Development of a Multiplex PCR to Detect the BCR-ABL1 Transcripts in Chronic Myeloid Leukaemia Patients in Indonesia

Abstract

Background

There are limited commercial kits available for the identification of BCR-ABL1 gene fusion transcripts in chronic myeloid leukaemia (CML) patients. Most kits are based on RT-PCR to detect e14a2 and e13a2 transcripts as major breakpoint segments, but do not detect minor transcripts. Therefore, this study aimed to develop a multiplex PCR approach for BCR-ABL gene transcripts.

Methods

Primers were designed to amplify four BCR-ABL1 transcripts from the cDNA reverse transcribed from 52 archived mRNAs from CML patients. The BCR-ABL1 transcripts were identified by multiplex PCR, and the assay performance was compared to Xpert^® BCR-ABL Ultra (GeneXpert).

Results

The multiplex TD-PCR successfully amplified four BCR-ABL1 transcripts, with detection limits of 10 fg/µL for e13a2 and e14a2 and 100 fg/µL for e1a2 and e19a2. Out of 52 archived RNA samples, 40 were confirmed positive. The most common transcript was e14a2, which was identified in 72.5% of the confirmed positive samples, followed by e13a2, identified in 25% of the positive samples. The e19a2 transcript was identified in 2.5% of the positive samples. No patients expressed the e1a2 transcript type or co-expressed both e14a2 and e13a2. The overall sensitivity and specificity of the multiplex PCR when compared to GeneXpert results were 97.56% (95% CI: 87.14–99.94%) and 100% (95% CI: 71.51–100%), respectively.

Conclusion

The multiplex PCR assay developed to identify the four BCR-ABL1 transcripts can be implemented in CML diagnostic settings.

Introduction

Chronic myeloid leukaemia (CML) is a malignant clonal disorder of pluripotent hematopoietic stem cells caused primarily by the Philadelphia chromosome t(9,22) (q34;q11), identified by the presence of the BCR-ABL1 fusion gene.^1,2 The annual incidence of CML is 1–2 cases per 100,000 people.^3–5 The BCR-ABL1 fusion gene is generated by the breakpoint of the BCR and ABL1 genes, producing a protein that constantly activates tyrosine kinase, resulting in uncontrolled granulocyte proliferation. The ABL1 gene (chromosome 9) breakpoint may occur anywhere within a > 300-kb section at the 5’ end of the gene, either upstream of the first alternative intron Ib, between exons Ib and Ia, or downstream of exon Ia. The breakpoint in the BCR gene (chromosome 22) can occur in a variety of locations, including inside a 5.8-kb area known as the main breakpoint cluster region (M-bcr), covering 5 exons formerly called b1 to b5, now known to be exons 12 to 16 of the BCR gene.^6,7

Patients with CML are classified into three groups based on the BCR-ABL1 transcript type: major BCR (M-bcr), minor BCR (m-bcr), and micro BCR (µ-bcr). Over 95% of CML patients have the M-bcr type, which is further subdivided into two breakpoints: e13a2 and e14a2, formed by fusion of BCR exon 13 (b2) or exon 14 (b3) to ABL1 exon 2 (a2). There is a difference of 75 base pairs or 25 amino acids between the e13a2 and e14a2 products.⁸ The m-bcr transcript is e1a2, formed by the fusion of BCR exon 1 to ABL1 exon 2 (a2). The µ-bcr transcript is e19a2, formed by the fusion of BCR exon 19 to ABL1 exon 2 (a2). The abnormal gene fusion p210 protein is synthesised by the major BCR-ABL1 transcript (210 kDa), the chimeric p190 protein (190 kDa) is synthesised by the minor BCR-ABL1 transcript e1a2, while the p230 abnormal fusion protein is synthesised by the micro BCR-ABL1 transcript e19a2 (230 kDa).^3,9–11

