SNPs of ABCG2 in Chinese Advanced NSCLC Patients

SNPs of ABCG2 in Chinese Advanced NSCLC Patients Introduction Lung cancer is one of the most prevalent and fatal malignant neoplasm all over the world and non-small-cell lung cancer (NSCLC) accounts for 80%–85% of all lung cancers(1). The majority of NSCLC patients, approximately 80%, presents in locally advanced (phase IIIA/B) or metastatic (phase IV) stages, which results in quite low 5-year survival rates, 8-14.1% for phase IIIA and 1-5% for phase IIIAB/IV (2). The standard treatment of advanced NSCLC, two-drug chemotherapy based on platinum, has reached a bottleneck with limited effect. Tyrosine kinase inhibitors (TKIs), a targeted drug of epidermal growth factor receptor (EGFR), have been recently introduced for the treatment of NSCLC. Clinical trials indicated that Gefitinib and Erlotinib treating advanced NSCLC patients with EGFR mutation could result a remission rate of 62.1%~84.6% and progression-free survival (PFS) of 8.4~13.1 months, which are significantly higher than that in chemotherapy group (32.2%~47.3% and 4.6~6.7 months , respectively), but not over survival(3-6). In order to implement accurate treatment of both chemotherapy and targeted therapy, it’s urgent to find other predictive targets of NSCLC patients to stratify for treatment. ATP binding cassette superfamily G member 2 (ABCG2), also known as breast cancer resistance protein, was demonstrated to be associated with the effect and prognosis of chemotherapy/targeted therapy in NSCLC (7-9). Because the single nucleotide polymorphisms (SNPs) of ABCG2 are supposed to affect the expression of ABCG2 protein and SNPs of ABCG2 in Asian population are different from other ethnicities (10), we conducted this study to evaluate the SNPs of ABCG2 in Chinese advanced NSCLC patients and its association with their prognosis of TKI therapy. Materials and methods Patients and treatment A total of 100 patients with pathology and cytology confirmed advanced or metastatic NSCLC were enrolled into this study between April 2012 and January 2014 in Hangzhou, China. The mutation of EGFR gene was assessable in 32 patients. Other patients were not assessed EGFR mutation. TKI targeted therapy was implemented in 70 NSCLC patients and other therapy was implemented in the other 30 patients. Patients with TKI targeted therapy were treated with Gefitnid (Astrazeneca pharmaceutical co., LTD) at a dose of 250 mg/day or Erlotinib (Roche pharmaceuticals co., LTD) at a dose of 150 mg/day or Icotinib (Zhejiang beida pharmaceutical co., LTD) at a dose of 375 mg/day. The patients’ characteristics were detailed in Table 1. All patients received chest CT every two months after 1 month of therapy. The efficacy of TKI therapy was clarified as complete response (CR), partial response (PR), stable disease (SD) and progression disease (PD) according to RECIST 1.1 [1]. Patients with CR or PR at more than 6 months were considered as responders. Patients with SD and PD at less than 6 months were considered as nonresponders.[A1] Progression-free survival (PFS) was defined as the duration from TKI therapy to disease progression. Overall survival (OS) was defined as the duration from diagnosis to death from any cause. All patients agreed to participate in this study and signed written informed consent. This study was approved by the Institutional Review Board of Nanjing Medical University and performed in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines[A2]. DNA extraction Blood samples were collected before chemotherapy and kept in a microcentrifuge tubes containing ethylenediamine tetra-acetic acid (EDTA). Genomic DNA was extracted using a DNA purification kit (Flexi Gene DNA Kit, Qiagen, Hilden, Germany). The concentration of genomic DNA was determined with NanoDrop 1000 (Thermo Scientific, Wilmington, USA) and then it was diluted to a standard of 25 ng/ÃŽ ¼l. Analysis of ABCG2 polymorphisms The ABCG2 34 G/A (dbSNP ID: rs2231137), 421 C/A (dbSNP ID: rs 2231142), 1143 C/T (dbSNP ID: rs2622604) and -15622 C/T (dbSNP ID: rs7699188) polymorphisms were amplified by PCR with the appropriate primers. The primers for PCR and single base extension (Table 2) were designed by the Sequenom Assay Designer 3.1 Software (San Diego, CA). The PCR reactions[A3] were performed at 95 °C for 2 min, followed by 40 cycles at 95 °C for 30 s, Tm for 30 s, and 72 °C for 60 s. After PCR amplification, single base extension reaction was performed following the method of Wiltshire et al [2]. Finally, polymorphisms of ABCG2 gene were tested and analyzed using matrix-assisted laser desorption/ionization timeof-flight mass spectrometry (MALDI-TOF MS) and Sequenom MassARRAY system (Sequenom, San Diego, CA, USA). Statistical analysis Allele frequencies of SNPs were calculated and their genotype distributions were assessed using Fisher’s exact test or chi-square test. PFS and OS were evaluated with censored survival time methods and 95% confidence intervals (CI) was obtained from multivariable logistic regression. Kaplan-Meier survival curves were plotted for OS and analyzed with log-rank test. All tests were performed 2-sided and a p-value were carried out using SPSS 18.0 (SPSS