation and poor cDNA
synthesis.
Recommendations for cDNA synthesis
Synthesis of complementary DNA (cDNA) follows RNA
extraction, using random hexamer primers and either Moloney
murine leukaemia virus reverse transcriptase (MMLV-RT),
Superscript III enzyme (Invitrogen) or High Capacity cDNA
RT kit (AB-Life Technologies). Throughout the RNA extrac-
tion and cDNA synthesis it is of paramount importance to take
care to avoid RNA degradation or cross-contamination of
samples (Appendix 3).
Recommendations for BCR-ABL1 quantification
Procedures related to this part of the BCR-ABL1 test will be
extensively discussed in this section. The use of in house and
published primers/probes as well as using different plasmids as
calibration standards are a great source of variation in results.
At least in the UK and Ireland the comparability of results has
been improved by using centrally provided plasmid dilutions
(NGRL plasmid; unpublished observations), using Europe
Against Cancer (EAC) primers and probes (Gabert et al, 2003)
and ‘high and ‘low’ RQ-PCR controls following procedures
developed by the Adelaide group (Branford & Hughes, 2006).
These RQ-PCR controls are provided as lysates of cell dilutions
of K562 into HL60 in RLT or TRIzol. Their extraction and
cDNA conversion control for several steps of the BCR-ABL1
analysis and help with the process of standardization. The use
of these common reagents across many laboratories is
currently being validated.
Real-time PCR: methodologies and platforms
To increase the sensitivity of detection it is desirable to use a
probe designed to hybridize to a region within the PCR
amplicon (typically a TaqMan dual-labelled hydrolysis probe;
Appendix 4).
Other types of fluorescent technology include Hybridization
Probes (Roche) and dsDNA intercalating dyes such as SYBR
Green (Molecular Probes), although the latter is not recom-
mended due to low specificity. Such methodologies have
replaced qualitative two-step ‘nested’ PCR (Kawasaki et al,
1988) or competitive PCR for BCR-ABL1 transcript quantifi-
cation in CML (Cross et al, 1993), being more accurate, faster,
less prone to contamination and, in many hands, equally
sensitive.
There are many real-time hardware platforms available for
RT-qPCR analysis, with the choice for each laboratory being
dictated mainly by cost and throughput. The precise protocol
for setting up the RT-qPCR may vary depending on the
platform that is employed but should nevertheless be consis-
tent with these guidelines.
Samples should be tested for an internal control gene to
assess the sample quality and quantity using the same amount
of cDNA as for the BCR-ABL1 gene amplification (usually 2Æ5–
Guideline
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology
3

Page 4

5 lL). For the choice of control gene and detailed procedure
see below. Different aspects of the RT-qPCR analysis will be
analysed individually in the next paragraphs.
Standard curves (SC).
1 SC must be derived from a minimum of three dilution
points, although four are recommended (each run in
triplicate or duplicate; single amplification is not acceptable
for the SC).
2 SC should cover the dynamic range of the target if possible;
(a)ABL1: 106–102(105–103for Ipsogen standard users); as
values below 103are not acceptable for ABL1, the use of
a 102dilution may be of limited use for ABL1.
(b)BCR-ABL1: 106–101. It is NOT acceptable to extrapolate
data beyond the points of the curve. Therefore mea-
surement of BCR-ABL1 molecules below ‘10’ (1–10)
may not be accurate.
3 SC should be run for both ABL1 and BCR-ABL1. It is
acceptable to run a standard curve each day for multiple
runs BUT these must use the same master mix as used for
the sample quantification. The threshold should remain
constant throughout.
4 Slopes must be between )3Æ20 and )3Æ60.
Assays should be optimized to yield similar intercept values
for both BCR-ABL1 and ABL1. Variable intercept values
between runs and/or between BCR-ABL1 and ABL1 indicate
suboptimal assay performance and should be addressed.
Threshold. The threshold is used to determine the threshold
cycle (Ct) and is typically set in the log-linear phase of the
amplification curve. Automatic threshold determination is not
recommended,except for
machines. In our collective experience, the use of automatic
threshold determination may lead to spurious results and thus
is not recommended unless thorough internal validation has
been performed. In addition, retaining a fixed threshold is
useful, as in this way evidence of drift can be obtained by
maintaining a record of Ct values for plasmids and high and
low quantification controls, because these should remain
constant over time.
Set at: 0Æ05–0Æ1 for ABI7900/7500/7700, but this may vary
for Corbett Rotor-Gene 6000/or RG3000 (0Æ02–0Æ2) or other
equipment. These values are subject to intra- laboratory
validation and must remain constant throughout the mea-
surements; any change needs to be validated by retesting old
samples.
The baseline delineates between the start of amplification
and the preceding period of background fluorescence. Most
modern RT-qPCR hardware will have an option to deter-
mine this automatically and this is the recommended
setting. For ABI 7900/7500, use auto-baseline; for ABI
7700, set upper limit of baseline at 3 Ct values below lowest
Ct of standard and for other hardware, e.g. Roche Light
Cycler, use automated settings for threshold and baseline
capillary-basedLightCycler
(which currently apply cut-off of 3 Cts below lowest Ct of
standard).
Sample controls
Every time a batch of RNA extraction is performed, negative
RT- and RT-qPCR blanks should be included.
Negative controls.
1 The water used for RNA elution from columns; or to
resuspend RNA pellets.
2Ethanol from RNA preparation (optional);
3RLT solution from each sample processing batch;
4 Non-Template Control (NTC, water).
5In many laboratories BCR-ABL1 negative samples that
are processed simultaneously are true ‘negative/non-
amplified controls’. Under the conditions used for
RT-qPCR it is unlikely to detect the extremely low
level of BCR-ABL1 transcripts in healthy individuals and
therefore these controls are adequate. In other labora-
tories, cell lines negative for the BCR-ABL1 fusion are
used and these procedures may depend on each
laboratory’s protocols.
Note: the NTC control (4) should be run on every plate,
whereas the first three (1–3) can be tested once with each
processing batch. The NTC control should be included on
every run, whereas the first three need to be tested only once
while testing with the batch of samples prepared using the
same reagents. Each laboratory can add or use negative
controls according to their internal standard operating
procedures, but controls for contamination derived from
reagents used in the RNA extraction and cDNA synthesis
must be included. Should contamination occur, all efforts
must be made to identify its source before further samples
are analysed.
Positive controls. Positive controls are required to safeguard
against false negative results due to failure in the amplification
of the most common fusion subtypes. For this purpose a
positive control (e.g. cell lines such as K562 for the e14a2
fusion or SD1 for e1a2; Lozzio & Lozzio, 1977; Dhut et al,
1991) can be included either as lysates to control for all steps in
the process or as cDNA to control specifically for the
amplification/detectionsteps.
monitored by comparing Ct values over time for defined
plasmid dilutions.
Thelatter mayalso be
RT-qPCR run quality control. Each RT-qPCR run should
conform to the following criteria:
1
2
Slope )3Æ20 to )3Æ60
R2> 0Æ980;
Consider rejecting the run if these criteria are not met.
Guideline
4
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology

Page 5

Evaluation of ABL1 amplification
Reference gene analysis. Amplification of an endogenous
control (e.g. BCR, ABL1, G6PD, B2L, GUSB) is used to act as
a reference gene in order to normalize BCR-ABL1 transcript
levels and to highlight poor quality samples. ABL1 remains the
most common reference gene used, due to its wide expression
level across different cell types and its stability (Beillard et al,
2003) and has been implemented by all UK testing laboratories.
The ability to amplify at least 10 000 ABL1 molecules per
reaction volume (i.e. the same volume of cDNA used to test for
BCR-ABL1) is a safeguard against poor sample quality and also
false negative results.
Protocol. All samples must be evaluated for amplification of a
reference gene (ABL1) to verify quality following these
stringent criteria.
1 ABL1 measurement should be done at least in duplicate;
single measurements are not acceptable and ABL1 should
show ‡10 000 ABL1 molecules per sample. The number of
ABL molecules per test refers to the number detected in the
same volume of cDNA used for each BCR-ABL reaction.
(a) The difference between the highest and lowest replicates
should be <0Æ5 Ct.
(b) If one triplicate with an outlying Ct value is removed,
the average is obtained by dividing remaining values by
2 (not by 3). This adjustment is not possible with only
duplicate reactions.
2 ABL1 molecules <10 000 and >5000: ABL1 amplification is
considered sub-optimal (low sensitivity and quantification
may be unreliable).
3 ABL1 molecules <5000 and >1000: only positive results
should be issued with a comment that MRD quantification
is not reliable and must be interpreted with caution. A
repeat sample is strongly recommended.
4 ABL1 < 1000: sample has failed; a repeat sample should be
requested. It is unlikely that even positive tests can be
considered reliable.
For any sample that provides <10 000 ABL1 molecules when
using 2Æ5–5 lL of cDNA either: 1. a new extraction can be
attempted from the original GTC/RLT or TRIzol stored
material; or 2. request a new sample.
However, in case of diagnostic samples it may still be
acceptable to interpret BCR-ABL1 results even when ABL1 has
failedbythecriteriaabove(if<10 000but>1000)andtoprovide
a qualitative and not quantitative analysis. In some cases,
singleplex or multiplex RT-PCR and RT-qPCR analysis can be
combined to confirm the type of pre-treatment fusion subtype.
QuantificationbyRT-qPCRisnotaccurateinpoorsamples.For
followupsamples,ABL1values<10 000caneasilyleadtoafalse
negative result and should be interpreted as ‘failed’ and a new
sample requested.
The ABL1 total copy number (e.g. ABL1 = 3Æ2 · 104
molecules) should be included in the final report. It is not
necessary to include the volume (lL) as the assessment of copy
number for BCR-ABL1 should be performed in the same
volume as ABL1. Some laboratories include Ct values in the
report, but since these may vary depending on equipment and
standard calibrators, Ct values alone should not be used for
inter-laboratory comparison.
Evaluation of BCR-ABL1 amplification
1 It is strongly recommended that BCR-ABL1 measurement
is performed in triplicate.
2 The same volume of cDNA used for the ABL1 gene test
should be used for the BCR-ABL1 quantification.
3One of the triplicates can be excluded if the following
parameters are not fulfilled:
(a) the difference between the highest and lowest replicates
should be <0Æ5 for Ct values up to 30;
(b) the difference between the highest and lowest replicates
should be <1Æ0 for Ct values between 30Æ1 and 33;
(c) the difference between the highest and lowest replicates
should be <1Æ5 Ct for Cts >33Æ1 and <37. Above 37Æ1 Cts,
replicates may show considerable variation and quanti-
fication may be unreliable, and can be reported as
positive outside the quantitative range (POQR);
4 If two of the triplicates are positive and negative controls
show no contamination: the result is positive. If triplicates
have reasonable copy numbers (Ct <38) consider PCR
drop-out for the negative well and adjust mean accord-
ingly, i.e. divide by 2, not by 3.
5A sample with a SINGLE positive result of a triplicate or
duplicate (triplicates recommended) with Ct < 38 MUST
be repeated to confirm positivity;
6If two of the triplicates are negative and one of three is
positivewithCt > 38theresultmustbereportedas‘lowlevel
positivity not quantifiable’. Evaluate whether to repeat test is
dependent on previous molecular and/or clinical history.
7If all triplicates are negative/undetectable, the test should be
reported as ‘undetectable’.
Note: The cut-off for positivity should correspond to the
guidelines set by the EAC programme, which sets the
Intercept + 1 (typically leading to cut-offs of 41–42 Ct (Gabert
et al, 2003), corresponding to a theoretical 0Æ5 molecules.
The above recommendations are valid only when NTC give
no amplification (except for well recognized non-sigmoidal
amplification creep) (Gabert et al, 2003). Its occurrence in
these wells can be non-specific (e.g. ‘creeping baseline’, see
Gabert et al, 2003) or specific as a result of contamination,
which may occasionally occur on a per-plate basis (as a result
of aerosol or splash) or more longitudinally because of reagent
contamination.
To avoid inter-run variation, it may be helpful to perform
the BCR-ABL1 and ABL1 quantification in the same RT-
qPCRrun or, alternatively, as a duplex reaction using differ-
entially labelled probes in the same well (Gerrard et al, 2010
2011, in press).
Guideline
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology
5

