This article is mentioned in:

Introduction

Lung cancer is the leading cause of cancer-associated mortality globally (1,2). The discovery and application of epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs) have proved beneficial to patients with sensitive EGFR mutations compared with chemotherapy (3–5). However, ~10% of patients with activated EGFR mutations demonstrate primary resistance to EGFR-TKI. Another group of patients with an immediate initial response inevitably developed acquired resistance to TKIs in an average time of ~8 months (6,7). Although, much is known about acquired resistance, only very few primary resistance cases have been elucidated, which involve de novo T790M mutations and Bcl-2-L11 (BIM) deletion polymorphisms (6,8,9). Thus, further study is still required to gain an improved understanding of EGFR-TKI resistance.

A previous study conducted by our group validated the inferior response to EGFR-TKIs in patients carrying the BIM deletion polymorphism (10). BIM is a pro-apoptotic protein that belongs to the B-cell lymphoma-2 (Bcl-2) family (11). Via the BH3 domain, BIM activates downstream pro-apoptotic proteins, including Bcl-2 homologous antagonist killer (BAK) and Bcl-2 associated X protein (BAX) and antagonizes anti-apoptotic proteins, including B-cell lymphoma 2 (Bcl-2). The stimulation of BAK and BAX induces the release of cytochrome c into the cytoplasm and consequently triggers the caspase cascade. Thus, BIM is involved in the mitochondrial apoptotic pathway (12). In EGFR TKI therapy, BIM is pivotal in apoptosis. The BIM deletion polymorphism impairs the expression of BH3-containing BIM isoforms and leads to primary resistance to TKIs (9,10). Epigenetic silencing of BIM, including promoter methylation and histone-deacetylation, has also been reported in hematological malignances; including Burkitt's lymphoma, non-Hodgkin lymphoma, chronic myeloid leukemia (12–14) and solid tumors, including renal carcinoma (15).

DNA promoter hyper-methylation and histone deacetylation are two major, closely associated epigenetic regulatory mechanisms. Histone deacetylation enhances the methylation-induced inhibition of transcription, by forming the transcription-inhibitory complex. The capacity of histone deacetylation depends on the number of methylated CpG islands (16). In cases where the gene promoter region was lacking in methylated CpGs to inhibit transcription, histone deacetylation acted as the major inhibitor in the transcription process. Conversely, methylated CpGs increased the activity of histone deacetylase (HDAC), which further contracted the chromatin and thus inhibited transcription (17). The aim of the present study was to investigate the impact of promoter methylation and histone deacetylation of BIM in NSCLC cell lines and patients with NSCLC, to identify an association with EGFR-TKI resistance.

Materials and methods

Cell culture and in vitro treatment

A total of 8 human NSCLC cell lines were used in the present study. PC9, A549, H1299, H292, H1975 and H520 were obtained from the American Type Culture Collection (Manassas, VA, USA) and 2 gefitinib-resistant PC9 cell lines, PC9/R and PC9/G2, were induced in the laboratory. The culture methods for each cell line have been previously reported in multiple publications and were employed in the present study (18,19). Cultured PC-9 cells were exposed to 2.5 µg/ml N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) for 24 h and then washed and cultured in medium containing 0.2 µM gefitinib for 7 days. Following exposure to gefitinib, cells were washed and cultured in drug-free medium for 14 days. When variable cells had grown, they were seeded in medium containing 0.3–0.5 µM of gefitinib on 96-well cultured plates for subcloning. After 21–28 days, the colonies were harvested and a single clone was obtained. The subcloned cells exhibited an 182-fold increase in resistance to the growth-inhibitory effect of gefitinib as determined by MTT assay, and the resistant phenotype has been stable for at least 6 months under drug-free conditions. H520 cells were cultured in RPMI-1640 medium (Thermo Fisher Scientific, Inc., Waltham, MA, USA) and the remaining cell lines were cultured in Dulbecco's modified Eagle's medium (Thermo Fisher Scientific, Inc.), each supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and penicillin/streptomycin (100 U/ml). All cells were passaged within 3 months from their renewal from frozen or early-passage stocks. A commonly used de-methylating agent, 5-aza-2′-deoxycytidine (AZA; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) which inhibits DNA methyltransferase and Trichostatin A (TSA; Sigma-Aldrich; Merck KGaA), an HDAC inhibitor (HDACi) were used as cell culture supplements. All six cell lines (PC9, A549, H1299, H292, H1975 and H520) were respectively seeded at a density of 2×105 cells/cm2 in 6- or 12-well plates and treated with the following reagents prior to further procedures. Group A cells were treated with 4 µm AZA or 0.01 µm DMSO for 96 h, then 3 µm gefitinib or continued 0.01 µm DMSO (Meilun, Dalian, Liaoning, China) for 72 h. Therefore, four subgroups were produced: Negative control, which received DMSO twice, gefitinib group, received only gefitinib treatment, AZA group, which only received AZA and a combination group, received AZA and subsequently geftinib treatment. Group B cells received 0.02 µm TSA or 0.01 µm DMSO treatment for 16 h, followed by 72 h of 3 µm gefitinib or continued 0.01 µm DMSO, generating four similar subgroups as aforementioned.