Inc., Chicago, IL, USA) software. Results ABCG2 gene polymorphisms The genotyping of ABCG2 34 G/A, 421 C/A, 1143 C/T and -15622 C/T were performed in all these 100 patients. For the ABCG2 34 G/A polymorphism, the frequencies of GG, GA and AA genotypes were 36%, 50% and 14%, respectively. The allele frequencies of G and A were 61% and 39%, respectively. The wide-type ABCG2 421 C/A genotype (CC) had a frequency of 53%, while the CA and AA genotypes were found in 43% and 4% of the patients, respectively. The allele frequencies of G and A were 74.5% and 25.5%, respectively. The frequencies of CC, CT and TT genotypes for ABCG2 1143 C/T were 66%, 29% and 5%, respectively. The allele frequencies of G and A were 80.5% and 19.5%, respectively. Regarding the ABCG2 -15622 C/T polymorphism, the TT genotype was observed in all patients. Therefore, polymorphism of ABCG2 -15622 C/T was not investigated in the following steps. Polymorphisms of ABCG2 and clinical characteristics Patients clinical characteristics were shown in Table 1, and the relationship between polymorphisms of ABCG2 and clinical characteristics were presented in Table 3. No significant correlations were found between ABCG2 polymorphisms (34 G/A, 421 C/A and 1143 C/T) and patients’ characteristics, including gender, age, smoking history, histology and EGFR mutation (p > 0.05). Although there was no significant relationship between ABCG2 421 C/A polymorphism and EGFR mutation, a trend that CA genotype was observed frequently in EGFR mutation positive patients (47.6% in positive patients vs. 18.2% in negative patients, p = 0.119). Then we calculated the allele frequency of A in these patients and a high frequency of allele A in positive patients (33.3% vs. 9.1%, p = 0.038) was observed. Polymorphisms of ABCG2 and clinical outcome of TKI The sensitivity of 70 patients to TKI treatment was shown in Table 4. NO significant correlation was found between ABCG2 polymorphisms (34 G/A, 421 C/A and 1143 C/T) and sensitivity (p > 0.05). As shown in Table 4, median PFS for carriers of the A-allele and GG genotype at position 34 of the ABCG2 gene who were treated with TKI therapy was 8.0 months (95% CI: 5.9-10.1, n = 45) and 6.5 months (95% CI: 4.1-8.9, n = 25), respectively. There was no significant difference in median PFS of NSCLC patients receiving TKI therapy between CC genotype and CA + AA genotype at position 421 of ABCG2 gene (p > 0.05). Median PFS of patients with CC genotype at position 1143 of ABCG2 gene was higher than those with CT and TT genotypes, but no significant difference was found (p > 0.05). The median OS of patients with ABCG2 34 G/A, 421 C/A, 1143 C/T polymorphisms was shown in Table 4. The median OS of patients with GG genotype at position 34 of the ABCG2 gene was 18 months (95% CI: 14.9-21.1, n = 25) and for those with other genotypes (GA and AA) was 31 months (95% CI: 22.9-39.1, n = 45). Figure 1 showed the Kaplan-Meier curve for OS for NSCLC patients receiving TKI therapy in relation to ABCG2 genotypes at 34 G/A (Figure 1A), 421 C/A (Figure 1B) and 1143 C/T (Figure 1C). There was significant difference between patients with GG genotype and those with GA + AA genotypes at position 34 of the ABCG2 gene (p difference between patients that were CC genotype regarding the position 421 of ABCG2 gene and carriers with other genotypes (CA + AA, p > 0.05). No significant difference was found in 1143 C/T polymorphism (p > 0.05). Discussion Our present study observed that three polymorphisms of ABCG2, 34G>A, 421C>A and 1143C>T occured more frequently compared with -15622C>T in Chinese advanced NSCLC patients. As for -15622C>T, all patients presented a TT genotype. Although no relationships were observed between different genotypes of ABCG2 polymorphisms and EGFR status, a higher frequency of allele A (421C>A) in EGFR mutation positive patients was observed. The other polymorphisms were not related to clinical characteristics. The sensitivity and PFS to TKI of 70 patients was not related to polymorphisms. However, the OS of patients with 34G>A mutant type (GA+AA) was significantly longer than those with wild type (GG). The ABCG2 protein is an important member of the ABC transporter superfamily, which has been suggested to be involved in multi-drug resistance (MDR) in cancer. Screening for SNPs in ethnically diverse subjects has identified more than 80 synonymous and nonsynonymous SNPs in the ABCG2 gene to date (12). The two most frequent polymorphisms identified were 34G>A (resulting in V12M) and 421C>A (resulting in a Q141K substitution) transitions (13). A novel diplotype of two polymorphic loci in the ABCG2 promoter involving -15622C>T and 1143C>T were identified recently (14). Introduction of other ABCG2 SNPs can be found in a recent review (15). Despite the similar allele frequency of 421C>A variant among East Asian populations including Chinese (34.2–35.0%) and Japanese (26.6–35.0%), the allele frequency is higher than that of Southeast Asians (15.0%), Middle Easterns (13.0%), Caucasians (8.7–12.0%) and African-Americans (2.3%) (10). Similarly, the allele frequency of the 34G>A variant in Chinese (20.0%), Koreans (19.8%) and Japanese (15.0-19.0%) is comparable. However, it