The diagnosis of CML involves clinical and laboratory tests, including physical examination (with a focus on spleen and liver size), complete blood cell count with differential blood counting, blood smear examination, bone marrow aspirate for cytogenetics, fluorescent in-situ hybridisation (FISH) in cases of Ph-negativity, and qualitative reverse transcriptase polymerase chain reaction (RT-PCR) to detect BCR-ABL1 transcripts.¹² RT-PCR is one of the most commonly used procedures due to its ease of use, speed, and sensitivity,^1,13 and is mandatory to detect the BCR-ABL1 transcript type according to European LeukemiaNet (ELN) and Indonesian guidelines in Indonesia.^12,14 Accurate identification of the BCR-ABL1 transcript type at diagnosis is clinically important, as different transcripts may be associated with distinct clinical features and can influence molecular monitoring strategies and interpretation of treatment response.^14,15 In high-income countries, commercial qualitative RT-PCR kits that can identify the main BCR-ABL1 transcript types are widely available and routinely used. In Indonesia, however, these kits are either not registered, not regularly stocked, or are prohibitively expensive for many laboratories. Hence, in the Indonesian setting, the lack of a robust, locally validated, affordable assay that can simultaneously detect the major BCR-ABL1 transcript variants represents a practical barrier to implementing current ELN and national guidelines. Therefore, this study aimed to develop a multiplex touchdown PCR method for the diagnosis and characterisation of the four most relevant BCR-ABL1 transcript types (e13a2, e14a2, e1a2, and e19a2) in CML patients. By amplifying multiple target sequences with multiple primer sets in a single reaction, the proposed multiplex PCR method is intended to (i) reduce turnaround time, (ii) decrease reagent and consumable costs, (iii) minimise hands-on time and contamination risk, and (iv) provide comprehensive transcript typing in a single assay that can be implemented in routine molecular laboratories in Indonesia that have only standard PCR equipment. This locally developed method is expected to fill the current diagnostic gap and support guideline-concordant care for CML patients in resource-limited settings.

Methods

This study was approved by the university ethics committee Universitas Padjadjaran (No 1396/UN6.KEP/EC/2022), and the samples used to validate the assay were archived RNA samples from CML patients. The patients whose tissues were used for this research had provided informed consent, in accordance with the Declaration of Helsinki.

Multiplex Touchdown PCR Development

Primer3 was used to design primers for the multiplex PCR, targeting the BCR gene as the internal control, as well as the e13a2, e14a2, e1a2, and e19a2 transcript types. Table S1 shows the primer sequences generated by Integrated DNA Technologies (IDT), and Figure S1 illustrates the primer attachment sites. The gBlocks™ gene fragment containing different transcript types was used as the cDNA template to develop and optimise the multiplex PCR for the different BCR-ABL1 fusion transcripts. This fragment was diluted to a concentration of 1 pg/uL and used to generate a positive control.

The multiplex PCR was optimised starting with the singleplex PCR for each transcript, followed by the duplex, triplex, and quadruplex PCR using gBlock™ gene fragments. Singleplex PCR was performed using 0.2 µM of forward and reverse primers, with the concentration of the reverse primer (a2R) in the triplex PCR increased to 0.24 µM. Multiplex PCR was performed in a 200 µL thin-walled tube using an iCycler ^® thermal cycler (Bio-Rad) containing 1x Mastermix (MyTaq™ Mix, Bioline), 0.2 µM two different forward primers (e13F and e19F) located in the BCR gene, 0.32 µM of the e1F primer and 0.32 µMof the reverse primer located in exon 3 of the ABL1 gene, 5 µL of a positive control and RNAse-free water in a 25 µL final reaction volume. The thermal cycling consisted of 1 minute of denaturation at 95°C, then 35 cycles of amplification consisting of 15 seconds at 95°C, 30 seconds at a temperature ramping down from 62°C to 53°C (decreased by 0.5°C every 2 cycles), and 15 seconds at 72°C, and an extension step of 7 minutes at 72°C. The same protocol was used for all optimisation processes.