Page 6

BCR-ABL1 reporting
BCR-ABL1 ratios are expressed as a percentage (%) as follows:
BCR-ABL1 molecules divided by the total number of ABL1
molecules and multiplied by 100;
For example: BCR-ABL1: 4500 molecules ABL1: 23 000
molecules; Ratio = 4500 divided by 23 000 (= 0Æ195) and
multiplied by 100 (= 19Æ5%).
1 If ABL1 copy number is ‡10 000, results should be both
reliable and sensitive and can be issued as ‘positive’ or
‘undetectable’.
If ABL1 copy number is <10 000 but > 5000: ‘undetectable
BCR-ABL1’ results should not be given (due to the risk of
false negative) and one of the following comments should
be included in the report:
‘The ABL1 measurement is suboptimal (<10 000 molecules)
consequentlyafalse negative
be excluded. A repeat sample should be sent for
testing’.
‘The ABL1 transcript level (<10 000) in this sample suggests
that the quantity/quality of the RNA extracted is
suboptimal, consequently the quantification of disease
level may lack precision and should be treated with
caution’.
3 If ABL1 copy number is between 5000 and 1000 a positive
result can be given BUT a repeat sample should be
requested and the following comment added to the report:
‘Due to the suboptimal quality of the sample, the MRD level
is unlikely to be accurate and no therapeutic decisions should
be based on this result. A repeat sample should be provided’.
No reliable negative result can be issued as this could be a
false negative test due to the limited number of amplifiable
cDNA molecules.
4 If ABL1 is <1000, irrespective of the level of BCR-
ABL1 positivity, the following comment should be
included in the report ‘Failed sample, please provide a
repeat.’
2
MRD resultcannot
Examples
1ABL1: 12 345 molecules; BCR-ABL1: 43; test: positive;
ratio: 0Æ35%*. *Results: e13a2/e14a2 amplification has
been detected by RT-qPCR.
ABL1: 6300 molecules; BCR-ABL1: 0; test: failed*; ratio:
NA. *Results: As fewer than 10 000 or <1E4 ABL1
molecules have been amplified from this specimen, MRD
results are not reliable and false negative results cannot
be excluded.A newsample
re-testing.
ABL1: 6300 molecules; BCR-ABL1: 15; test: failed*; ratio:
0Æ24%*. *Results: Because of the poor sample quality
(ABL1 < 10 000 4), MRD results must be interpreted
with great caution and a repeat sample is recommended
for retesting.
2
isrecommended for
3
Conversion
validated conversion factor should also consider converting
their BCR-ABL1/ABL1 values to the International Scale.
However this should be clearly specified in the final report
to the clinicians.
factor. Laboratories withan appropriately
Useful additional guidelines
It is very useful to record the Ct values of the standards on
every run. This should remain constant over time providing
the same threshold for the particular real-time machine (0Æ05–
0Æ1 for TaqMan and 0Æ02–0Æ2 for Rotor-Gene Q) is maintained.
This enables the laboratory to check for run-to-run variability
and drift over time, and gives confidence when comparing
results over long periods. It is useful to look at the mean Ct
values for the previous 4–6 month period and compare to
previous periods to look for evidence of drift.
The Ct for equivalent data points for different transcripts
should be the same if reaction efficiencies are similar (this
justifies the use of the DDCt method, or use of a single
standard curve, i.e. using the ABL1 SC to quantify both
BCR-ABL1 and ABL1 transcripts). If this is not the case, it
suggestsat least oneof
optimized.
thereactionsneedstobe
Acknowledgements
DG gratefully acknowledges the support of the Minimal
Residual Disease Workpackage (WP12) of the European
LeukaemiaNet. We are most grateful to the members of the
NCRI CML Working Party (Prof R Clark, Prof J Apperley, Prof
T Holyoake and Prof S O’Brien) for their useful comments to
the manuscript.
Appendix 1. Sample processing and red cell lysis
Reagents
Red cell lysis buffer (RCLB);
1· Phosphate buffer saline (PBS);
RLT buffer from RNeasy Mini Kit (contains GTC);
b-mercaptoethanol;
Centrifuge (capable of spinning volumes of up to 40 mL);
Syringes (2Æ0 mL) (from BD plasticware);
18G Blunt end mixing needles (Terumo; cat no BN-1838).
x10 Red cell lysis buffer (RCLB)
This can be purchased or prepared in house according to the
protocol below. Make sure that the instruction and procedure
are followed accurately. This step is vital for accurate and
reliable RNA extraction and errors in this step will result in
poor RNA yield.
Guideline
6
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology

Page 7

ProductSupplier Cat
no.
Weight/
volume
Working concentration
Ammonium
Chloride
EDTA
0Æ5 mol/L
(pH 8)
Potassium
bicarbonate
Merck 100173D248Æ7 g 1Æ55 mol/L
Gibco 15575-038 6Æ0 mL 0Æ01 mol/L
SigmaP-9144 30Æ03 g 0Æ1 mol/L
Prepare to scale depending on volume requirement
Measure out the dry chemicals and add 2Æ0 L of sterile RNase-free (e.g.
pharmacy) water and then add EDTA. Once dissolved, make up to
3Æ0 L with pharmacy water and transfer to a 2Æ5 L Winchester vessel
and store in cold room. The remaining 500 mL can be transferred to
another clearly labelled 500 mL vessel and stored at 4?C.
x1 Red cell lysis buffer
Make 1 L of x10 RCLB up to 10 L with sterile pharmacy water and
leave overnight at 4?C, or in a cold room. Adjust the pH to 7Æ4 with
concentrated HCl and return to the cold room overnight. The
following morning check the pH and adjust accordingly. The pH meter
must be standardized prior to use.
Prevention of cross contamination
There should be physical separation of the working areas for primary
sample handling, pre-PCR (cDNA synthesis and PCR set up) and post-
PCR, qualitative and quantitative. Care should be exercized when
dispensing plasmid standards. All reagents used in RNA isolation,
cDNA synthesis and PCR should be aliquoted in smallest possible
working volumes. Aerosol-resistant pipette tips must be used for all
procedures. It is recommended that all tube transfers should be
checked by a colleague.
Protocol
Sample preparation and RLT cell lysis.
should ideally be performed within a Class II cabinet. Use aseptic
techniques throughout and make sure the tubes are firmly capped.
Reagents should be stored in small aliquots.
Note: all possible steps
1 Transfer whole sample (BM, PB and bone marrow harvests) to
50 mL suitably labelled polypropylene (Falcon) tubes. The tube
should carry at least two identifiers for the patient (Surname and
laboratory number)
Note. For samples >10 mL it is recommended that the sample be
centrifuged at 400 g on a bench centrifuge using a swing out rotor for
10 min and the plasma discarded using a sterile pasteur pipette. The
buffy coat is collected and transferred to a 50 mL Falcon tube, using a
sterile pasteur pipette. This is achieved by gently disturbing the top of
the packed cells with the sterile pasteur pipette. The buffy coat is then
collected by skimming across the surface in a series of to and fro
motions. Usually the buffy coat collected is approximately 1 mL in
volume.
1 Add x1 ice-cold RCLB up to 40 mL
2 Vortex sample vigorously and incubate on ice for 10 min, making
sure that the tube is firmly capped before proceeding with the
vortexing to avoid any spillage.
3 Centrifuge sample at 400 g for 7 min, making sure the tube is capped
firmly. Discard supernatant at the end of the run.
4 Add a second 40 mL x1 ice-cold RCLB, vortex vigorously and place
on ice for 10 min.
5 Centrifuge at 400 g for 7 min. Discard the supernatant as above.
6 After each centrifugation and before adding the new aliquot of RCLB
or later PBS, make sure the pellet is dislodged from the bottom of the
tube to maximize efficiency of each step.
(a) NB: Two washes with 1· RCLB are usually sufficient to wash away
the haemoglobin/red cells from the cellular pellet. However, some
samples may require further x1 RCLB washes and the process
should then be repeated until the nuclear pellet is devoid of pink-
red coloration.
7 Wash the cellular pellet once with 30 mL x1 PBS, vortex vigorously
and centrifuge at 400 g for 7 min.
8 Discard the supernatant and drain the excess by inverting the tube on
to paper tissue.
9 In a Falcon tube measure out the RLT buffer (contains GTC),
allowing at least 1 mL buffer per sample plus 1 mL for a blank
control then add 10 lL b-mercapatoethanol for each mL measured
out. GTC is highly toxic and extra precautions must be applied when
handling it. Work should ideally be carried out in a Class II cabinet.
Wear laboratory coats and gloves. Wash hands on completion.
10 Resuspend the nuclear pellet in 1 mL of RLT using a 2 mL syringe
and 18G blunt needle. Homogenize the suspension by passing it
rapidly to and fro through the needle until it loses viscosity, thus
degrading high molecular weight DNA. Take care when handling
needles and do not re-sheath. Discard needles unsheathed on
completion of homogenization. In cases where the nuclear pellet is
larger/smaller than normally expected, it may be necessary to adjust
the RLT/b-mercaptoethanol volume according to the size of the
cellular pellet. This will help to minimize the occurrence of overly
viscous (or alternatively too dilute) sample preparations. A sample
suspension of correct viscosity can be aspirated using 1 mL pipette
(Gilson or other similar manufacturer) and plugged tip with ease.
11 Using sterile pasteur pipettes transfer the RLT (GTC) lysates into
screw cap Eppendorf tubes correctly labelled with patient details.
12 Store lysates at )20?C for short-term storage, and at )80?C for long
term storage.
13 Whilecompletingextractionremembertoprepareblanktubesforthe
control of contamination derived from this step, which includes:
(a) 70% Ethanol control from the RNA extraction protocol.
(b) Sterile water control (25 lL), an aliquot of sterile water set-aside
from the RNA extraction protocol.
(c) GTC control saved from the sample processing. (see below)
Guideline
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology
7