is much lower than that in Southeast Asians (45%) and higher than other ethnic groups including Caucasian (1.7–10.3%), African-American (6.3%) and Middle Eastern (5.0%) populations (10). The allele frequency of 421C>A variant in our studied population was 25.5%, which was comparable to other Asian populations. However, the allele frequency of 34G>A variant was 39.0%, which was higher than other reports from Asian populations. We found that the allele frequency of 1143C>T variant and -15622C>T variant in our study was 19.5% and 100%, respectively. In Caucasians, it was reported to be 22% and 28%, respectively (16). We unexpectedly observed that all the included patients presented TT genotype of -15622C>T. As far as we known, this gene has not been investigated in other Asian populations. Future studies could be conducted to determine the polymorphism of -15622C>T in Asian po pulation and its potential impact. Physiologically, ABCG2 protein is highly expressed in the blood-brain barrier and gastrointestinal tract, where it is thought to play a role in protection against xenobiotic exposure. High ABCG2 expression has also been found in a variety of tumors and correlated with multidrug resistance and poorer clinical outcomes, as this transporter has the ability to extrude its drug substrates out of the cells, thereby decreasing their intracellular accumulation (17, 18)[16]. Primary structural variations of ABCG2 are associated with its drug-transporter function (15). Therefore, SNPs in the ABCG2 gene would influence the pharmacological effects differently in different patients. It has been demonstrated that 421C>A polymorphisms may express low amounts of ABCG2 (19-22) while the influence of 34G>A polymorphisms on ABCG2 expression remains controversial (22, 23). And regarding to 1143C>T and -15622C>T, some researchers found a decreased protein expression related to these two polymorphisms (21 ) and others found no relation between them (24). Moreover, 421C>A polymorphism has been demonstrated to be associated with ATPase activity and drug transport (18). Thus, several clinical studies have investigated the relation between ABCG2 polymorphism and clinical outcome of NSCLC. Mà ¼ller and colleagues (25) found that carriers of the ABCG2 421 A-allele treated with platinum-based drugs showed a significantly worse OS in all lung cancer patients. However, this effect was not statistically significant in the smaller subgroups of SCLC patients or NSCLC patients with platinum-based treatment. They did not found an association between 34G>A polymorphism and prognosis. Another study of 129 unresectable NSCLC cases treated with first-line platinum-based chemotherapy suggested that ABCG2 SNPs rs2725264 and rs4148149 were associated with OS (26). On the other side, there was also evidence showing that ABCG2 polymorphisms were not related to response or prognosis of NSCLC patients treated with gefitinib (24), erlotinib (27) and gemcitabine and/or platinum-based drugs and/or other drugs (28). In our present study, we found the OS of patients with 34G >A mutant type (GA+AA) was significantly longer than those with wild type (GG). However, we did not observe significant differences concerning other polymorphisms including 421C>A, which was found to be associated with prognosis of other cancer by other study (29). Interestingly, it was reported that ABCG2 34 GA/AA genotypes were associated with poor prognosis of Chinese patients with acute leukaemia (30). Polymorphisms of 34G>A seems to have an opposite impact in different types of cancer. The mechanisms are worthy to be investigated in future large studies. Moreover, ABCG2 SNPs was demonstrated not only related to TKI resistance, but also to TKI induced side effects. Cusatis and colleaguesinvestigated associations between allelic variants ofABCG2 with diarrhea and skin toxicity ingefitinib-treated patients. They found that 16 patients heterozygous forABCG2 421C>A developed diarrhea, versus only 13 (12%) of 108 patients homozygous for the wild-type sequence. However, this SNP was not associated with skin toxicity (28). A recent study found that patientscarrying anABCG2 -15622 TT genotype or harboring at least one TT copy in theABCG2 (1143CT, -15622CT) haplotype developed significantly more grade 2/3 diarrhea (23). In our present study, we did not perform the analysis on side effects. However, this is a serious concern which should be taken into consideration in future studies. In Conclusion, Our findings demonstrate a strong association between the ABCG2 34G>A polymorphism and the overall survival of NSCLC patients treated with TKIs, including Gefitnib, Erlotinib and Icotinib. Since these polymorphisms can be assessed with a simple blood test, it might potentially improve the stratification of patients for TKI treatment by identifying genetically high-response subgroups. Therefore, larger prospective trials are warranted to validate these findings. [A3]The PCR reactions were performed in 20 ÃŽ ¼l volumes on 384-well plates (cat. No. TF-0384/W, ABgene, USA) with 20 ng DNA, 10 pmol for each primer and 1 Ãâ€" PCR-Buffer (Sequenom, San Diego, CA, USA). à §Ã‚ ¼Ã‚ ºÃƒ ¤Ã‚ ¸Ã…“à ¨Ã‚ ¥Ã‚ ¿Ãƒ £Ã¢â€š ¬Ã¢â‚¬Å¡