Reverse transcription quantitative-polymerase chain reaction (RT-qPCR.)

BIM RNA expression levels were quantified by RT-qPCR. PC9, PC9/R, and PC9/G2 cell lines were randomly assigned into groups A and B and treated as aforementioned. Total RNA was extracted using an RNAiso Plus kit (Takara, Bio, Inc., Otsu, Japan) according to the manufacturer's protocol. Reverse transcription was performed with 1 µg total RNA using the RevertAid First Strand cDNA Synthesis kit (Fermentas; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Gene quantification was analyzed by qPCR, run with 2 µl cDNA and SYBR Premix Ex Taq (Takara, Bio, Inc.), on an Mx3000P qPCR system (Agilent Technologies, Santa Clara, CA, USA). GAPDH was used as the reference gene. The primer sequences were as follows: Bim forward, 5′-AGTGGGTATTTCTCTTTTGACACAG-3′ and reverse: 5′-TCAATGCCTTCTCCATACCAGACG-3′; GAP DH forward, 5′-TCGACAGTCAGCCGCATCTTCTTT-3′ and reverse, 5′-ACCAAATCCGTTGACTCCGACCTT-3′. The reaction conditions were as follows: 94°C for 30 sec, followed by 40 cycles of 95°C for 5 sec, 62°C for 20 sec, with a final extension of 72°C for 10 min. Gene quantification was performed by automatic calculation using LightCycler software (version 1.5; Roche Applied Science, Penzberg, Germany) and BIM expression in the PC9 cell line was considered the base line, as this was the only cell line harboring activated EGFR mutations.

Cell cytotoxicity assay

Cells were seeded at a density of 4,000 cells/well in 96-well plates and treated as aforementioned prior to performing the cytotoxicity assay. Cells in each well were incubated with 100 µl medium and 10 µl Cell Counting Kit-8 reagent (CCK-8; Dojindo Molecular Technologies, Inc., Kumamoto, Japan) at 37°C for 1 h. Following this, the absorbance was measured at 450 nm using a microplate reader. Data were calculated as the percentage of the total absorbance of treated cells/absorbance of non-treated cells and were expressed as the mean ± standard error of the mean.

Methylation analysis

DNA was extracted from every cell line using DNeasy Blood & Tissue kit or QIAamp DNA FFPE Tissue kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's protocol. A total of 500 ng extracted cell line or plasma DNA was converted and purified using an EZ DNA Methylation-Gold kit (Zymo Research, Corp., Irvine, CA, USA), according to the manufacturer's protocol. The EpiTect PCR Control DNA set (Qiagen, Inc., Valencia, CA, USA) was used for methylated and unmethylated controls. Bisulfite converted DNA was analyzed by Methylation-Specific PCR (MSP), nested quantitative methylation specific PCR (nested q-MSP), and direct bisulfite pyrosequencing. For MSP, PCR amplification was performed in a total volume of 25 µl with an ExTaq Polymerase premix PCR kit (Takara, Bio, Inc.), according to the manufacturer's protocol. Primer sequences for methylated and unmethylated BIM promoters from the research of Eneriz et al (12) were used. Reaction conditions were as follows: 95°C for 10 min, followed by 30 cycles of 94°C for 30 sec, 60°C for 30 sec, 72°C for 60 sec, and a final extension of 72°C for 10 min. PCR products were then separated by electrophoresis on 3% agarose gel with ethidium bromide dye.

Nested q-MSP

DNA samples were amplified twice as in the aforementioned MSP protocol to amplify the signal and acquire the second round Cq value. The EpiTect PCR Control DNA Set was used to generate a standard curve to correct the different amplification efficacy of methylated and unmethylated promoters. Bisulfite converted DNA samples were subjected to the same protocol as control DNA mix, and the methylation index (MI) was calculated with the following formula: MI=M/(M+U)=1/[1+ (U/M)]=1/(1+2CtM-CtU).

