==== Front Medicine (Baltimore)Medicine (Baltimore)MEDIMedicine0025-79741536-5964Wolters Kluwer Health 29995771MD-D-17-0768010.1097/MD.0000000000011346113465700Research ArticleObservational StudyAssociation between advanced NSCLC T790 M EGFR-TKI secondary resistance and prognosis A observational studyWang Yuli MSaWei Yuan MSbMa Xiaoping MSaMa Xinyu MScGong Ping MDa∗Guo. Weimin a The First Affiliated Hospital, School of Medicine, Shihezi University, Shihezib Changji Branch Hospital, the First Affiliated Hospital of Xinjiang Medical University, Changjic School of Medicine, Shihezi University, Shihezi, Xinjiang, China.∗ Correspondence: Ping Gong, the First Affiliated Hospital, School of Medicine, Shihezi University, North 2nd Road, Shihezi, Xinjiang 832003, China (e-mail: 15001635050@163.com).7 2018 13 7 2018 97 28 e113467 12 2017 2 6 2018 Copyright © 2018 the Author(s). Published by Wolters Kluwer Health, Inc.2018This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. http://creativecommons.org/licenses/by-nc-nd/4.0.Supplemental Digital Content is available in the text Abstract Epidermal growth factor receptor (EGFR) mutations for EGFR-tyrosine kinase inhibitors (EGFR-TKI) in non-small cell lung cancer (NSCLC) patients are with clinical benefits. Nevertheless, eventual resistance to EGFR-TKI is almost inevitable. In about 50% patients, EGFR-TKI develops a secondary mutation, which is often the T790 M mutation. We aimed to investigate the relationship between EGFR gene status in the peripheral blood and prognosis (progression-free survival [PFS] and overall survival [OS]) in advanced lung adenocarcinoma patients and the 20 exon 790 site mutation (T790 M) and acquired resistance to EGFR-TKI. A total of 49 patients with EGFR-TKI resistance and advanced lung cancer who visited the Shihezi University School of Medicine between 12/2013 and 12/2014 were enrolled in this study. Peripheral blood plasma DNA was isolated after EGFR-TKI resistance and the EGFR exon 20 T790 M mutation was detected using the probe amplification refractory mutation system method. The T790 M mutation rate was 30.6% (15/49). There was no association between T790 M mutation and age, gender, smoking, clinical stage, Eastern Cooperative Oncology Group rating, initial EGFR mutation, and EGFR-TKI drugs, but EGFR-TKI resistance was associated with progression (P = .009). Median progression-free survival (PFS) of patients with T790 M mutation was 9.6 months and median overall survival (OS) was 17.6 months, compared to 6.8 and 12.7 months in controls (P = .018 and P = .027). Multivariate analysis showed that T790 M mutations independently affected the PFS (risk ratio, RR = 0.653, 95% confidence interval, CI: 0.069–0.886, P = .032) and OS (RR = 0.847, 95% CI: 0.208–2.696, P = .008). T790 M mutation and EGFR-TKI resistance are independent factors to affect PFS and OS of non-small cell lung cancer patients. Keywords epidermal growth factorfree DNAnon-small cell lung cancerprognosisOPEN-ACCESSTRUE ==== Body 1 Introduction Lung cancer is one of the most common cancers worldwide, and over 80% of them was non-small cell lung cancer (NSCLC).[1] Surgery and combined therapy are mainly used for early lung cancer treatment while radiotherapy and chemotherapy are the main treatments for advanced lung cancer with 15% to 40% efficiency[2–4] and the median survival of 9 months.[5] The guidelines of National Comprehensive Cancer Network (NCCN) pointed out that epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (EGFR-TKI) can be selected as the first-line treatment for advanced NSCLC when EGFR mutation is detected. EGFR-TKI treatment in NSCLC patients with EGFR mutations showed clinical benefit rates around 70%, progression-free survival (PFS) of 6 to 12 months, and overall survival (OS) of 2 to 24 months.