49
Članek prispel / Received
4. 2. 2020
Članek sprejet / Accepted
9. 4. 2021
Abstract
Purpose: The objective of this study 
was to introduce the GeneXpert Ultra 
cartridge-based BCR-ABL diagnostic test 
into routine molecular monitoring of mi-
nimal residual disease for chronic myelo-
id leukemia (CML) patients who are on 
tyrosine kinase inhibitors (TKIs). 
Methods: BCR-ABL1 transcript 
quantification was performed on 100 
peripheral blood samples from 33 CML 
patients at different stages of the disease 
using automated GeneXpert BCR-ABL 
Ultra assay and following a standardized 
laboratory protocol (SP). 
Results: Analysis results were expressed 
as % on the International Scale (IS) for 
Izvleček
Namen: Cilj študije je bil uvesti dia-
gnostični test BCR-ABL, ki temelji na 
GeneXpert Ultra sistemu, za rutinsko 
molekularno spremljanje minimalne 
preostale bolezni (MRD) pri bolnikih s 
kronično mieloidno levkemijo (CML), 
ki so na zaviralcih tirozin kinaze (TKI).
Metode: Kvantifikacija transkriptov 
BCR-ABL1 je bila izvedena v 100 vzor-
cih periferne krvi 33 bolnikov s CML 
v različnih fazah bolezni z avtomati-
ziranim GeneXpert BCR-ABL Ultra 
testom (Cepheid, ZDA) in s standardizi-
ranim laboratorijskim protokolom (SP).
Rezultati: Rezultati obeh analiz so 
bili izraženi kot % na mednarodni le-
Ključne besede: 
kronična mieloična levkemija, 
fuzijski gen BCR-ABL1, GeneXpert, 
minimalna preostala bolezen, 
molekularni odziv.
Key words: 
chronic myeloid leukemia, BCR-
ABL1 fusion gene, GeneXpert, 
minimal residual disease, molecular 
response 
Natančnost in hitrost poročanja o 
molekularnem odzivu pri KML bolnikih z in 
vitro diagnostičnim testom Xpert BCR-ABL
Accuracy and speed of molecular response 
reporting with Xpert BCR-ABL Ultra in vitro 
diagnostic test  in CML patients
Avtor / Author Špela Stangler Herodež1,2,  Nadja Kokalj-Vokač1,2,  Zlatko Roškar3,  
Mojca Dreisinger3
Ustanova / Institute 1Univerzitetni klinični center Maribor, Klinika za ginekologijo in perinatologijo, Laboratorij 
za medicinsko genetiko, Maribor, Slovenija; 2Univerza v Mariboru, Medicinska fakulteta, 
Katedra za molekularno biologijo, Maribor, Slovenija; 3Univerzitetni klinični center Maribor, 
Klinika za interno medicino, Oddelek za hematologijo in hematološko onkologijo;
1University Medical Centre Maribor, Clinic for Gynecology and Perinatology, Laboratory 
of Medical Genetics, Maribor, Slovenia; 2University of Maribor, Faculty of Medicine, 
Department for Molecular Biology, Maribor, Slovenia; 3University Medical Centre Maribor, 
Clinic for Internal Medicine, Department for Haematology and Haematological Oncology, 
Maribor, Slovenia.
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
50
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
INTRODUCTION
Chronic myeloid leukemia (CML) is a clonal 
malignant disease of pluripotent stem cells directed 
towards granulopoiesis. According to the World 
Health Organization (WHO), it is classified as chronic 
myeloproliferative disease with an annual incidence 
of 1.6/100,000, where one out of every 62,500 men 
and women will be diagnosed with CML during their 
lifetime (1). The average age at diagnosis is 50–60 years, 
although the disease can occur during all stages of life, 
even in childhood. Exposure to ionizing radiation is the 
only known risk factor for the occurrence of the disease. 
