OCT-1 Expression in Patients with Chronic Myeloid Leukemia: A Comparative Analysis with Respect to Response to Imatinib Treatment

Abstract

The introduction of tyrosine kinase inhibitors (TKI) has resulted in a significant improvement in the treatment of CML patients. However, some CML patients are resistant to imatinib therapy, the initial TKI therapy in the CML. Therefore, it is important to find prognostic markers for resistance. The OCT-1 gene involved in imatinib uptake is also suspected to cause imatinib resistance. The aim of this study was to investigate the role of OCT-1 in imatinib resistance by comparing OCT-1 expression levels in imatinib resistant and imatinib sensitive patients with chronic myeloid leukemia (CML). This study was conducted on 101 patients with CML [imatinib sensitive (n = 51) and imatinib resistant (n = 50)] who were treated with imatinib. Gene expression analysis was done using QRT-PCR. The relative expression levels of OCT-1 were calculated using 2(−ΔΔCT) method. OCT1 mRNA expression levels were 0.149 (0.011–2.532) and 0.119 (0.008–2.868) in imatinib-sensitive group and imatinib-resistant group, respectively. OCT-1 expression levels were not significantly different in the imatinib-sensitive group when compared to imatinib resistant group (p > 0.05). OCT-1 expression was also similar in BCR-ABL1 kinase domain mutation positive and negative cases (p > 0.05). The imatinib-resistant group had a higher rate of hydroxyurea or interferon-alpha treatment prior to imatinib therapy and a lower rate for first-line imatinib as the only treatment than the imatinib-sensitive group (p = 0.002 and p = 0.002, respectively). According to the results of our study, OCT-1 does not have a biomarker feature in the evaluation of imatinib response. In addition, the study should be performed in larger patient groups.

Introduction

Chronic myeloid leukemia (CML) is a clonal myeloproliferative disease characterized by the reciprocal translocation between chromosomes 9 and 22, producing the Philadelphia (Ph) chromosome [1]. This leads to formation of the BCR-ABL1 gene encoding the BCR-ABL1 oncoprotein with constitutively active tyrosine kinase activity, critical to the development of CML and the most effective target for the therapy used against this disease [1, 2]. Imatinib mesylate is the first-line therapy for CML that potently inhibits the tyrosine kinase activity of BCR-ABL1 via competitive binding to the ATP-binding domain of BCR-ABL1 [1–3]. Despite the clinical success of imatinib therapy with excellent cytogenetic and molecular responses for the majority of CML patients, approximately 20–30% of patients fail to achieve an optimal response (primary resistance), or lose their initial response (secondary resistance) [4–6]. The most commonly identified molecular resistance mechanisms are associated with alterations in BCR-ABL1, including mutations and overexpression, while mechanisms independent of BCR-ABL1 are also likely to involve factors associated with drug metabolism and drug transport [5–9].

The entrance of imatinib into the cells is a process dependent on membrane transporters including organic cation transporter 1 (OCT-1) and organic cation/carnitine transporter 2 (OCTN2). Accordingly, membrane transporters have also emerged as important contributors to poor treatment response as their low expression leads to a reduction in intracellular imatinib concentration [9, 10]. Hence, as the main influx transporter involved in imatinib uptake in CML cells OCT-1 has been considered a key determinant of intracellular levels of imatinib achieved in cells contributing to imatinib resistance in patients with CML [10–12]. However, there are conflicting data regarding the impact of OCT-1 on the response to imatinib therapy in terms of OCT-1 genetics, OCT-1 expression levels, and cellular imatinib uptake as well as the usefulness of OCT-1 expression or activity as a prognostic marker in CML patients [10, 13–18]. Identifying prognostic indicators of poor response to imatinib is considered of critical importance in the development of new therapeutic approaches to circumventing imatinib resistance or to the provision of tailored therapy for patients at risk [8, 19, 20]. The aim of this study was to investigate the role of OCT-1 on imatinib resistance by comparing OCT-1 expression levels in imatinib resistant and imatinib sensitive patients with CML treated with imatinib.

Materials and Methods

Study Population

A total of 101 CML patients treated with imatinib who were followed up in the Hematology outpatient clinic of Istanbul Medical Faculty between 2015 and 2017 were included on a voluntary basis in this retrospective study. Patients were divided into two groups according to treatment response including imatinib sensitive (n = 51) and imatinib resistant (n = 50) groups using European Leukemia Net (ELN) criterias [21] for molecular-monitoring data retrieved from hospital records. CML patients aged > 18 years, who were under imatinib treatment or had previously been treated with imatinib were included in the study.