[6,7] It is currently widely accepted that EGFR gene mutations are a key predictor for the efficacy of EGFR-TKI, and it is necessary to detect the EGFR mutations before treatment planning. Indeed, some evidence suggests that EGFR-TKI are not effective for all patients with NSCLC and EGFR mutations.[8–13] Wrong treatment might lead to treatment delays, making the patients lose the best treatment timing and suffer economic losses. Therefore, how to further improve the prediction of EGFR-TKI efficacy has become an urgent problem to be solved. Resistance to EGFR-TKI treatment is almost inevitable.[11,14] At present, the mechanisms of EGFR-TKI resistance include the EGFR 20 exon 790 site mutation (T790 M), c-Met amplification, and phosphatidylino-sitol 3-kinases (PIK3CA) mutation.[15,16] These mutations can be overcome by additional drugs, but with additional toxicities and financial burden. For example, mesenchymal epithelial transition factor inhibitor (ARQ197) was developed to overcome the amplification of c-Met gene leading to EGFR-TKI resistance. A phase II clinical trial in 2011 investigated the efficacy of erlotinib ± ARQ197 in advanced NSCLC patients with failed previous treatment, and confirmed that the combination significantly reduced the risk of disease progression, but without significant improvement of median PFS.[17] Another drug, afatinib, interferes with EGFR autophosphorylation and inhibits the growth of EGFR-TKI resistant NSCLC.[18] It is also effective in the presence of the T790 M mutation.[19] Nevertheless, EGFR mutations detecting is needed to prescribe the best treatment. For EGFR gene mutations screening, a certain amount of fresh tumor tissues is required, which is not always possible. Under this circumstance, the screening for EGFR gene mutations becomes even more difficult. Indeed, most patients with NSCLC at an advanced stage are treated with chemotherapy, targeted therapy, and radiation therapy, without surgery. Therefore, fresh tumor samples are not obtained. The use of blood samples could solve this issue. Circulating tumor deoxyribonucleic acid (DNA) can be found in peripheral blood,[9,20–22] allowing for the determination of the EGFR gene status and guiding treatment. The non-invasive detection of the EGFR T790 M mutation in peripheral blood is feasible in cases where tumor biopsy is difficult to obtain. It has been shown that the agreement between the EGFR status in the tumor and the blood reached 80% to 90%.[6,7,11] In order to analyze its association with clinical features, therapy response, and prognosis, free tumor DNA from the peripheral blood using the Amplified Refractory Mutation System (ARMS) assay to determine the presence of the T790 M mutation in patients with EGFR-TKI resistant advanced NSCLC. The results could help to explain the causes of EGFR-TKI resistance and provide some basis for subsequent selection of drug treatment options. 2 Materials and methods 2.1 Specimen source A total of 114 patients with advanced NSCLC receiving EGFR-TKI therapy (all patients had indications for EGFR-TKI therapy) at the First Affiliated Hospital of Shihezi University School of Medicine between December 2013 and December 2014, and 60 patients were with EGFR-TKI resistance (either primary or acquired). Primary resistance was defined as the absence of any curative effect with EGFR-TKI. Acquired resistance was defined as an initial response to the EGFR-TKI, but that progressed to resistance after at least 6 months.[15] The inclusion criteria were: histological or cytological stage IIIB/IV NSCLC; the patient accepted voluntary EGFR-TKI targeted therapy and treatment before EGFR genetic testing, specifically for EGFR gene mutation, and were able to comply with the study and follow-up; the patient underwent CT imaging every 2 months after starting targeted therapy until disease progression, which was defined as a relative increase of at least 20% of the index lesion (minimum of 5 mm); and blood samples were collected (n = 49 patients). The exclusion criteria were: Eastern Cooperative Oncology Group (ECOG) score >3 points; or unknown EGFR gene status. This study was approved (AF/SC-08/01.0) by the Ethics Committee of Shihezi University School of Medicine at the First Affiliated Hospital (Reference number: 2015–129–01). All patients signed informed consent before blood collection. 