Page 8

Sample preparation and TRIzol cell lysis
Protocol
1 Pour off supernatant after final red cell lysis step being careful not to
lose the pellet. Flick tube to resuspend white cell pellet and add
?14 mL PBS, spin at 580 g for 5 min.
2 Pour off supernatant, resuspend pellet in 1 mL TRIzol. Using a sterile
pasteur pipette, pipette up and down vigorously to homogenize. If
sample is very viscous, extra TRIzol may be added in 1 mL aliquots.
In our hands homogenization with needles as for RLT decreases
the DNA contamination as does freezing the lysates prior to
RNA extraction. Exceeding the recommended cell:TRIzol ratio
(5–106:1 mL of TRIzol) results in considerable DNA contamination.
3 Transfer 1 mL of homogenate to each labelled 1Æ5 mL Eppendorf
tube. Incubate for 3–5 min at room temperature (15–30?C) to
permit the complete dissociation of nucleoprotein complexes.
4 Freeze samples at )80?C or continue as described in Appendix 2.
Appendix 2. RNA extractions
Precautions for handling RNA
Gloves must be worn at all times when handling reagents and
consumables used for RNA extraction and manipulation. Ribonuclease
is present on skin and is highly stable and present on all laboratory
surfaces and is resistant to autoclaving.
RNA extraction using Qiagen kit
Reagents:
Qiagen RNeasy Mini Kit (Catalogue No. 74106);
Qiagen Rotor adaptors (cat no. 990394);
Qiagen elution tubes (included with above);
Qiagen 2 mL sample tubes RB (cat no. 990381);
Qiacube machine with RNeasy mini;
QIAcube filter tips 1000 lL (cat no. 990352);
Micro-centrifuge Sterile plugged tips;
1Æ5 mL micro-centrifuge tubes;
P1000 and P200 Pipettes;
70% Ethanol;
Protocol adjusted to elution volume of 60 lL.
Protocol
Manual RNA extraction (Qiagen columns).
work within the laminar flow cabinet throughout the procedure to
protect both yourself and the samples. RNase is highly stable and is
detectable on the fingertips.
Wear gloves and
1 Thaw RLT (GTC) lysates (stored at )20?C overnight or at )80?C for
longer period) for RNA extraction.
2 Include negative controls, i.e. water, ethanol and RLT/GTC controls.
3 Pipette 350 lL of 70% ethanol into appropriately labelled 1Æ5 mL
micro-centrifuge tube using 1000 lL plugged tip.
4 On thawing, vigorously vortex GTC one at a time.
5 Transfer 350 lL to appropriately labelled 1Æ5 mL micro-centrifuge
tube using 1000 lL plugged tip, changing the tip between each
sample.
6 Return the remaining GTC lysates to )20?C and then )80?C for
longer storage.
7 Transfer each 700 lL suspension to an appropriately labelled
RNeasy spin column.
8 In addition to the samples, include a blank ethanol control, i.e. add
700 lL of 70% ethanol to a separate spin column.
9 Centrifuge the columns for 15 s at 7800 g in a microfuge. This is
achieved by setting the timer on the micro-centrifuge for 1 min
and the speed at 7,826 g. When 15 s have lapsed stop the cen-
trifuge For g to rpm conversion see http://cabinet.weblog.com.pt/
arquivo/TR0040d44-Centrifuge-speed.pdf
Note: DO NOT hold down the PULSE SPIN button as this will spin at
15,339 g.
10 Discard the flow-through from the collection tubes.
Note:
Qiagen RNeasy kit), aliquot enough buffer from the stock bottle into a
universal tube to perform a single experiment. This will minimize access
to the stock solutions and so minimize contamination risk.
For both the RW1 Buffer and RPE Wash Buffer (supplied in
11 Transfer650 lL RW1Buffertothespincolumns,centrifuge for15 s
at 7826 g and discard the flow-through and collector tubes.
12 Place column on new collector tube.
13 Transfer 500 lL RPE Wash Buffer to the spin columns, changing
the tip between each sample, centrifuge for 15 s at 7826 g and
discard the flow-through.
14 Add 500 lL RPE Wash Buffer to the spin columns and centrifuge
for 2 min at 15,339 g.
15 Discard the flow-through and the collection tubes.
16 Place the columns upside down onto tissue paper in the hood,
keeping the lid CLOSED, for 2 min. This will allow any residual
70% ethanol at the bottom of the column to evaporate.
17 Transfer the columns to new labelled, 1Æ5 mL micro-centrifuge
tubes. Open the lids of the spin columns slightly, and leave for
approximately 20 min. This will allow any remaining 70% ethanol
present within the spin columns to evaporate.
Note: Take an aliquot of RNase-free water from the stock solution
(provided with the Qiagen RNeasy kit), enough for an entire single
experiment. Once again, this is done to minimize the risk of
contamination. Also set aside 60 lL of water into a 1Æ5 mL micro-
centrifuge tube, which will be used later as a control for
subsequent cDNA synthesis.
18 Apply 60 lL of RNase-free water to the centre of each column. Try
to apply it directly on to the membrane, taking care not to touch
the membrane itself or the sides of the column.
19 Elute the RNA from the membrane by centrifuging for 2 min at
15,339 rpm. As the spin column is sitting within a micro-centrifuge
Guideline
8
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology

Page 9

tube, bend the lid of the micro-centrifuge tube back against the
centrifuge rotor, so as to prevent it being snapped off whilst
spinning.
20 Remove and discard each spin column. Close the lid of the micro-
centrifuge tube.
21 It is recommended that the cDNA should be synthesized imme-
diately, as RNA is very labile. If this is not possible, store the RNA
at )80?C.
Automated RNA extraction (Qiacube).
automated RNA extraction using the Qiacube machine is as follows:
(Operators should read the Qiacube user manual version 1 before
using the machine).
The procedure for
Wear gloves – transfer samples and prepare plastic ware within the
laminar flow cabinet.
1 Transfer 350 lL of RLT (GTC) lysate to an appropriately labelled
2 mL RB sample tube.
2 Also aliquot a blank ethanol control (350 lL of 70% ethanol) and
have the tube transfers checked by a colleague
3 Prepare the required number of rotor adaptors by inserting one spin
column and 1 appropriately labelled elution tube into each adaptor
(see Qiacube user manual page 5–11 for details – leave the middle
position empty).
4 Open the Qiacube door; place the rotor adaptors in the centrifuge
and the 2 mL sample tubes into the Qiacube shaker rack (see Qia-
cube user manual page 5–13 for details). Check the labelled sample
tube position in the shaker rack corresponds with the labelled elu-
tion tube/rotor adaptor position in the centrifuge (see Qiacube user
manual appendix B-1 for loading instructions).
5 Remove reagent bottle lids and fill the bottles to the marked level
with appropriate reagent from RNeasy mini kit; return bottles to
their correct positions in the reagent rack (see Qiacube user manual
page 5–9).
6 Ensure that there are >16 Qiacube filter-tips (1000 lL – NOT wide
bore) in the Tip rack.
7 Close the Qiacube door.
8 Select the RNeasy mini protocol from the on screen menu. Follow
on screen instructions. The programme takes approximately 30 min.
9 Once finished, open the Qiacube door and carefully bring the rotor
adaptors to the hood.
10 Separate the elution tubes and close their lids. Pour off GTC waste
to a container and discard used plastic-ware to a clinical waste bin.
11 Aswiththemanualmethod,proceedtocDNAsynthesisimmediately
or store RNA at )80?C. (See Appendix 3: cDNA synthesis).
TRIzol RNA extraction.
026; 100 mL) is a mono-phasic solution of phenol and guanidine
isothiocyanate. Addition of chloroform followed by centrifugation
separates the solution into aqueous and organic phases. RNA remains
exclusively in the aqueous phase, which is then recovered by
precipitation with isopropyl alcohol/isopropanol.
TRIzol (GibcoBRL catalogue no.15596-
At the end of Stage 1 the samples can be frozen at )80?C. Stage 2 of the
process involves the precipitation and resuspension of the RNA into
RNase free water. Every step of this procedure is undertaken on ice
using RNA-free consumables. We recommend a maximum batch size
for Stage 2 of 12 samples. This is to allow all samples to keep cool
during the process.
Equipment
Bench-top centrifuge;
Refrigerated microfuge; For g to rpm conversion please see http://
cabinet.weblog.com.pt/arquivo/TR0040dh4-Centrifuge-speed.pdf
Rotamixer;
Pipettes – p1000, p200, p20 plus filter tips;
Cool block;
Individual wrapped sterile pastettes;
15 mL Falcon tubes, graduated, RNAse free;
1Æ5 mL Eppendorf tubes.
Reagents
We recommend a separate stock of chemicals for use in RNA buffers;
Injection grade sterile water, 5 mL injection vials or DEPC water
(Thistle Scientific);
Isopropanol/Propan-2-ol; VWR AnalaR NORMAPUR (UN1219)
20842.312;
Chloroform; AnalaR BDH (UN1888) 100776B;
75% Ethanol: Dilute 37Æ5 mL of 100% Ethanol with 12Æ5 mL injection
grade water.
Protocol
1 Turn on the RNA microfuge and fast cool to 4?C.
2 Remove samples from freezer and place in cool block once defrosted.
3 Label 1 · 1Æ5 mL Eppendorf tube for each sample.
4 Once the samples have defrosted, add 200 lL chloroform and vortex
for a minimum of 15 s. Make sure samples have been mixed
homogenously. Beware: samples will layer again if left for any time
before centrifuging.
5 Spin at 15 800 g for 15 min at 4?C.
6 Remove the tubes from the centrifuge and place in the cool block. A
self-tubetransfercheckisadvisedatthispointtoavoidsamplemixup.
7 Transfer the upper aqueous layer into the Eppendorf tube. Be careful
not to disturb the interphase layer (white layer – DNA/protein). If
the phases are not well separated, add a further 50 lL chloroform,
vortex and re-spin. Discard the bottom pink layer according to the
individual laboratory procedures.
8 Add an equal volume of cold isopropanol (approximately 400–
600 lL). Mix up and down slowly approximately 10 times.
9 Leave at )20?C for at least 1 h.
10 Spin at 15 800g for 15 min at 4?C, discard isopropanol supernatant
appropriately.
11 Wash in 1 mL of cold 75% ethanol.
12 Spin at 15 800 g for 5 min at 4?C.
Guideline
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology
9