Direct pyrosequencing

A total of 50 ng converted DNA was amplified using the PyroMark PCR kit (Qiagen GmbH, Hilden, Germany) and the reaction conditions were 95°C for 15 min, followed by 45 cycles of 94°C for 30 sec, 56°C for 30 sec, 72°C for 60 sec, and a final extension of 72°C for 10 min. The primers used for pyrosequencing were: Forward primer, 5′-ATAATGGGGTAGGAGTAGGGAA-3′ and reverse primer, 5′-ATACTCTTTACCCAAAACAAACTT-3′. The specific PCR products were subjected to quantitative pyrosequencing analysis using a Biotage PyroMark Q96 system (Qiagen GmbH) according to the manufacturer's protocol. The results were analyzed using Pyro Q-CpG 1.0.9 software (Qiagen GmbH).

Patient details and EGFR-TKI treatment

This section of the present study was conducted retrospectively. Eligible patients were recruited if the following criteria were met: NSCLC histologically confirmed; received EGFR mutation status analysis; received EGFR-TKI therapy and regular follow up. A total of 32 patients were consecutively identified from the Shanghai Pulmonary Hospital (Shanghai, China) database according to the aforementioned criteria, from November 2011 to September 2012. Patients gave written informed consent for the scientific use of blood samples upon the first occasion of hospitalization. A complete medical history interview, p hysical examination, laboratory test and radiology examination were performed for each case prior to starting treatment. All patients received 250 mg gefitinib (Astrazeneca, Cambridge, UK) or 150 mg erlotinib (Roche Applied Science, Penzberg, Germany) daily until disease progression or the appropriate toxic effect was observed. Tumor response was evaluated every 6 weeks according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (20), and progression free survival was the primary endpoint of the present study.

The present study was approved by the Ethics Committee of Shanghai Pulmonary Hospital Affiliated to Tongji University (Shanghai, China) and was performed in accordance with the World Medical Association's Declaration of Helsinki. Written informed consent was obtained from each patient prior to initiation of any study-associated procedure.

Statistical analysis

All analyses were carried out using PASW 18.0 (SPSS Statistics; IBM Corp., Armonk, NY, USA). Statistical analyses were conducted using one-way analysis of variance to determine statistical significance between RNA expression levels groups in the in vitro studies and Bonferroni method was adopted as post-hoc test. The Kaplan-Meier method was used to plot the PFS curves and a log-rank test was used to compare the differences between groups in clinical data. χ2 test was used to determine the difference in the BIM methylation distribution pattern in the baseline characteristics of patients. P<0.05 was considered to indicate a statistically significant difference.

Results

BIM methylation status of NSCLC cell lines

The methylation status of the BIM promoter of PC9, PC9/R, and PC9/G2 cell lines using MSP was investigated (Fig. 1). PC9, PC9/R, and PC9/G2 cell lines all demonstrated hypo-methylation in BIM promoters. Pyrosequencing and nested q-MSP were used to further confirm the BIM promoter methylation status in the 8 cell lines. The CpG islands analyzed were selected based on the results from a previous study (12). The resulting methylation index (MI) of pyrosequencing is listed in Table I. According to a previous study on quantitative methylation analysis (12), DNA samples with MI≥50% were identified as hyper-methylated; those with MI≥20% were identified as partially methylated and MI<20% were considered hypo-methylated. Taken together, the results from MSP, pyrosequencing and nested q-MSP indicated that the hypo-methylated status of BIM existed in all tested cell lines; including gefitinib sensitive PC9; gefitinib induced drug-resistance cell lines PC9/R and PC9/G2; EGFR TKI-primary resistance adenocarcinoma cell lines H292, A549, H1975; large-cell carcinoma H1299 and squamous carcinoma H520 (Table I and Fig. 1C). The two quantitative testing methods resulted in different methylation indexes. The divergence may reflect the difference between global and regional methylation status of the BIM promoter region.

Figure 1. BIM promoter methylation analysis. (A) MSP analysis of the BIM promoter in the PC9 cell lines. PC9 and its two derivative, gefitinib resistant cell lines all presented with hypo-methylated BIM. (B) Pyrosequencing analysis of all studied cell lines. Blue regions mark the analyzed positions. (C) Comparison of the BIM methylation index by pyrosequencing and nested q-MSP in all studied cell lines. BIM, bcl-2-like protein 11; MSP, methylation specific polymerase chain reaction; q-, quantitative; M, methylated; U, unmethylated.