The PCR products (5 µL) were separated on a 2% TBE agarose gel containing RedSafe™ nucleic acid staining solution (Intron) and visualised with a GelDoc UV illuminator (Bio-Rad). The predicted bands for e13a2, e14a2, e1a2, and e19a2 were 381 bp, 456 bp, 693 bp, and 585 bp, respectively. The PCR products were sequenced to validate their accuracy and identity, and the results were analysed using BLAST.

Limit of Detection of the Multiplex Touchdown PCR

The analytical sensitivity of the developed multiplex assay was assessed by a serial dilution (10-fold dilution) of the gBlocks™ gene fragment. The limit of detection (LoD) was determined as the lowest concentration at which the positive control was amplified, ranging from 1 pg/µL to 1 fg/µL. The internal control, consisting of forward and reverse primers located in exon 2 and exon 11 of the BCR gene, respectively, was included to ensure the accuracy and specificity of the reverse transcription and amplification steps, at a concentration of 0.23 µM.

Multiplex Touchdown PCR Performance in Detecting CML

The performance of the developed Multiplex Touchdown PCR assay was evaluated using 52 archived RNA samples (collected from January 2022 to August 2024 and stored at −80°C from CML patients) and compared to Xpert^® BCR-ABL Ultra (GeneXpert) results. Only high-quality RNA samples (sample volume ≤20 µL and 260/280 ratio <1.8) were reverse transcribed using the SensiFAST™ cDNA Synthesis Kit (Bioline) according to the manufacturer’s instructions.

Statistical Analysis

The diagnostic performance of the multiplex BCR-ABL1 PCR assay was evaluated using the GeneXpert assay as the reference standard. Qualitative concordance of two methods was evaluated using a 2×2 contingency table. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall diagnostic accuracy were calculated according to standard diagnostic test evaluation methods.

Results

Multiplex Touchdown PCR Development

The singleplex PCR results in Figure 1 show that the e13a2 transcript produced a band of 381 bp, the e14a2 transcript produced a band of 456 bp, while the e1a2 and e19a2 transcripts produced bands of 693 bp and 585 bp, respectively, as predicted by in silico primer design software. The results of Sanger sequencing of each amplicon are shown in Figure 2, confirming the correct sequence of the breakpoints of each amplicon. The duplex, triplex, and quadruplex PCRs results are shown in Figure 3, confirming that all optimised PCRs produced the expected band sizes for each breakpoint.

Figure 1.

The singleplex TD-PCR results: (a) e13a2 transcript produced a band of 381 bp, (b) e14a2 transcript produced a band of 456 bp, (c) e1a2 transcript produced a band of 693 bp, and (d) e19a2 transcript produced a band of 585 bp.

Abbreviation: ntc, non-template control.

Figure 2.

Electropherograms of different transcripts are analysed using Sanger sequencing; the red line indicates the location of the breakpoint of each BCR-ABL fusion transcript. e13a2, e14a2, e1a2 and e19a2 are fusions of exon 13, 14, 1 and 9 of the BCR gene, respectively, with exon 2 of the ABL gene.

Figure 3.

The results of the (a) duplex TD-PCR, (b) triplex TD-PCR, and (c) quadruplex TD-PCR.

Abbreviations: L, 100 bp ladder; ntc, non-template control.

The LoD was calculated as the lowest concentration at which the positive control was amplified using the developed multiplex PCR (Figure 4), indicating that e13a2 and e14a2 had a sensitivity of 10 fg/µL, whereas e1a2 and e19a2 had a sensitivity of 100 fg/µL.

Figure 4.

Limit of detection of the multiplex TD-PCR. Assessment of multiplex TD-PCR LoD obtained from positive control of serial dilution (1 pg-1 fg per µL) for the four transcript targets. The limit of detection of e13a2 and e14a2 is 10 fg/µL, and the limit of detection for e1a2 and e19a2 is 100 fg/µL.