Page 10

13 Discard ethanol supernatant using p200 and p10 tips. Pulse spin
the sample or use a vacuum centrifuge, if required, to remove the
last of the supernatant. Note that over drying may results in a pellet
that is difficult to resuspend.
14 Resuspend the pellet in 20 or 40 lL injection grade sterile RNase-
free water depending on the size of pellet obtained. Allow the pellet
to re-dissolve for 10 min at room temperature then either flick the
tube several times or vortex briefly and pulse spin.
15 Quantify RNA (optional; may be unreliable due to DNA
contamination). The optimum RNA concentration is between 500
and 1000 ng/mL. If >1500 ng/mL add a further 20 lL and leave for
an additional 10 min at room temperature. Re-OD sample.
Appendix 3. cDNA synthesis
The protocol below applies to samples eluted from Qiagen columns
into 60 lL of double distilled H2O (ddH2O) or RNA resuspended in
water following TRIzol RNA extraction. Some laboratories use the
entire eluate from 1/3 of the original RLT or GTC lysed samples, others
use a specified amount of RNA (after OD reading). We would
recommend that at least 2 lg of total RNA is used for each cDNA
synthesis. Irrespective of the protocol used, the aim is to be able to
measure a minimum of 10 000 ABL1 molecules for each 2Æ5–3 lL of
cDNA in each RT-QPCR reaction.
Reagents
5· Reverse Transcriptase Buffer (usually supplied) with MMLV-RT
(Invitrogen):0Æ25 mol/L Tris-HCl,
KCl, 15 mmol/L MgCl2. This may vary depending on the supplier
used.
pH8Æ3,0Æ375 mmol/L
0Æ1 mol/L dithiothreitol (DTT) usually supplied with MMLV-RT.
25 mmol/L dNTP stock: Mix an equal volume of ultrapure 100 mmol/
L dATP, dCTP, dGTP and dTTP.
Hexamer primers/random primers: 5 mg/mL random (Invitrogen
48190-011) although this may vary with individual suppliers.
Reverse transcriptase (usually supplied at 200 u/lL and RNasin
(Promega 40 u/lL).
cDNA master mix preparation
Prepare a master mix, the volume of which may depend on the
total number of samples to be processed (scale up or down to
requirement). An example of the final concentration for each re-
agent is provided below. However, we suggest that each laboratory
may follow the protocol recommended by the manufacturer sup-
plying the Reverse Transcriptase purchased by the individual users
and we refer to the EAC guidelines for additional information
(Gabert et al, 2003).
It is also relevant to note that a RT-negative sample (to control for
DNA amplification) is not routinely recommended. This test, however,
should be carried out while validating any RT-qPCR to exclude
amplification from DNA contaminating the RNA preparation when
new primers are tested. This is not a problem for BCR-ABL1 and ABL1
amplification and therefore this control is not required routinely.
A master mix should be prepared to achieve the following recom-
mended working concentrations in each reaction:
Working concentration
5· buffer*
0Æ1 mol/L DTT
dNTP 25 mmol/L
Random primers
RNasin
Reverse transcriptase
ddH2O
x reaction
10 mmol/L
0Æ5 mmol/L
500 ng
50 units
400 units
Volume will vary depending on final
volume reaction and volume of RNA used
Large master mixes can be prepared in advance and stored at )80?C.
However, in these cases the RNasin and Reverse Transcriptase is
omitted and only added to the mixture before use.
*As supplied from the RT-enzyme; this composition may vary
depending on supplier.
Protocol
Wear gloves at all times whilst working within the laminar flow cab-
inet, throughout the procedure. In addition to the samples, cDNA
synthesis should also include the following blank controls (as
mentioned above):
1 70% ethanol control from the RNA extraction protocol. Comments
as before
2 Sterile water control (25 lL), an aliquot of sterile water set-aside
from the RNA extraction protocol.
3 GTC control saved from the sample processing. Comments as before
The RNA is resuspended in ddH2O at the bottom of the tube used to
collect the eluate during the Qiagen protocol or it will be in a tube at
the end of the TRIzol procedure.
Label cDNA tubes with numbered stickers, remembering that each
tube must carry two ID items (patient’s name and unique laboratory
number identifier). Label tops and sides of tubes. Pre-print labels can
be purchased from specific suppliers.
1 Incubate the RNA at 65?C for 10 min.
2 Transfer to ice immediately, to avoid formation of RNA secondary
structure.
3 Transfer tubes to a micro-centrifuge and pulse spin to collect the
contents to the bottom of the tube. While the RNA is incubating at
65?C, preparecDNAmastermixas aboveandhaveicereadyin atray.
4 Thaw cDNA mix or prepare cDNA master mix sufficient for number
of samples (including the blank controls) at room temperature as
per protocol above.
5 Add RNasin to the cDNA mix, and vortex briefly.
6 Add MMLV-RT to the remaining cDNA mix, and vortex briefly.
Pulse spin to collect mix at bottom of tube.
7 Transfer cDNA mix to the required final volume (in lL) to each of
the other samples/controls. Mix by gently pipetting up and down a
couple of times into the RNA at the bottom of the tube.
Guideline
10
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology

Page 11

8 Incubate at 37?C for 2 h (as better efficiency of synthesis is achieved
with longer incubations). Check the heating block is at the required
temperature.
9 Incubate at 65?C for 10 min to inactivate the reverse transcriptase.
10 Pulse spin samples in a micro-centrifuge.
11 Store cDNA samples at )80?C.
Use 3 lL per reaction to test for ABL1 control amplification prior to
BCR-ABL1 testing. Only test for BCR-ABL1 sample with >10 000 ABL1
molecules. The final volume of the RT-qPCR will depend on the
platform used but it should be noted that, in general, the analysis of
smaller samples will compromise the sensitivity with which residual
disease can be detected.
Note. To reduce variability between BCR-ABL1 and ABL1 tests, the
team at the Hammersmith Hospital has now developed a ‘duplex RQ
PCR test where all samples are tested simultaneously for both targets in
the same well (Gerrard et al, 2011, in press).
Appendix 4. RT-qPCR set up
Reagents
Taqman X2 universal PCR master mix (Applied Biosystems) part no.
4304437;
cDNA: recommended volume 3 lL (can be up to 5 lL, depending on
local policy). Some laboratories use up to 5 lL, equivalent to 100 ng of
total RNA used in the cDNA synthesis.
Protocol
For both the target gene and reference gene, prepare a mix containing
(per well):
12Æ5 lL of Taqman X2 universal PCR master mix;
300 nmol/L primers;
200 nmol/L probe (some laboratories use 100 nmol/L);
Water to final volume of 22 lL (if using 3 lL cDNA – otherwise adjust
accordingly).
Calculate the amount of reagents required for all samples plus controls
(accounting for analysis in triplicate/duplicate). Prepare the master
mix under sterile conditions and aliquot 20 or 22 lL into each well
(depending on the cDNA volume to be used) of a 96-well plate or
tubes depending on equipment used. Then add cDNA, this should also
be carried out under sterile conditions. It is recommended that the
same volume of cDNA is used for BCR-ABL1 and ABL1 tests.
Addition of plasmid reference into the relevant control wells must be
carried out in a separate room and hood to avoid contamination.
Pulse-spin plates (or tubes) so that contents gather at the bottom of
wells. Place the reactions into the real time machine. Procedure may
change depending on the equipment.
Plate or tubes must be spun shortly (1 min 400 g for a plate in a
suitable bench centrifuge with adequate adaptors) on in a bench
microfuge (tubes). Place the reactions into the real time machine.
Procedure may change depending on the equipment.
Cycling conditions: an example of PCR program for ABI real-time
machines:
95?C: 10 min followed by 50 cycles of 95?C for 15 s, 60?C: 1 min.
Some laboratories use up to 45 cycles.
References
Baccarani, M., Cortes, J., Pane, F., Niederwieser,
D., Saglio, G., Apperley, J., Cervantes, F.,
Deininger, M.,Gratwohl,
Hochhaus,A.,Horowitz,
Kantarjian, H., Larson, R., Radich, J., Simons-
son, B., Silver, R.T., Goldman, J. & Hehlmann,
R.; European LeukemiaNet (2009). Chronic
myeloid leukemia: an update of concepts and
managementrecommendations
LeukemiaNet. Journal of Clinical Oncology, 27,
6041–6051.
Beillard, E., Pallisgaard, N., van der Velden, V.H.,
Bi, W., Dee, R., van der Schoot, E., Delabesse, E.,
Macintyre, E., Gottardi, E., Saglio, G., Watzinger,
F., Lion, T., van Dongen, J.J., Hokland, P. &
Gabert, J. (2003) Evaluation of candidate control
genes for diagnosis and residual disease detection
in leukemic patients using ‘real-time’ quantitative
reverse-transcriptase polymerase chain reaction
(RQ-PCR) – a Europe against cancer program.
Leukemia, 17, 2474–2486.
Branford, S. & Hughes, T. (2006) Diagnosis and
monitoring of chronic myeloid leukemia by
qualitative and quantitative RT-PCR. Methods in
Molecular Medicine, 125, 69–92.
Branford, S., Cross, N.C., Hochhaus, A., Radich,
J., Saglio, G., Kaeda, J., Goldman, J. & Hughes,
T. (2006) Rationale for the recommendations
A.,
M.,
Guilhot,
Hughes,
F.,
T.,
ofEuropean
for
detecting BCR-ABL transcripts in patients with
chronic myeloidleukaemia.
1925–1930.
Branford, S., Fletcher, L., Cross, N.C., Muller, M.C.,
Hochhaus, A., Kim, D.W., Radich, J.P., Saglio,
G., Pane, F., Kamel-Reid, S., Wang, Y.L., Press,
R.D., Lynch, K., Rudzki, Z., Goldman, J.M. &
Hughes, T. (2008) Desirable performance char-
acteristics for BCR-ABL measurement on an
international reporting scale to allow consistent
interpretation of individual patient response and
comparison of response rates between clinical
trials. Blood, 112, 3330–3338.
Burmeister, T. & Reinhardt, R. (2008) A multiplex
PCR for improved detection of typical and
atypical BCR-ABL fusion transcripts. Leukemia
Research, 32, 579–585.
Bustin, S.A. (2010) Developments in real-time
PCR research and molecular diagnostics. Ex-
pert Review of Molecular Diagnostics, 10, 713–
715.
Cross, N.C. (2009) Standardisation of molecular
monitoring for chronic myeloid leukaemia. Best
Practice and Research. Clinical Haematology, 22,
355–365.
Cross, N.C., Feng, L., Chase, A., Bungey, J., Hughes,
T.P. & Goldman, J.M. (1993) Competitive poly-
merase chain reaction to estimate the number of
BCR-ABL transcripts in chronic myeloid leuke-
harmonizing currentmethodologyfor
Leukemia, 20,
mia patients after bone marrow transplantation.
Blood, 82, 1929–1936.
Cross, N.C., Melo, J.V., Feng, L. & Goldman, J.M.
(1994) An optimized multiplex polymerase chain
reaction (PCR) for detection of BCR-ABL fusion
mRNAs in haematological disorders. Leukemia,
8, 186–189.
Dhut, S., Gibbons, B., Chaplin, T. & Young, B.D.
(1991) Establishment of a lymphoblastoid cell
line, SD-1, expressing the p190 bcr-abl chimaeric
protein. Leukemia, 5, 49–55.
Druker, B.J., Tamura, S., Buchdunger, E., Ohno, S.,
Segal, G.M., Fanning, S., Zimmermann, J. &
Lydon, N.B. (1996) Effects of a selective inhibitor
of the Abl tyrosine kinase on the growth of
Bcr-Abl positive cells. Nature Medicine, 2, 561–
566.
Druker, B.J., Talpaz, M., Resta, D.J., Peng, B.,
Buchdunger, E.,Ford,
Kantarjian, H., Capdeville, R., Ohno-Jones, S. &
Sawyers, C.L. (2001) Efficacy and safety of a
specific inhibitor of the BCR-ABL tyrosine kinase
in chronic myeloid leukemia. New England
Journal of Medicine, 344, 1031–1037.
Foroni, L., Gerrard, G., Nna, E., Khorashad, J.S.,
Stevens, D., Swale, B., Milojkovic, D., Reid, A.,
Goldman, J. & Marin, D. (2009) Technical aspects
and clinical applications of measuring BCR-ABL1
transcripts number in chronic myeloid leukemia.
American Journal of Hematology, 84, 517–522.
J.M., Lydon,N.B.,
Guideline
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology
11