2.2 Sample collection, separation, and processing Fasting blood samples (5 mL) were collected from all patients with disease progression. The blood sample was added to a centrifuge tube (radius of 8 cm) with a serum separation gel and then centrifuged at 3000 rpm for 10 minutes. The upper serum layer was collected. Free DNA was extracted from peripheral blood within 4 to 5 hours using the QIAamp Blood Mini kit (Qiagen, Venlo, the Netherlands) and stored at –80°C. The concentration and purity of the DNA samples were tested at A260/280 nm with a UV spectrophotometer. Using a probe amplification refractory mutation system (Amplification Refractory Mutation System, ARMS) method, the patient's peripheral blood plasma DNA was tested with the human ADx-EGFR mutation detection kit (Xiamen Eide Biological Pharmaceutical Co. Ltd., Beijing, China). Test sample DNA concentration was adjusted to 1 to 3 ng/μL. The PCR was conducted by an ABI7500 real-time PCR instrument (Applied Biosystems, Foster City, CA). PCR cycling parameters were shown as follows: 95°C denaturation for 5 minutes, 1 cycle; 95°C denaturation for 25 s, 64°C annealing for 20 seconds, and 72°C extension for 20 seconds, for a total of 15 cycles; 93°C denaturation for 25 seconds, 60°C annealing for 35 seconds, and 72°C extension for 20 seconds, for 31 cycles. The EGFR gene mutation status was considered positive when the mutation Ct values was <28 and the T790 M ΔCt value mutation was ≤7. Otherwise it was considered negative or below the detection limit of the kit. 2.3 Follow-up After starting treatment, CT imaging was conducted every 2 months. Response evaluation criteria in solid tumor (RECIST) version 1.1 was used to determine complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). CR + PR were used to determine the response rate. PFS was defined as the first day of starting EGFR-TKI treatment to disease progression, death, or last follow-up. OS was defined as the first day of starting EGFR-TKI therapy to death or last follow-up. All 49 patients were followed up by telephone and outpatient visit. Follow-up was censored on June 20, 2015. 2.4 Statistical analysis Normality of the distribution of continuous variables was determined using the Kolmogorov–Smirnov test. Continuous data were presented as mean ± SD or median (range or interquartile range, IQR), and analyzed using the student t test or the Wilcoxon test, as appropriate. Categorical data were presented as n (%) and analyzed using the Fisher exact test. The SPSS 17.0 software (IBM, Armonk, NY) was used. PFS and OS were analyzed using the Kaplan–Meier method and the log-rank test. Factors independently associated with survival were analyzed using the Cox multivariate regression analysis (backward). Two-sided P-values < .05 were considered statistically significant. 3 Results 3.1 Characteristics of the patients A total of 49 patients (28 males and 21 females), aged 39 to 78 years (median age: 64 years) met the inclusion criteria. Eighteen patients had a history of smoking. ECOG score was 0/1 in 38 patients and 2/3 in 11 patients. According to the American Joint Committee on Cancer tumor node metastasis (AJCC TNM) staging, 8th edition, nine patients were stage IIIB and 40 were stage IV. All 49 patients were tested positive for EGFR mutation before EGFR-TKI treatment; the 19-Del mutation was observed in 29 patients and the 21-L858R mutation in 20 patients; the T790 M mutation was not observed before treatments. EGFR-TKIs given as second-line therapy and above accounted for 65.3% (32/49) of the patients. Initial EGFR-TKI treatment median PFS was 11.3 months. 