Assessments

Data on patient demographics (age and gender), age at diagnosis, pre-imatinib treatments, tyrosine kinase regimens, presence of BCR-ABL1 kinase domain (KD) mutation and OCT-1 expression levels were compared in treatment response groups. OCT-1 expression was also evaluated with respect to treatment groups with and without mutations.

Treatment Response

Patients, who achieved a complete molecular response (CMR), or a major molecular response (MMR) (BCR-ABL1 (IS) < 1%, as assessed by QRT-PCR) were considered to be part of the imatinib-sensitive responder group, while the imatinib-resistant non-responder group was composed of those without CMR or MMR BCR-ABL1 (IS) ≥ 1%, as assessed by QRT-PCR).

RNA Extraction and cDNA Synthesis

Total RNA was isolated from whole blood using a High Pure RNA Isolation Kit (Roche Diagnostics, Mannheim, Germany). cDNA synthesis was done using a Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnostics). cDNA was then used for PCR amplification.

Quantitative Real-Time Polymerase Chain Reaction (QRT-PCR)

Gene expression analysis of the target gene OCT-1 was performed on the LightCycler 480II (Roche Diagnostics) instrument with a QRT-PCR using hypoxanthine phosphoribosyltransferase 1 (HPRT1) as the reference gene. Primer pairs used for the OCT-1 gene were: forward 5′-CTGAGCTGTACCCCACATTCG-3 and reverse 5′-CCAACACCGCAAACAAAATGA-3. The following amplification program was used: Ten min at 95 °C, followed by 40 cycles involving 15 s at 95 °C, 30 s at 60 °C, and 10 s at 72 °C. SYBR Green 1 dye binding method was used. Primer pairs used for the HPRT1 gene were forward 5′-GACCAGTCAACAGGGGACAT-3 and reverse 5′-TGTCAATTATATCTTCCACAATCAAG-3′. Standard curves were created for each amplification using ten-fold serial dilutions of a cDNA mixture expressing HPRT1. Relative gene expression levels were measured using the Livak formula based on the 2(−ΔΔCT) method [22].

Statistical Analysis

Statistical analysis was conducted using IBM SPSS Statistics for Windows, version 21.0 (SPSS Inc., Chicago, IL, USA). Chi-square test and Fisher’s exact test were used for the comparison of categorical data. Student’s t-test and Mann–Whitney U test were used for the analysis of parametric variables, where appropriate. p < 0.05 was considered statistically significant.

Results

Mean age in the overall study population was 55.0 ± 13.4 years, while age at disease diagnosis was 44.9 ± 13.4 years. Males composed 51.5% (52/101) of the study population. Imatinib sensitive and imatinib resistant groups were homogenous in terms of age, gender and age at diagnosis (p = 0.723, p = 0.843 and p = 0.427, respectively). Overall, 40.6% (41/101) of patients had received hydroxyurea or interferon-alpha treatment prior to imatinib therapy. First-line imatinib was the only treatment in 62.4% (63/101) of patients, being followed by second-line and third-line regimens in 34.7% (35/101) and 1.9% (2/101), respectively. The imatinib-resistant group as compared with the imatinib-sensitive group was associated with a higher rate of hydroxyurea or interferon-alpha treatment prior to imatinib therapy (28/41 (56.0%) vs. 13/41 (25.5%), p = 0.002) and a lower rate for receipt of first-line imatinib as the only treatment (30.0% vs. 96.0%, p = 0.002). KD mutation was not assessed in 61 (60.4%) patients, whereas the presence of mutation was noted in 7 (6.9%) of 40 patients. No significant difference was noted between imatinib-sensitive and imatinib-resistant groups in terms of KD mutation positivity (p = 0.565). Demographics, treatment characteristics and BCR-ABL1 kinase domain mutation in study groups are given in Table 1.

Table 1.