Abbreviation: ntc, non-template control.

Characterisation of the BCR-ABL1 Transcripts in CML Patient Samples

The developed Multiplex Touchdown PCR was used to identify the BCR-ABL1 transcripts in 52 archived mRNA samples from CML patients. Demographic, clinical and laboratory characteristic of CML patients enrolled in this study was shown in Table 1. The multiplex PCR of BCR-ABL1 showed that the most common transcript was e14a2 (72.5%), followed by e13a2 (25%) and only 1 CML patient was identified with e19a2 (2.5%) (Table 2 and Figure 5). The performance of the developed Multiplex Touchdown PCR was compared to the GeneXpert results, the overall sensitivity and specificity of the Multiplex BCR-ABL1 were 97.56% (95% CI: 87.14–99.94%) and 100% (95% CI: 71.51–100%), respectively (Tables 3 and S2). Overall agreement between PCR multiplex BCR-ABL and GeneXpert results was 98.08%.

Table 1.

Characteristic of CML Patients (n = 52)

Characteristic	Value
Age, years
Median (range)	36 (2–74)
Mean ± SD	36.9 ± 19.28
Age group, n (%)
<40 years	28 (53.9)
40–59 years	19 (36.5)
≥60 years	5 (9.6)
Sex, n (%)
Male	33 (63.5)
Female	19 (36.5)
Disease phase, n (%)
Chronic phase (CP)	38 (73.1)
Accelerated phase (AP)	4 (7.7)
Blast phase (BP)	8 (15.4)
N.A	2 (3.8)
Current TKI therapy, n (%)
Imatinib	41 (79)
Nilotinib	9 (17)
Ponatinib	2 (4)
Laboratory parameters, mean ± SD (range)
Hemoglobin, g/dL	12 ± 1.96 (5.5–13.4)
White blood cell count (×103/uL)	16 ± 27.7 (17.9–607.4)
Platelet count (×109/L)	528 ± 820 (18–5658)
BCR-ABL1 IS (%), median (range)	98% (0.018–193%)

Abbreviation: N.A, The data is not available.

Table 2.

BCR-ABL1 Transcripts Identified in CML Patients Using Multiplex -PCR

BCR-ABL Transcript Type	Total Positive (n=40)
e13a2	10 (25%)
e14a2	29 (72.5%)
e1a2	0 (0%)
e19a2	1 (2.5%)

Figure 5.

RedSafe-stained gel electrophoresis image of multiplex PCR. Each well contains the BCR‑ABL fusion transcripts identified from patients (Patient ID indicated on each well) and the BCR transcript as internal control (IC). L, 100 bp ladder, e13a2 transcript (381 bp), e14a2 transcript (456 bp), internal control (858 bp), and (+): multiplex PCR for all BCR-ABL fusions controls.

Table 3.

Performance of the Multiplex BCR-ABL

Multipleks PCR BCR-ABL	Gene-Xpert (+)	Gene-Xpert (-)	Sensitivity % (95% CI)	Specificity % (95% CI)	PPV (95% CI)	NPV (95% CI)	Accuracy (95% CI)
BCR-ABL (+)	40	0	97.56 (87.14–99.94)	100 (71.51–100)	100 (91.9–100)	91.67 (61.52–99.79)	98.08 (89.74–99.95)
BCR-ABL (-)	1	11

Discussion

The common commercial kit in Indonesia for BCR-ABL is a quantitative RT-PCR kit, which only detects and quantifies the two most common transcripts, e13a2 and e14a2, and cannot determine the specific transcript. Herein, we developed a Multiplex RT-PCR using three BCR primers and one ABL primer to detect the various BCR-ABL transcripts, with the BCR transcripts serving as an internal control. Multiplex RT-PCR uses multiple primer sets in a single reaction to detect and amplify multiple target genes or transcripts¹⁶ and thus can be an important diagnostic tool to identify BCR-ABL transcripts in CML patients. The Multiplex Touchdown PCR was highly sensitive, capable of detecting the four transcripts even at a low concentration of 100 fg/µL.