3.2 T790 M resistance mutation status analysis after EGFR-TKI treatment DNA test results from serum samples of the 49 secondary-resistant patients showed that the EGFR gene mutation rate was 46.9% (23/49). The T790 M mutation rate was 30.6% (15/49, Supplemental Figure 1), the 19-Del mutation rate was 6.1% (3/49, Supplemental Figure 2), the 21-L858R mutation rate was 10.2% (5/49, Supplemental Figure 3), including three cases of 21-L858R + T790 M mutation (Supplemental Figure 4), and the EGFR mutation-negative rate was 53.1% (26/34, Supplemental Figure 5). As shown in Table 1, the T790 M mutation did not correlate with advanced age, gender, smoking status, clinical stage, ECOG score, initial EGFR mutation, and EGFR-TKI drugs, while the association between EGFR-TKI resistance and progression was significant (P = .009). Table 1 The correlation between T790 M mutation and basic features in NSCLC patients. 3.3 T790 M mutation relationship status and the efficacy of EGFR-TKI Among the 49 patients with secondary EGFR-TKIs resistance, the response rate was 55.1% (27/49). There was no difference in the frequency of the T790 M mutation between patients with CR+PR compared with patients with SD+PD (P = .647). 3.4 Relationship between PFS and T790 M mutation Among the patients with acquired resistance to EGFR-TKI, the median PFS was 8.2 months and the median OS was 15.4 months. The patients with the T790 M mutation had a median PFS of 9.6 months and median OS of 17.6 months, compared to 6.8 and 12.7 months in patients without the mutation (P = .018 and P = .027) (Figs. 1 and 2). The impact of gender, age, smoking status, clinical stage, ECOG score of PFS, and OS was not statistically significant. The Cox multivariate analysis showed that T790 M mutations were independent factors for PFS and OS (P = .032 and P = .008) (Table 2). Figure 1 Comparison of progression-free survival between patients with and without the T790 M mutation. Figure 2 Comparison of overall survival between patients with and without the T790 M mutation. Table 2 PFS and OS of 49 advanced NSCLC patients with acquired resistance to EGFR-TKI by multivariate cox model analysis. 4 Discussion As we know, circulating tumor DNA (ctDNA) is released by tumor cells, and could effectively monitor tumor volume and cellular turnover increase 1. ctDNA is highly degraded (∼166 bp) and firstly released to the circulation, plasma and then other body fluids, such as cerebrospinal fluid, saliva and urine 2 to 5. However, because of the kidney barrier filtration, urine trans-renal DNA (tr-DNA) is in the form of more shorter fragments (less than 100 bp) than plasma ctDNA, which bring out technical difficulties to detect the highly fragmented and low abundant tumor-specific DNA in urine 6. Therefore, so far clinical liquid biopsies are most commonly applied to plasma-derived ctDNA. In this study, the T790 M mutation in exon 20 of the EGFR gene in serum free DNA of patients with advanced NSCLC and EGFR-TKI resistance was detected using the ARMS method after the occurrence of drug resistance. Ma et al[23]suggested that EGFR mutation status tested by ARMS in plasma cannot replace a tumor tissue biopsy though positive EGFR mutation results detected in plasma are fairly reliable. However tumor biopsy also has its limitations due to heterogeneity of the tumor cells.