Table I. MI of tested cell lines by pyrosequencing and q-MSP.

Table I.

MI of tested cell lines by pyrosequencing and q-MSP.

	MI (by pyrosequencing)
Cell lines	Position 1 (%)	Position 2 (%)	MI (by q-MSP) (%)
PC9	2	4	6.22±0.16
PC9/R	3	4	5.99±0.06
PC9/G2	3	3	10.85±0.56
H1299	6	6	16.59±0.93
H292	2	3	16.70±2.64
A549	3	4	15.52±1.37
H1975	8	5	15.21±1.25
H520	7	7	15.94±1.14

[i] q-MSP, quantitative methylation specific polymerase chain reaction; MI, methylation index.

BIM RNA expression level changes following treatment with AZA or TSA

In response to gefitinib treatment, two gefitinib resistant cell lines presented with reduced BIM mRNA levels compared with PC9 cells. Following AZA treatment, three cell lines demonstrated increased BIM transcript levels (Fig. 2A), although no significant differences were observed. Conversely, cell lines receiving TSA treatment demonstrated markedly elevated BIM mRNA levels compared with cells with no prior TSA stimulation (P<0.01; Fig. 2B).

Figure 2. In vitro assays. BIM gene expression levels in PC-9 cell lines following (A) AZA or (B) TSA treatments. Analysis of (C) PC9, (D) PC9/R and (E) PC9/G2 cell viability, following pre-treatments with TSA or AZA prior to gefitinib treatment. Cells pre-treated with TSA demonstrated lower tolerance to gefitinib compared with those that received gefitinib treatment alone. AZA pre-treatment, however, resulted in higher cell line resistance. Progression free survival curves of (F) the entire study group and (G) EGFR mutant groups. **P<0.01 vs. cells treated with DMSO and gefitinib. BIM, bcl-2-like protein 11; AZA, 5-Aza-2′-deoxycytidine; TSA, Trichostatin A; EGFR, epidermal growth factor receptor; PFS, progression free survival; NC, negative control.

Cell cytotoxicity analysis

PC9, PC9/R and PC9/G2 cells were treated with the same treatment regime as described in the previous section. Following treatments, Cell Counting Kit-8 (CCK-8) was utilized to test cell toxicity (Fig. 2C-E). In all three cell lines, cells pre-treated with TSA presented with a lower tolerance to gefitinib compared with cells that received gefitinib treatment alone. The effects of AZA pre-treatment, however, resulted in a higher resistance in all three tested cell lines.

Distribution of BIM methylation indexes in patients with NSCLC

The clinicopathological characteristics of the 32 patients with NSCLC recruited into the present study are listed in Table II. Amongst the 32 patients, the lowest methylated BIM index identified was 5.88, whilst the highest was 106.58. Patients were divided into three groups according to standards described in the ‘BIM methylation status of NSCLC cell lines’ section, and the subsequent associated data are listed in Table III. Patients with low BIM MI comprised the majority of the study group (78.1%), whilst those with medium and high MI accounted for 15.6 and 6.3% of all patients, respectively. Data from the latter two groups were merged into one group (methylated group, group M) in the following analysis due to the small identified sample size, and patients with low BIM were named the non-methylated group (group N) in the subsequent analysis. The distribution between the M and N groups revealed no significant difference when assessed using a χ2 test assessing multiple clinicopathological characteristics, including gender (P=0.96), smoking status (P=0.98), performance status (P=0.96), histological type (P=0.80) and EGFR mutant status (P=0.24).

Table II. Clinical characteristics and MI distribution of patients with NSCLC with a median age of 56 (range, 32–78).

Table II.

Clinical characteristics and MI distribution of patients with NSCLC with a median age of 56 (range, 32–78).