CML accounts for 20–25% of all leukemia cases. It is 
Naslov za dopisovanje /  
Correspondence
Doc. dr. Špela Stangler Herodež, univ. 
dipl. inž. kem. teh., Univerzitetni 
klinični center Maribor, Klinika za 
ginekologijo in perinatologijo, Laboratorij 
za medicinsko genetiko, Ljubljanska 5, 
2000 Maribor, Slovenija
Telefon  +386 23212737
Fax        +386 23212755
E-pošta: 
Spela.STANGLERHERODEZ@ukc-mb.si
stvici (IS) in so bili ocenjeni z moleku-
larnim odzivom (MR) v skladu z labora-
torijskimi priporočili za oceno globokih 
molekulskih odzivov po zdravljenju 
kronične mieloične levkemije. Njihova 
primerjava je pokazala, da ni bilo stati-
stično pomembne razlike v rezultatih, iz-
raženih kot % na IS, p = 0,098. Vzorce 
glede na ocenjeno MR smo razdelili v 6 
skupin (MR1, MR2, MR3, MR4,  MR4.5 
in MR5 ). 74 od 100 vzorcev je imelo 
z obema metodama popolnoma enak 
MR. 12 vzorcev je imelo višji MR, pri-
dobljen z GeneXpertom, medtem ko je 
imelo 14 vzorcev nižji MR v primerjavi s 
SP. Pri 3 od 74 vzorcev se je GeneXpert 
razlikoval od SP za dva MR razreda, 
medtem ko se je pri ostalih  razlikoval 
za en MR razred. Analiza glavnega mo-
lekularnega odziva (MMR) je pokazala, 
da smo dosegli 82 % ujemanje. V osmih 
vzorcih so bile z GeneXpertom zaznane 
minimalne količine kopij BCR-ABL1, 
medtem ko jih SP ni zaznal. Ključno 
je, da pri nobenem izmed bolnikov re-
zultat, pridobljen z GeneXpertom, ni 
vplival na drugačno klinično odločitev o 
zdravljenju s TKI kot rezultat, ki je bil 
pridobljen po SP. 
Zaklju~ek: Glede na ugotovitve lah-
ko Xpert BCR-ABL Ultra test šteje-
mo kot koristno in zelo hitro klinično 
orodje za molekularno spremljanje 
bolnikov na TKI.
BCR-ABL1 real-time PCR quantifica-
tion and estimated using molecular re-
sponse (MR) according to the laboratory 
recommendations for scoring deep MRs 
following CML treatment. A compari-
son of the two methods showed that there 
was no statistically significant difference 
in the results expressed as % on the IS 
(p = 0.098). Samples were divided into 
six groups (MR1, MR2, MR3, MR4, 
MR4.5, and MR5) according to the esti-
mated MR. A total of 74/100 samples 
had a completely identical MR with 
both methods, 12/100 samples had a 
higher MR obtained with GeneXpert, 
and 14/100 samples had a lower MR 
compared to the SP. In addition, 2/74 
samples had a different MR for two clas-
ses obtained with GeneXpert compared 
to the SP, while the others 72/74 sam-
ples differed in one MR class. Analysis 
of the major molecular response resulted 
in 82% agreement. Minimal amounts 
of BCR-ABL1 copies were detected 
with GeneExpert in eight samples while 
the SP did not detect them. None of the 
GeneXpert results had any effect on the 
clinical decision for treatment with TKIs 
obtained using the SP.
Conclusion: The Xpert BCR-ABL Ul-
tra assay can be considered a useful and 
very fast clinical tool for molecular follow-
-up of patients on TKIs.