[24] Therefore the ARMS method, as a non-invasive method to detect EGFR mutations may have its clinical advantages. The relationship of the T790 M mutation with the efficacy and prognosis of targeted therapy was investigated. The results showed that the T790 M mutation rate was 30.6% (15/49) in 49 patients with secondary drug resistance, which was lower than that reported by Kuang et al,[25] who reported 45.2% (19/42) positivity in peripheral blood specimens from patients receiving EGFR-TKIs and using the Scorpions Amplification Refractory Mutation System (Scorpions ARMS) whole-genome amplification approach. This difference might be due to the different detection methods and to the study population. Because of the high cost of this method, it is not widely used in the clinical setting in China. The serum-free DNA in blood samples might be contaminated with mutated and wild type DNA from cancer cells, as well as with wild type DNA from normal tissues. In addition, there is genetic heterogeneity between tumor cells. Therefore, how to obtain reliable cfDNA for the detection of T790 M mutation in blood samples remains to be studied further. The T790 M mutation is the main factor for EGFR-TKI resistance in patients with EGFR sensitive mutation. Therefore, it is clinically significant to detect and investigate the T790 M mutation in patients with NSCLC. Bai et al[26] detected the EGFR gene in serum and tissue samples, and they showed that serum free DNA could replace tissue DNA for EGFR gene mutation detection. With the continuous improvement of the T790 M mutation detection methods, the sensitivity and specificity of detection were greatly improved.[27] The ARMS method has the advantages of convenient operation and good repeatability, and has been applied in the clinical setting.[21] Therefore, in this study, we detected the T790 M mutation in patients with EGFR-TKI resistance using ARMS. The results of the present study indicated that the T790 M mutation was not associated with age, gender, smoking, clinical stage, ECOG score, initial EGFR mutation types, and EGFR-TKI types in patients with advanced NSCLC, but was associated with the prognosis after EGFR-TKI resistance (P = .032). Oxnard et al[10] reported that EGFR-sensitive patient plasma showed continuous relationship between T790 M mutation and radiological tumor changes within 16 weeks before radiographic progression of the disease, suggesting that the dynamic monitoring of T790 M gene status during EGFR-TKI treatment may better predict drug resistance than the RECIST evaluation, which could facilitate individualized treatment of patients with NSCLC, and also help to identify the T790 M mutation TKI resistance induction mechanism and heterogeneity.[28] The present study showed that the PFS and OS of patients with the T790 M mutation were higher than those of patients without the T790 M mutation. The Cox multivariate analysis showed that the T790 M mutation was independently associated with PFS and OS. These results were consistent with the studies by Kuiper et al[29] and Li et al.[30] A retrospective study in Korea observed that the patients with the T790 M mutation had a better prognosis than those without.[16] Uramoto et al[13] found that the 5-year survival rate was 86.7% in patients with the T790 M mutation and 13.3% in those without. Therefore, in patients resistant to EGFR-TKI, the T790 M mutation demonstrates a relatively good prognosis. This could be due to the presence of the T790 M indicating an inert tumor, the inactivation of the other mechanisms of resistance (K-ras gene mutation, C-met gene amplification, BRAF gene mutation, BIM deletion polymorphism, and so on), or to a better EGFR-TKI efficacy.[10] A recent study showed a TKI effectiveness rate of 53% and disease control rate of 82% in T790 M mutation-positive patients, with a clear association between the T790 M mutation in EGFR-TKI therapy and three generations.