Demographics, treatment characteristics and BCR-ABL kinase domain mutation in study groups

	Total (n = 101)	Imatinib-sensitive (n = 51)	Imatinib-resistant (n = 50)	P value
Age, mean	55.0 ± 13.4	55.5 ± 13.6	54.6 ± 13.2	0.723a
Age at diagnosis, mean	44.9 ± 13.4	46.0 ± 13.8	43.8 ± 13.1	0.427a
Gender, n (%)
Female	49 (48.5%)	24 (47.1%)	25 (50.0%)	0.843b
Male	52 (51.5%)	27 (52.9%)	25 (50.0%)
Pre-imatinib treatment
Hydroxyurea/interferon-alpha	41 (40.6%)	13 (25.5%)	28 (56.0%)	0.002b
Tyrosine kinase inhibitors
First-line imatinib treatment	63 (62.4%)	48 (96.0%)	15 (30.0%)	0.002b
Second-line therapy	35 (34.7%)	2 (4.0%)	33 (66.0%)
Third-line therapy	2 (1.9%)	–	2 (4.0%)
BCR-ABL kinase domain mutation
Not assessed	61 (60.4%)	46 (90.2%)	15 (30.0%)
No mutation	33 (32.7%)	5 (9.8%)	28 (56.0%)	0.565c
Presence of mutation	7 (6.9%)	0	7 (14.0%)

Data are presented as mean ± SD

SD standart deviation, n number of samples

^aStudent’s t-test

^bχ2 test

^cFisher’s exact test

OCT1 mRNA expression levels were 0.149 (0.011–2.532) and 0.119 (0.008–2.868) in imatinib responders and nonresponders, respectively. There was no significant difference in OCT-1 expression levels with respect to treatment response (p = 0.940) (Fig. 1). OCT-1 expression levels were also similar between KD mutation positive and negative patients in the overall study population and in the imatinib resistant group (p = 0.653). Imatinib resistant patients without KD mutation also had similar OCT-1 expression levels with imatinib sensitive patients (p = 0.231). BCR-ABL1 kinase domain mutations and OCT-1 expression levels relative to treatment response are given in Table 2.

Fig. 1.

OCT-1 expression in imatinib sensitive and imatinib resistant groups

Table 2.

BCR-ABL kinase domain mutations and OCT-1 expression levels relative to treatment response

	Presence of BCR-ABL kinase domain mutation				Treatment response
	Total (n = 40)		Imatinib-resistant (n = 35)
OCT-1 expression median (min–max)	Absent	Present	Absent	Present	Imatinib-sensitive	Imatinib-resistant
	(n = 33)	(n = 7)	(n = 28)	(n = 7)	(n = 51)	(n = 50)
	0.149 (0.027–2.868)	0.097 (0.014–1.157)	0.201 (0.027–2.868)	0.097 (0.014–1.157)	0.149 (0.011–2.532)	0.119 (0.008–2.868)
P value	0.557		0.323a	0.433b	0.940

Mann–Whitney U test

Data are presented as median (min–max). n: number of samples

^aCompared to imatinib sensitive group

^bCompared to imatinib resistant group with no mutations

Discussion

Our findings revealed a higher likelihood of receiving treatments before and after the first-line imatinib therapy in the imatinib resistant group than in the imatinib sensitive group. There was no significant difference between the imatinib resistant and the imatinib sensitive groups in terms of OCT-1 gene expression levels and KD mutation rates.

While T315I mutation has been the most common KD mutation reported in past studies [23], F359V mutation (in 2 of 7 patients) was the most frequent KD mutation in our cohort and T315I mutation was not observed. In addition, although BCR-ABL1 is suggested to have an inhibitory effect on OCT-1 expression [24], our findings revealed no association between KD mutation and OCT-1 expression. Similarly, in a cell line model study based on ectopic expression of BCR-ABL1, authors reported no evidence that BCR-ABL1 directly affected OCT-1 expression and function [19].

In our cohort, imatinib sensitive and imatinib resistant patients had similar levels of OCT-1 expression, suggesting no relevance of the OCT-1 expression levels in response to therapy among CML patients.

In their studies among CML patients, some investigators reported that high OCT1 expression levels in responders were associated with improved patient response and lower expression/or activity of OCT-1 associated with poor response to treatment. [8, 12, 19, 25–30]. Accordingly, measuring expression/functional activity of OCT-1 at diagnosis has been considered a prognostic indicator for both short- and long-term clinical response to imatinib, based on studies reporting superior complete cytogenetic response rates, and progression-free and overall survival [10, 26, 27, 29, 31–33].