51
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
the first malignant disease in which a change leading 
to its development was recognized at a molecular level 
(2). It is caused by the formation of the Philadelphia 
chromosome, a reciprocal translocation of the 
chromosomal parts between chromosomes 9 (9q34.1) 
and 22 (22q11.2), resulting in the formation of the 
BCR-ABL1 fusion gene. The resulting fusion oncogene 
carries the BCR-ABL1 (tyrosine kinase) protein that 
is responsible for the development of CML because 
it maintains the activity of the signaling pathway in 
the cell, thereby accelerating the proliferation of stem 
cells, interfering with maturation, and preventing 
apoptosis. Genetically modified cells overwhelm the 
bone marrow. Genetic instability of the cell clone may 
cause resistance to the treatment and progression of the 
disease to secondary acute leukemia. Therefore, BCR-
ABL1 can be employed as a specific leukemic marker 
for minimal residual disease (MRD) follow-up (3). After 
treatment with tyrosine kinase inhibitors (TKIs), BCR-
ABL1 quantification may predict the outcome in CML 
patients, especially in those who achieve a complete 
cytogenetic response (4). 
Monitoring of BCR-ABL transcript levels with real-
time quantitative polymerase chain reaction (qRT-
PCR) in patients on TKI therapy is the standard of 
care in the management of CML. Furthermore, it 
is very important to achieve sensitivity of 4.5 logs 
below baseline in patients who are considered for TKI 
discontinuation (5). In order to assess the treatment 
response, it is essential to ensure that variability in 
testing methodologies is tightly controlled. This 
was the cause of BCR-ABL molecular measurement 
standardization across laboratories, which introduced 
the International Scale (IS) defined by WHO (5). 
The GeneXpert system utilizes microfluidics to directly 
process peripheral blood samples in a cartridge for RNA 
extraction, reverse transcriptase PCR (RT-PCR), and 
fluorescence detection in an “all-in-one” self-contained 
instrument (6). The GeneXpert BCR-ABL system 
uses ABL1 as its control gene, which is recommended 
for molecular response (MR) follow-up. It calculates 
the BCR-ABL1/ABL1 ratio using the delta-crossing 
threshold (Ct) comparative-quantitative method (7). 
Jobbagy et al. and Winn-Deen et al. performed an early 
comparison of the GeneXpert assay with standard qRT-
PCR assay and found that the system had the potential 
to translate into clinical practice despite the differences 
in quantitative methods (8, 9). However, the system 
did not report BCR-ABL1 values on the IS and result 
interpretation and inter-laboratory comparison were 
hindered. Enjeti et al. developed a conversion factor 
specific to GeneXpert assay and used it to adjust the 
BCR-ABL1/ABL1 ratio to the IS. A batch-specific 
standard curve and an efficiency value were also 
introduced to ensure cartridge lot-specific calibration 
(10). This very important quality control assay 
improvement enables inter-laboratory comparison. It 
also achieved an accurate measurement of 0.01% of 
BCR-ABL1 on the IS, a level sufficient to detect MR4 
(molecular response at 4-log reduction) (11, 12). 
In our study, we introduced the GeneXpert Ultra 
cartridge-based BCR-ABL diagnostic test into routine 
molecular monitoring of MRD for CML patients who 
are on TKIs. 
MATERIALS AND METHODS
Study sample
Peripheral blood samples were obtained from CML 
patients on TKI therapy referred to our laboratory of 
Medical Genetics from the Department of Haematology 
and Haematological Oncology, Clinic for Internal 
Medicine, University Medical Centre Maribor, Maribor, 
Slovenia for monitoring of BCR-ABL transcript levels. 
All samples were from individuals of Caucasian descent 
who were residents of different geographical areas of 
east Slovenia. The study was approved by the ethical 
committee (No. UKC-MB-KME-24-08/17). Informed 
consent was obtained from all participating individuals. 
Assay design 
An automated GeneXpert BCR–ABL Ultra car-
tridge System (Cepheid Europe S.A.S., Vira-Solelh, 
Maurens-Scopont, France) consists of an instrument 
and a personal computer with software for running 
tests and viewing results. It simplifies generation of 
very accurate and reproducible results using integra-
tion of nucleic acid extraction, target amplification, 
and quantification of the BCR-ABL e13a2/b2a2 and 
e14a2/b3a2 transcripts relative to ABL in peripher-
al blood samples. Disposable GeneXpert cartridges 
52
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
contain the RT-PCR and nested PCR reagents and 
host the RT-PCR and nested PCR processes. Cross-
contamination concerns are minimized because the 
cartridges are self-contained.