[12] The limitations of this study include small sample size, absence of dynamic and continuous detection of T790 M mutation, and subsequent clinical monitoring of blood to determine TKI resistance. In addition, the number of patients with the T790 M mutation was small, preventing subgroup analyses. Finally, tumor tissue was not available for most patients and no agreement between the tumor and the circulating tumor DNA could be made. Thereby, larger sample pool will be collected for further study, following clinical monitoring will be processed, and further sampling, sequencing and analysis would be performed. In summary, the T790 M mutation detection has tremendous potential clinical value in patients with lung cancer. The T790 M mutation can facilitate timely assessment of treatment efficacy, patient prognosis, and individualized treatment plans. Author contributions Conceptualization: Yuli Wang, Yuan Wei, Ping Gong. Data curation: Yuli Wang, Xiaoping Ma. Formal analysis: Yuli Wang, Yuan Wei, Xiaoping Ma, Xinyu Ma. Funding acquisition: Xiaoping Ma. Investigation: Yuan Wei, Xinyu Ma, Ping Gong. Methodology: Yuli Wang, Xinyu Ma. Project administration: Ping Gong. Resources: Yuan Wei. Software: Xiaoping Ma, Xinyu Ma. Supervision: Ping Gong. Validation: Xinyu Ma. Writing – original draft: Yuli Wang. Writing – review & editing: Yuan Wei, Xiaoping Ma, Xinyu Ma, Ping Gong. Supplementary Material Supplemental Digital Content Abbreviations: ABI7500 = Applied Biosystems7500, ADx-EGFR = Automatic Data Exchange System-EGFR, AJCC = American Joint Committee on Cancer, ARMS = Amplified Refractory Mutation System, ARQ197 = mesenchymal epithelial transition factor inhibitor, BIM = BCL2-like 11, cfDNA = cell-free DNA, CR = complete response, Ct = cyclic threshold, DNA = deoxyribonucleic acid, ECOG = Eastern Cooperative Oncology Group, EGFR = epidermal growth factor receptor, EGFR-TKI = epidermal growth factor receptor tyrosine kinase inhibitor, IBM = International Business Machines Corporation, IQR = interquartile range, NCCN = National Comprehensive Cancer Network, NSCLC = non-small cell lung cancer, OS = overall survival, PCR = polymerase chain reaction, PD = progressive disease, PFS = progression-free survival, PIK3CA = phosphatidylino-sitol 3-kinases, PR = partial response, RECIST = response evaluation criteria in solid tumor, SD = stable disease, SPSS = Statistical Package for Social Sciences, T790 M = 20 exon 790 site mutation, TNM = tumor node metastasis. YW and YW contributed equally to this work. This study was supported by National Natural Science Foundation of China (grant number 81560381). The authors declare no conflicts of interest. Supplemental Digital Content is available for this article. ==== Refs References [1] Zhang SW Chen WQ Zheng RS An analysis of cancer mortality in China, 2003–2007 . Chin Cancer 2012 ;21 :171–8 . [2] Ahn MJ Park SY Kim WK A single nucleotide polymorphism in the phospholipase D1 gene is associated with risk of non-small cell lung cancer . Int J Biomed Sci 2012 ;8 :121–8 .23675264 [3] Benaiges D Zanui M Chillaron JJ Two cases of pituitary metastases as initial presentation form of small cell lung cancer . Invest Clin 2012 ;53 :402–7 .23513490 [4] Shrestha S Joshi P Gefitinib monotherapy in advanced non-small-cell lung cancer: a retrospective analysis . JNMA J Nepal Med Assoc 2012 ;52 :66–71 .23478732 [5] Jemal A Bray F Center MM Global cancer statistics . CA Cancer J Clin 2011 ;61 :69–90 .21296855 [6] Cataldo VD Gibbons DL Perez-Soler R Treatment of non-small-cell lung cancer with erlotinib or gefitinib . N Engl J Med 2011 ;364 :947–55 .21388312 [7] Rosell R Moran T Queralt C Screening for epidermal growth factor receptor mutations in lung cancer . N Engl J Med 2009 ;361 :958–67 .19692684 [8] Hui Z Xinjie Y Xinyong Z Advanced lung EGFR mutation status monitoring and analysis . Chin J Cancer Prev Control 2014 ;21 :1517–20 . [9] North J Jing L Ping G Non-small cell lung cancer patients with EGFR gene mutation serum circulating DNA detection . Chin J Cancer Prev 2014 ;21 :29–33 . [10] Oxnard GR Arcila ME Sima CS Acquired resistance to EGFR tyrosine kinase inhibitors in EGFR-mutant lung