However, similar to our findings, other studies in CML patients have reported no significant difference in OCT-1 expression levels between responders and non-responders along with no significant correlation of OCT-1 expression with imatinib response and achievement of MMR [13, 17, 34–37], There was no significant difference in imatinib response among CML patients when patients were divided into ‘equal’, ‘high’ and ‘low’ OCT-1 expression groups [36]. Moreover, OCT-1 expression was considered not to be a good predictor of molecular response in CML patients in some studies [15, 38], in conjunction with the likelihood that the cellular uptake of imatinib is independent of OCT-1 in others, suggesting OCT-1 expression may not be a good biomarker for imatinib response [15].

Therefore, conflicting data exist in the current literature on the impact of OCT-1 expression in predicting treatment responses in CML patients [13, 17, 18]. One of the factors implicated in this notable lack of concordance among studies has been the consideration that the OCT-1 gene is highly polymorphic in ethnically diverse populations, leading to the formation of OCT-1 variants that differ in frequency as well as drug pharmacokinetics [9, 16]. The genetic variability in OCT-1 expression is therefore considered important in the relation of OCT-1 expression to treatment response, since some described polymorphisms can lead to normal expression, but poor protein activity [39]. Moreover, OCT-1 protein activity rather than OCT-1 expression was suggested to be a strong determinant of treatment response in CML patients in some studies [30, 32]. Another explanation for the several conflicting results involved in the measurement of the expression of the OCT-1 gene seems to be the majority of the cell population at diagnosis used for monitoring OCT-1 expression, since OCT-1 expression was reported to be significantly higher in peripheral blood polymorphonuclears compared to mononuclears [19, 24].

In addition, the involvement of both OCT-1 and OCTN2 transporters in the resistant phenotype has been suggested, since studies with OCT-1 inhibitors showed that the influx of the imatinib transporter is not exclusively mediated by OCT-1, suggesting the involvement of other drug transporters [9].

Recently, imatinib was reported to have a non-linear concentration-dependent effect on OCT-1 expression in an imatinib-sensitive cell line that occurs within 24 h and remains stable for up to 72 h, suggesting that the transport activity of OCT-1 is dependent not only on OCT-1 expression levels, but also on other currently unknown rate-limiting physiological factors [40]. In addition, growth of new leukemic cell populations derived from imatinib resistant silent stem cells in the human body is also considered as one of the potential resistance mechanisms that act independently of OCT-1 expression [41].

Notably, the alteration in OCT-1 expression levels in relation to imatinib exposure is considered to be a dynamic process [14] with direct correlation between the increase in OCT-1 gene expression and treatment duration [27] as well as the decrease in OCT-1 expression levels at the time of imatinib resistance as compared with pretreatment levels [42]. This seems to emphasize higher compatibility of OCT-1 expression when analysis is based on repetitive samples collected at specific time intervals with the dynamic expression process, rather than a single point measurement as performed in the current study.

The large sample size comprising 101 CML patients is the major strength of the current study, in terms of the increased likelihood of reliability of our findings. However, certain limitations to this study should be considered. Firstly, due to the retrospective single center design of the present study, establishing the temporality between cause and effect is not possible. Second, OCT-1 expression analysis in the current study was based on single-point measurements, which is another limitation given the dynamic nature of OCT-1 expression. Third, the lack of a healthy control group is another limitation, which would otherwise enhance the level of knowledge achieved in the current study.

In conclusion, our findings revealed no significant difference in OCT-1 expression levels with respect to imatinib response or KD mutation in CML patients. This seems to indicate that OCT-1 mRNA expression levels may not correlate with response to imatinib therapy among CML patients. Given that development of a treatment resistance phenotype is a multifactorial process underlined by more than one mechanism [43], there is a need for further prospective studies addressing the genetic, molecular and pharmacokinetic mechanisms in the potential association of imatinib transporters with clinical response in CML patients.

Declarations

Conflict of interests

The authors declare that they have no conflict of interest.

Ethıcal Approval

The study was conducted in accordance with the ethical principles stated in the “Declaration of Helsinki”. The study was approved by the Clinical Research Ethics Committee of the Istanbul Medical Faculty, Istanbul University (Decision number/date: 1330/June 26, 2015).

Informed Consent

Written informed consent was obtained from each subject following a detailed explanation of the objectives and protocol of the study.