The BCR-ABL1 transcript type is important for predicting treatment outcomes and the rate of treatment-free remission.^16–18 The e1a2 transcript is mainly found in lymphoid blast phase CML and acute lymphoblastic leukaemia, occurring at a rate of about 1.8% in chronic phase CML.¹⁹ Limited studies suggest that m-bcr may be associated with increased monocytes and a more aggressive disease. In cases of lymphoid or mixed lineage blast crisis, e1a2 is a high-risk condition with a poor response to tyrosine kinase inhibition.²⁰

The rare e19a2 fusion transcript is typically found in chronic myelogenous leukaemia with prominent neutrophilic maturation. Some researchers have also referred to these atypical malignancies as chronic neutrophilic leukaemia.²¹ The e19a2 fusion gene was first associated with neutrophilic CML, a mild Ph-myeloproliferative illness that advances relatively infrequently, but later reports detected e19a2 fusion gene transcripts in classical CML and acute myeloid leukaemia.^21,22 However, it has been shown that e19a2 type CML does not differ significantly from M-BCR in terms of clonal evolution and chromosomal abnormalities during blastic crisis.²³

About one-quarter of the patients in this study expressed the e13a2 transcript, with the most common transcript being e14a2. A rare BCR-ABL1 rearrangement including exon 19 on the BCR gene (e19a2) was discovered in one patient (2.5%). The lack of e13a2/e14a2 co-expression and e1a2 transcript type in any of the Indonesian patients in our study may be related to racial distinctions or the sensitivity of the detection method. These findings are similar to a study in Yogyakarta, Indonesia, that found a higher proportion of e14a2 transcripts compared to the e13a2 transcript.²⁴ Although a small number of studies reported the opposite results.²⁵ However, a review of 34 studies including 54,034 patients showed a higher overall estimated prevalence of e14a2 compared to e13a2 (54% vs 39%, respectively).²⁴ In the present study, male patients (55%) tended to express e14a2, in contrast to the reports of a relationship between sex and transcript type, with male patients having a higher tendency to express e13a2 and females having a higher tendency to express e14a2.^26,27 Given the small sample size using in this study, the distribution of BCR-ABL1 fusion in this cohort should interpret with caution, as this might be no reflect BCR-ABL1 fusion profile in much boarder CML patients in Indonesia. As this study was intended as a proof of concept for a new multiplex PCR assay and its preliminary clinical validation, rather than to deliver conclusive molecular epidemiological data. Consequently, it is essential to conduct larger, multicenter studies with a wider variety of patient populations to thoroughly define the molecular epidemiology of BCR-ABL1 fusion transcripts in Indonesia and to confirm the clinical efficacy of the assay further.

The performance of the developed Multiplex Touchdown PCR was compared to the GeneXpert results and demonstrated high overall concordance. Of the 52 archived RNA samples analysed only one yielded a discordant result, testing positive by GeneXpert but negative by multiplex TD-PCR. This sample was obtained from a patient undergoing treatment and exhibited a very low BCR-ABL1 transcript level (International Scale [IS] <0.3%). The discordance is most likely attributable to the superior analytical sensitivity of the Xpert assay for detecting low-abundance transcripts. These findings indicate that the developed multiplex TD-PCR is suitable for qualitative identification of BCR-ABL1 transcript types at the time of diagnosis but is not intended for the detection or monitoring of minimal residual disease (MRD), for which highly sensitive quantitative molecular assays are required.

Conclusion

The developed Multiplex Touchdown PCR is a reliable method for identifying the BCR-ABL1 transcript type and thus serves as a useful diagnostic tool for CML patients.

Disclosure

Delita Prihatni reports support for the manuscript from Support from Padjadjaran University (Faculty of Medicine) during the conduct of the study. The authors report no other conflicts of interest in this work.