cancer: distinct natural history of patients with tumors harboring the T790 M mutation . Clin Cancer Res 2011 ;17 :1616–22 .21135146 [11] Rosell R Bivona TG Karachaliou N Genetics and biomarkers in personalisation of lung cancer treatment . Lancet 2013 ;382 :720–31 .23972815 [12] Thress KS Paweletz CP Felip E Acquired EGFR C797S mutation mediates resistance to AZD9291 in non-small cell lung cancer harboring EGFR T790 M . Nat Med 2015 ;21 :560–2 .25939061 [13] Uramoto H Yamada T Yano S Prognostic value of acquired resistance-related molecules in Japanese patients with NSCLC treated with an EGFR-TKI . Anticancer Res 2012 ;32 :3785–90 .22993320 [14] Majem M Remon J Tumor heterogeneity: evolution through space and time in EGFR mutant non small cell lung cancer patients . Transl Lung Cancer Res 2013 ;2 :226–37 .25806236 [15] Jackman D Pao W Riely GJ Clinical definition of acquired resistance to epidermal growth factor receptor tyrosine kinase inhibitors in non-small-cell lung cancer . J Clin Oncol 2010 ;28 :357–60 .19949011 [16] Ji W Choi CM Rho JK Mechanisms of acquired resistance to EGFR-tyrosine kinase inhibitor in Korean patients with lung cancer . BMC Cancer 2013 ;13 :606.24369725 [17] Azuma K Hirashima T Yamamoto N Phase II study of erlotinib plus tivantinib (ARQ 197) in patients with locally advanced or metastatic EGFR mutation-positive non-small-cell lung cancer just after progression on EGFR-TKI, gefitinib or erlotinib . ESMO Open 2016 ;1 :e000063.27843623 [18] Wu YL Zhou C Hu CP Afatinib versus cisplatin plus gemcitabine for first-line treatment of Asian patients with advanced non-small-cell lung cancer harbouring EGFR mutations (LUX-Lung 6): an open-label, randomised phase 3 trial . Lancet Oncol 2014 ;15 :213–22 .24439929 [19] Cross DA Ashton SE Ghiorghiu S AZD9291, an irreversible EGFR TKI, overcomes T790M-mediated resistance to EGFR inhibitors in lung cancer . Cancer Discov 2014 ;4 :1046–61 .24893891 [20] Haber DA Velculescu VE Blood-based analyses of cancer: circulating tumor cells and circulating tumor DNA . Cancer Discov 2014 ;4 :650–61 .24801577 [21] Jiang T Ren S Zhou C Role of circulating-tumor DNA analysis in non-small cell lung cancer . Lung Cancer 2015 ;90 :128–34 .26415994 [22] Newman AM Bratman SV To J An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage . Nat Med 2014 ;20 :548–54 .24705333 [23] Ma M Shi C Qian J Comparison of plasma and tissue samples in epidermal growth factor receptor mutation by ARMS in advanced non-small cell lung cancer . Gene 2016 ;591 :58–64 .27370697 [24] Hata A Katakami N Yoshioka H Spatiotemporal T790 M heterogeneity in individual patients with EGFR-mutant non-small-cell lung cancer after acquired resistance to EGFR-TKI . J Thorac Oncol 2015 ;10 :1553–9 .26309190 [25] Kuang Y Rogers A Yeap BY Noninvasive detection of EGFR T790 M in gefitinib or erlotinib resistant non-small cell lung cancer . Clin Cancer Res 2009 ;15 :2630–6 .19351754 [26] Bai H Mao L Wang HS Epidermal growth factor receptor mutations in plasma DNA samples predict tumor response in Chinese patients with stages IIIB to IV non-small-cell lung cancer . J Clin Oncol 2009 ;27 :2653–9 .19414683 [27] Ma C Wei S Song Y T790 M and acquired resistance of EGFR TKI: a literature review of clinical reports . J Thorac Dis 2011 ;3 :10–8 .22263058 [28] Oxnard GR Paweletz CP Kuang Y Noninvasive detection of response and resistance in EGFR-mutant lung cancer using quantitative next-generation genotyping of cell-free plasma DNA . Clin Cancer Res 2014 ;20 :1698–705 .24429876 [29] Kuiper JL Heideman DA Thunnissen E Incidence of T790 M mutation in (sequential) rebiopsies in EGFR-mutated NSCLC-patients . Lung Cancer 2014 ;85 :19–24 .24768581 [30] Li W Ren S Li J T790 M mutation is associated with better efficacy of treatment beyond progression with EGFR-TKI in advanced NSCLC patients . Lung Cancer 2014 ;84 :295–300 .24685306