Page 12

Gabert, J., Beillard, E., van der Velden, V.H., Bi, W.,
Grimwade, D., Pallisgaard, N., Barbany, G.,
Cazzaniga, G., Cayuela, J.M., Cave ´, H., Pane, F.,
Aerts, J.L., De Micheli, D., Thirion, X., Pradel, V.,
Gonza ´lez, M., Viehmann, S., Malec, M., Saglio,
G. & van Dongen, J.J. (2003) Standardization and
quality control studies of ‘real-time’ quantitative
reverse transcriptase polymerase chain reaction of
fusion gene transcripts for residual disease
detection in leukemia – a Europe Against Cancer
program. Leukemia, 17, 2318–2357.
Gerrard, G., Mudge, K., Stewart, R., Foskett, P.,
Stevens, D., Khorashad, J.S., Szydlo, R., Rezvani,
K., Marin, D., Reid, A., Apperley, J., Goldman, J.
Foroni, L. (2011) Duplex quantitative PCR for
molecular monitoring of BCR-ABL1 associated
haematologicl malignancies. American Journal of
Hematology, 86, 313–315
Hochhaus, A., Reiter, A., Skladny, H., Melo, J.V.,
Sick, C., Berger, U., Guo, J.Q., Arlinghaus, R.B.,
Hehlmann, R., Goldman, J.M. & Cross, N.C.
(1996) A novel BCR-ABL fusion gene (e6a2) in a
patient with Philadelphia chromosome-negative
chronic myelogenous leukemia. Blood, 88, 2236–
2240.
Hughes, T., Deininger, M., Hochhaus, A., Branford,
S., Radich, J., Kaeda, J., Baccarani, M., Cortes, J.,
Cross, N.C., Druker, B.J., Gabert, J., Grimwade,
D., Hehlmann, R., Kamel-Reid, S., Lipton, J.H.,
Longtine, J., Martinelli, G., Saglio, G., Soverini,
S., Stock, W. & Goldman, J.M. (2006) Monitor-
ing CML patients responding to treatment with
tyrosine kinase inhibitors: review and recom-
mendations for harmonizing current methodol-
ogy for detecting BCR-ABL transcripts and kinase
domain mutations and for expressing results.
Blood, 108, 28–37.
Kantarjian, H., Shah, N.P., Hochhaus, A., Cortes, J.,
Shah, S., Ayala, M., Moiraghi, B., Shen, Z.,
Mayer, J., Pasquini, R., Nakamae, H., Huguet, F.,
Boque ´, C., Chuah, C., Bleickardt, E., Bradley-
Garelik, M.B., Zhu, C., Szatrowski, T., Shapiro,
D. & Baccarani, M. (2010) Dasatinib versus
imatinibin newlydiagnosed
chronic myeloid leukemia. New England Journal
of Medicine, 362, 2260–2270.
Kawasaki, E.S., Clark, S.S., Coyne, M.Y., Smith,
S.D., Champlin, R., Witte, O.N. & McCormick,
F.P. (1988) Diagnosis of chronic myeloid and
acute lymphocytic leukemias by detection of
leukemia-specific mRNA sequences amplified in
vitro. Proceedingsof the National Academia of
Science, U S A, 85, 5698–5702.
Lozzio, B.B. & Lozzio, C.B. (1977) Properties of the
K562 cell line derived from a patient with chronic
myeloid leukemia. International
Cancer, 19, 136.
Melo, J.V. (1996) The diversity of BCR-ABL fusion
proteins and their relationship to leukemia
phenotype. Blood, 88, 2375–2384.
Melo, J.V. & Chuah, C. (2007) Resistance to
imatinib mesylate in chronic myeloid leukaemia.
Cancer Letters, 249, 121–132.
Nowell, P.C. & Hungerford, D.A. (1960) Chromo-
some studies on normal and leukemic human
leukocytes. Journal of National Cancer Institute,
25, 85–109.
O’Hare, T., Walters, D.K., Stoffregen, E.P., Jia, T.,
Manley, P.W., Mestan, J., Cowan-Jacob, S.W.,
Lee, F.Y., Heinrich, M.C., Deininger, M.W. &
Druker, B.J. (2005) In vitro activity of Bcr-Abl
inhibitors AMN107 and BMS-354825 against
clinically relevant imatinib-resistant Abl kinase
domain mutants. Cancer Research, 65, 4500–4505.
Rowley, J.D. (1973) Identificaton of a transloca-
tion with quinacrine fluorescence in a patient
with acute leukemia. Annals of Genetics, 16,
109–112.
Saglio, G., Kim, D.W., Issaragrisil, S., le Coutre, P.,
Etienne, G., Lobo, C., Pasquini, R., Clark, R.E.,
chronic-phase
Journalof
Hochhaus, A., Hughes, T.P., Gallagher, N.,
Hoenekopp, A., Dong, M., Haque, A., Larson,
R.A. & Kantarjian, H.M.; ENESTnd Investigators
(2010) Nilotinib versus imatinib for newly diag-
nosed chronic myeloid leukemia. New England
Journal of Medicine, 362, 2251–2259.
Shtivelman, E., Lifshitz, B., Gale, R.P. & Canaani, E.
(1985) Fused transcript of abl and bcr genes in
chronic myelogenous leukaemia. Nature, 315,
550–554.
van der Velden, V.H., Boeckx, N., Gonzalez, M.,
Malec, M., Barbany, G., Lion, T., Gottardi, E.,
Pallisgaard, N., Beillard,
Hoogeveen, P.G., Gabert, J. & van Dongen, J.J.;
Europe Against Cancer Program (2004) Differ-
ential stability of control gene and fusion gene
transcripts over time may hamper accurate
quantification of minimal residual disease – a
study within the Europe Against Cancer Pro-
gram. Leukemia, 18, 884–886.
Westbrook, C.A., Hooberman, A.L., Spino, C.,
Dodge, R.K., Larson, R.A., Davey, F., Wurster-
Hill, D.H., Sobol, R.E., Schiffer, C. & Bloomfield,
C.D. (1992) Clinical significance of the BCR-ABL
fusion gene in adult acute lymphoblastic leuke-
mia: a Cancer and Leukemia Group B Study
(8762). Blood, 80, 2983–2990.
White, H.E., Matejtschuk, P., Rigsby, P., Gabert, J.,
Lin, F., Wang, Y.L., Branford, S., Muller, M.C.,
Beaufils, N., Beillard, E., Colomer, D., Dvorak-
ova, D., Ehrencrona, H., Goh, H.G., El Housni,
H., Jones, D., Kairisto, V., Kamel-Reid, S., Kim,
D.W., Langabeer, S., Ma, E.S., Press, R.D.,
Romeo, G., Wang, L., Zoi, K., Hughes, T., Saglio,
G., Hochhaus, A., Goldman, J.M., Metcalfe, P. &
Cross, N.C. (2010) Establishment of the 1st
World Health Organization International Genetic
Reference Panel for quantitation of BCR-ABL
mRNA. Blood, 116, 111–117.
E.,Hop,W.C.,
Guideline
12
ª 2011 Blackwell Publishing Ltd, British Journal of Haematology