		MI
Clinicopathological characteristics	Case (%)	M	N	P-value
Sex				0.96
Male	14 (43.8)	3	11
Female	18 (56.3)	4	14
Smoking status				0.98
Smoker	9 (28.1)	5	18
Non-smoker	23 (71.9)	2	7
PS				0.96
1	18 (56.3)	4	14
2	14 (43.8)	3	11
Histological cell type				0.80a
Adenocarcinoma	24 (75)	5	19
Squamous cell	3 (9.4)	1	2
Adenosquamocarcinoma	2 (6.3)	0	2
Othersb	3 (9.4)	1	2
EGFR status				0.23c
Wild type	9 (28.1)	3	6
19Del	12 (37.5)	6	6
L858R	9 (28.1)	6	3
Others	2 (6.3)	1	1

a Adenocarcinoma vs. non-adenocarcinoma

b NSCLC-NOS (not otherwise specified) and large cell carcinoma

c EGFR wild type vs. EGFR mutant. MI, methylation index; PS, performance status; NSCLC, non-small cell lung carcinoma; EGFR, epidermal growth factor receptor; M, methylated; N, non-methylated.

Table III. BIM promoter methylation status in patients with non small cell lung carcinoma.

Table III.

BIM promoter methylation status in patients with non small cell lung carcinoma.

Variable	Hypo-methylated	Partially-methylated	Hyper-methylated
Case (%)	25 (78.1)	5 (15.6)	2 (6.3)
Mean MI (%)	9.08	22.67	80.48
(95% CI)	(7.32–10.84)	(19.69–25.66)	(−251.23–412.18)
Minimum-maximum MI (%)	5.88–18.11	20.81–26.41	54.37–106.58
Median MI (%)	6.32	21.17	80.48

[i] MI, methylation index; CI, confidence interval.

Analysis of BIM MI and EGFR TKI efficacy

All 32 patients received EGFR TKI, either gefitinib or erlotinib. By the cut-off point of the present study, none of the patients had stopped taking the medication due to intolerable toxicity prior to progression of the disease (PD). Of those patients, 31 (96.9%) experienced PD, and 1 still had a stable disease (SD). The median PFS (mPFS) for the entire group was 5.75 months, with the shortest 0.5 months and longest 38 months. The EGFR TKI efficacy in the entire group and EGFR mutant subgroup with different BIM MI is listed in Table IV. In both the entire study population and EGFR mutant patients, PFS did not significantly differ between the M and N groups (Table IV, Fig. 2F-G).

Table IV. Efficacy of EGFR-tyrosine kinase inhibitor treatment in patients with non small cell lung cancer.

Table IV.

Efficacy of EGFR-tyrosine kinase inhibitor treatment in patients with non small cell lung cancer.

Group	Median PFS (95% CI)
	Study population	EGFR mutants
N	5.500 (0.000–12.844)	5.000 (0.000–12.563)
M	6.000 (3.434–8.566)	6.000 (0.000–13.201)
Total	5.500 (0.000–11.506)	6.000 (0.000–12.104)

[i] N, hypo-methylated BIM; M, partial- and hyper-methylated BIM; EGFR, epidermal growth factor receptor.

Discussion

The aim of the present study was to investigate the impact of BIM promoter methylation and histone deacetylation on EGFR TKI resistance. In all NSCLC cell lines examined, and samples from the majority of patients with NSCLC, BIM gene promoter regions were hypo-methylated. The demethylating agent, AZA, did not affect BIM mRNA expression in EGFR TKI-resistant cell lines. HDACi, TSA significantly increased BIM mRNA levels and reversed the TKI resistance of PC9/R and PC9/G2 cell lines.

Epigenetic BIM silencing has been previously reported to be associated with various malignancies (12,13). Furthermore, abnormal methylation has been identified as a hallmark of cancer (21,22). In previous years, the involvment of epigenetic regulation in the acquisition of drug resistance has been addressed due to development of cancer genome charts (23). In NSCLC, genes including multiple tumor suppressor 1 (p16), adenomatous polyposis coli (APC) and death associated protein kinase 1 were proven to be hyper-methylated in the promoter region (24,25).

In the present study, BIM was observed to be hypo-methylated in NSCLC, as evaluated by various methods, indicating the difference between NSCLC and hematological malignancies or renal carcinoma. Despite the heterogeneity in the tumorigenesis and development of NSCLC, due to the discrepancies observed, it was not possible to exclude epigenetic alteration of cell lines following in vitro culture. However, further exploratory research using circulating free DNA from patients with NSCLC confirmed that the hypo-methylated status of BIM was observed in the majority of patients (78.1%). The collective in vitro and in vivo results suggested that BIM was hypo-methylated in the majority of NSCLC cases and that promoter methylation did not contribute to EGFR TKI resistance. Increased resistance to gefitinib following the AZA-induced treatment in PC9, PC9/R, and PC9/G2 cell lines revealed AZA's protection of cell lines from gefitinib. Such protection suggested that anti-apoptotic and/or cell cycle regulating genes may be switched on following AZA treatment, thereby rescuing epigenetic silencing.