Assay procedure 
Peripheral venous blood samples were collected in 
standard vacutainer tubes containing ethylenedi-
amine tetraacetic acid (EDTA) anticoagulant. A total 
of 4 mL of whole peripheral blood was mixed with 
100 µL of proteinase K (Cepheid Europe S.A.S.). Af-
ter 1 min of incubation at room temperature, 2.5 mL 
of lysis reagent (Cepheid Europe S.A.S.) was added to 
inactivate nucleases in order to release nucleic acids 
from the cells, followed by vortexing and two incuba-
tion periods of 5 min at room temperature. After the 
second incubation, 1 mL of lysate was transferred to a 
new tube and mixed with 1.5 mL of lysis reagent. Af-
ter 10 min of incubation at room temperature, 2 mL 
of pure ethanol (Merck KgaA, Darmstadt, Germany) 
was added to the sample and 4.5 mL of mixture was 
transferred to the test cartridge. Washing reagent (Ce-
pheid Europe S.A.S.) was then added to the designat-
ed port in the test cartridge, the lid was closed, and the 
cartridge was loaded into the GeneXpert System. The 
GeneXpert System isolates the total RNA and runs a 
nested RT-PCR. Total assay time was approximately 2 
h and 20 min. 
Assay quality control 
Each GeneXpert BCR-ABL Ultra kit lot is standard-
ized using in-house secondary standards calibrated ac-
cording to the WHO (Figure 1). 
13
Figure 1. Sample of the BCR-ABL 
ΔCt standard curve. EΔCt=1.96 
(lot-specific efficiency value) is derived 
from the slope of the standard curve; 
IS-SF = 1.26 (lot-specific scaling fac-
tor) corrects the quantitative output 
of the assay to the international scale 
established based on WHO Interna-
tional Genetic Reference Panel.
Figure 1: Sample of the BCR-ABL ΔCt standard curve. EΔCt=1.96 (lot-specific efficiency value) is 
derived from the slope of the standard curve; IS-SF = 1.26 (lot-specific scaling factor) corrects the 
quantitative output of the assay to the international scale established based on WHO International 
Genetic Reference Panel. 
14
Figure 2. Calibration to WHO IS.
A lot-specific conversation factor is applied to each lot 
(Figure 2).
Statistical Analysis 
MRs were compared using the t-test, where p ≤ 0.05 
was considered statistically significant.
RESULTS 
BCR-ABL1 transcript quantification was performed on 
100 peripheral blood samples from 33 CML patients at 
different stages of the disease in two different laboratories. 
The first method used the automated GeneXpert BCR-
ABL Ultra assay (Cepheid Europe S.A.S.), while the 
second used a standardized laboratory protocol (SP). 
The blood samples were collected over three years in the 
period from 2017 to 2020. 
The negative control group included 14 blood samples 
obtained from patients with hematological disorders 
53
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
DISCUSSION 
A major advantage of the GeneXpert platform is 
its simplified work flow, which takes approximately 
2 h and 20 min. The system thus provides a faster 
turnaround, reduces hands-on time, and requires less 
technical expertise compared to the standard BCR-
ABL1 qRT-PCR assay. The GeneXpert automates 
both the pre-analytical and analytical phases. The risk 
of contamination with the equipment or with other 
samples is insignificant because the assay is performed 
inside a closed cartridge. All of these factors support 
the choice of GeneXpert for BCR-ABL1 quantification 
and CML monitoring. Cepheid has also introduced 
the GeneXpert BCR-ABL Ultra, a high-sensitivity 
assay, which demonstrates a limit of detection reaching 
MR4.5 and below (13). BCR-ABL assays that are 
sensitive in their measurement of deep-level response 
may aid in the identification of potential candidates 
for treatment discontinuation. GeneXpert BCR-ABL 
Ultra detects the most common BCR-ABL transcripts 
below MR4.5 or 0.0032%, which is widely accepted as 
the clinical threshold that defines candidates who can 
safely discontinue TKI therapy (13).
This important GeneXpert BCR-ABL1 assay 
improvement has led our laboratory to introduce the 
system into routine molecular monitoring of MRD for 
CML patients who are on TKIs, because the ability to 
consistently detect low levels of BCR-ABL transcripts 
in CML patients is important for the assessment of 
treatment outcomes in patients on TKI therapy. For 
that reason, GeneXpert BCR-ABL Ultra results in the 
present study were compared to the SP results. The 
concordance analysis showed that 74% of samples 
agreed at the same logarithmic interval and had a similar 
distribution of GeneXpert disagreement in values 
above and below those obtained using the SP. Analysis 
by intervals did not show any significant difference in 
concordance (p = 0.098). Analysis of MMR showed an 
82% agreement. MR was demonstrated to be properly 
determined using GeneXpert when compared to the 
SP. Our results also highlight the fact that GeneXpert 
was able to detect a low number of copies in cases 
that turned out to be negative when determined using 
the SP. This indicated a higher sensitivity as a result 
of the fact that GeneXpert provides a more sensitive 
Table 1. Concordance between standardized laboratory 
protocol (SP) and GeneXpert Ultra assay results at 
different logarithmic intervals and overall.
MR (%IS) 
Number of samples at the 
interval Concordance 
between 
methods
All samplesSP GeneXpert Ultra
MR1 (>1) 7 5 5/7 (71 %)
MR2 (≤1)   5 (50.0) 34 (18.7) 6/6 (100 %)
MR3 (≤0.1) 0 (0.0) 22 (12.1) 13/16 (81 %)
MR4 (≤0.01) 0 (0.0) 13 (7.1) 5/7 (71 %)
MR4.5 
(≤0.0032) 10 (100.0) 182 (100.0) 4/6 (67 %)
MR5 (≤0.001) 0 (0.0) 13 (7.1) 41/50(82 %)
Not detected 10 (100.0) 182 (100.0) 0/8 (0 %)
Total 0 (0.0) 13 (7.1) 74/100 (74 %)
unrelated to BCR-ABL gene fusion. The negative 
predictive value was 100%. A total of 74 of 100 (74%) 
samples had a completely identical MR using both 
methods (Table 1). Furthermore, 12 of 100 samples had 
a higher MR obtained with GeneXpert, while 14 of 100 
samples had a lower MR compared to the SP. Analysis 
by intervals did not show any significant difference in 
concordance (p = 0.098). 
Furthermore, only two of 74 (2.7%) samples had a 
different MR for two classes obtained with GeneXpert 
compared to the SP. The others 72 samples differed in 
one MR class. Analysis of the major molecular response 
(MMR) resulted in 82% agreement (Table 1). Minimal 
amounts of BCR-ABL1 copies were detected in eight 
samples that were not detected by the SP (Table 1). None 
of the GeneXpert results had any effect on the clinical 
decision for treatment with TKIs obtained using the SP.
54
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
assay (nested PCR) that is able to detect reduction of 
up to 5 log in BCR-ABL1 transcript concentrations. 
The negative control group also had a high specificity 
(100%) using GeneXpert. 
It can therefore be concluded that the GeneXpert 
BCR-ABL Ultra assay is a reproducible and robust 
alternative to laboratory-developed assays. Its ease of use 
may allow some laboratories to offer BCR-ABL testing 
to the patients they serve. It is also important that the 
short assay time enables same-day results for treating 
clinicians. Such fast sample turnaround can reduce 
patient anxiety and allow a faster treatment adjustment 
if there is an increase in BCR-ABL concentration.