The two PC9 gefitinib-resistant cell lines presented with impaired BIM mRNA expression, indicating that epigenetic factors may still regulate BIM. Thus, histone deacetylation, another epigenetic regulating mechanism associated with methylation, was further investigated. TSA pre-treatment led to a marked upregulation of BIM and gefitinib tolerance reduction in PC9 gefitinib-resistant cell lines, indicating that histone deacetylation, as opposed to methylation may be involved in the epigenetic BIM silencing. Another HDACi, verinostat, has been demonstrated to increase BIM expression in EGFR mutant NSCLC cell lines, harboring either wild type or deletion polymorphism BIM under TKI treatment (26). However, BIM methylation and histone deacetylation status were not directly detected in the aforementioned study. A previous study conducted on Burkitt lymphoma revealed the co-existence of promoter methylation and histone deacetylation of BIM in cell lines; and demonstrated that verinostat was able to reverse chemotherapy resistance by overcoming epigenetic silencing (13). However, in a clinical trial comparing erlotinib, with or without HDACi entinostat in patients with advanced stage NSCLC (27), the combined regime did not prove to be beneficial, which may highlight the importance of patient selection. Another previous study analyzed the single and combined effects of AZA and TSA on a multi-drug resistant cell line, and demonstrated that the anti-resistant effect of each single drug is not additive but interactive (28), offering a new viewpoint for our follow-up study on the investigation of combined AZA and TSA to treat NSCLC.

To the best of our knowledge, the present study is the first to use a quantitative method to analyze methylation status using circulating free DNA (cfDNA). Over previous decades, researchers have identified the existence of cfDNA (29,30). Tumor-specific changes including point mutation, DNA methylation, and microsatellite instability revealed that these cfDNAs originated from tumors (31,32). Since peripheral blood is considerably easier to acquire compared with tissue samples, the use of cfDNA may be beneficial to detect and/or monitor the methylation profile of genes of interest. Multiple studies have been conducted which detect methylation using cfDNA, including APC, RAS, and the Ras association domain family 1A gene in ovarian cancer (33); tumor suppressor gene BLU, Cadherin 13, fragile histidine triad and p16 in lung cancer (34). MSP, which provides qualitative results, is a commonly used method in such studies, and the incidence of methylated DNA detected in serum is commonly reduced compared with that in tumor tissue. In the present study, the aim was to combine nested PCR with quantitative MSP in order to acquire precise methylation information, using cfDNA.

In the present study, the small proportion (6.3%) of patients with hyper-methylated BIM indicated that patients with NSCLC often present with hypo-methylated BIM. These observations are consistent with the results presented from the in vitro study cell lines. PFS analyses revealed that no significant differences were observed between the M and N patient groups in the entire study group (P=0.859) and EGFR mutant subgroup (P=0.395); indicating that there was no existing association between BIM methylation and EGFR TKI efficacy in patients with NSCLC. There are several potential explanations for the results presented here. As previously stated, BIM methylation status may not influence EGFR TKI efficacy in patients with NSCLC. In addition, the BIM methylation index detected in cfDNA may not reflect the in vivo tumor tissue status, since the phenomenon of methylated DNA detected in serum was reduced compared with that in tumor tissue, which was also observed in several previous studies (25,33,34). Furthermore, although the present study was exploratory, the population size was small and the test efficiency weak. Hence, a paired study comparing methylation profiles in cfDNA and tissue with increased case numbers may reflect the function of BIM methylation in NSCLC target therapy more accurately.

In conclusion, the present study detected the BIM methylation profile in NSCLC cell lines and patients with NSCLC, and the cell lines and patients collectively presented with hypo-methylated BIM. The histone deacetylation inhibitor, TSA, but not the methylation inhibitor, AZA, reversed the resistance to EGFR TKI in acquired resistance cell lines PC9/R and PC9/G2. Thus, histone deacetylation as opposed to promoter methylation, may contribute in the epigenetic silencing of BIM and lead to EGFR-TKI resistance in NSCLC.

Acknowledgements

The authors would like to thank Amoy Diagnostics Company, Ltd., for technical support. The abstract was presented at Annual Meeting of The European Society for Medical Oncology, September 27th 2014 in Madrid, Spain and published as abstract no. 1320P in Annals of Oncology 25 (supplement 4): 2014.

References