ACKNOWLEDGEMENTS 
This work was supported by the IRP-2017/01-05 
internal research project of University Medical Centre 
Maribor, Maribor, Slovenia. The authors are grateful 
to all patients and individuals in the study who made 
this work possible and would especially like to thank 
the members of the Department for Haematology and 
Haematological Oncology, Clinic for Internal Medicine, 
University Medical Centre Maribor, Maribor, Slovenia 
and the members of the Clinical Department for 
Hematology, Clinic for Internal Medicine, University 
Medical Centre Ljubljana, Ljubljana, Slovenia for their 
helpful contributions to this study.
REFERENCES
1. Siegel RL, Miller KD, Jemal A. Cancer statistic, 
2019. CA Cancer J Clin  2019; 69: 7-34.
2. Höglund M, Sandin F, Simonsson B. Epidemiol-
ogy of chronic myeloid leukaemia: an update. Ann 
Hematol 2015; 94(Suppl 2): S241-7.
3. Branford S, Haughes TP. Practical considerations 
for monitoring patients with chronic myeloid leu-
kemia. Semin Hematol 2010; 47: 327-34. 
4. Hughes TP, Hochhaus A, Branford S, Müller MC, 
Kaeda JS, Foroni L. Long-term prognostic signifi-
cance of early molecular response to imatinib in 
newly diagnosed chronic myeloid leukemia: an 
analysis from the International Randomized Study 
of Interferon and ST1571 (IRIS). Blood 2010; 
116: 3758-65.
5. Day GJ, Lockwood C, Payton JE, Uy GL, Schiffer 
C, Feldman G et al. Development of Xpert® 
BCR-ABL Ultra, an Automated and Standard-
ized Multiplex Assay with Required Performance 
Characteristics for BCR-ABL1 Quantitative Mea-
surement on an International Reporting Scale. 
Blood 2015; 126: 2793. 
6. Dufresne S, Belloni D, Levy N. Tsongalis G. Quan-
titative assessment of the BCR-ABL transcript us-
ing the Cepheid Xpert BCR-ABL monitor assay. 
Arch Pathol Lab Med 2007; 131: 947-50.
7. Beillard E, Pallisgaard N, Van der Velden VH, Bi 
W, Dee R. 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 2007; 17: 2474-86.
8. Jobbagy Z, Van Atta R, Murphy K, Eshleman J, 
Gocke C. Evalution of the cepheid Gene Xpert 
BCR-ABL assay. J Mol Diag 2007; 9: 220-7.
9. Winn-Deen E, Helton B, Van Atta R, Wong W, 
Peralta J, Wang J et al. Development of an inte-
grated assay for detection of BCR-ABL RNA. Clin 
Chem 2007; 53: 1593-600.
10. Enjeti A, Granter N, Ashraf A, Fletcher L, Bran-
ford S, Rowlings P et al. A longitudinal evaluation 
of performance of automated BCR-ABL1 quan-
titation using cartridge-based detection system. 
Pathology 2015; 47: 570-4. 
11. Cayuela J, Macintyre E, Darlington M, Abdelali R, 
Fund X, Villarese P et al. Cartridge-based auto-
mated BCR-ABL1 mRNA quantification: solving 
the issues of standardization, at what cost? Hae-
matological 2011; 96: 664-71.
12. O'Dwyer M, Swords R, Nagler A, McMullin A, Le 
Coutre P, Langabeer S et al. Nilotinib 300 mg 
55
Laboratorijska študija / Laboratory study
ACTA MEDICO-BIOTECHNICA
2021; 14 (1): 25–37
bid as front-line treatment of CML: prospective 
analysis of the Xpert BCR-ABL monitor system 
and significance of 3-month molecular response. 
Leuk Res 2014; 38: 310-5.
13. Ferrand C, Deschamps M, Levitas A, Lockwood 
C, Payton JE, Uy GL et al. Xpert® BCR-ABL Ultra, 
a High Sensitivity Assay, Demonstrates a Limit 
of Detection Reaching MR4.5 and below. Blood 
2017; 130: 1610.