The Value of Next-Generation Sequencing for Treatment in Non-Small Cell Lung Cancer Patients: The Observational, Real-World Evidence in China

Background Great success has been made in the targeting therapy of advanced non-small cell lung cancer (NSCLC). Nowadays, next generation sequencing (NGS) is acquirable and affordable in developed area of China. Using this feasible and accurate method of detecting therapeutic genes would help to select optimal treatments to extend patients survival. Here, we identified somatic mutations by NGS and analyzed the value for treatment of NSCLC in a real-world clinical setting. Methods NGS was carried out on biopsy samples obtained from 66 advanced unresectable NSCLC patients who had not received any treatment. 23 patients received liquid biopsy after failure of first-line targeted treatment. The mutation profiling as well as associations between mutations and clinicopathological characters was analyzed. The study also assessed the values of NGS for choosing treatment options and predicting prognosis in NSCLC patients. Results 152 somatic mutations were identified in 45 (68.18%) tissue samples. The most frequently mutated genes were EGFR (42.42%), TP53 (31.82%) and KRAS (15.15%). Specifically, the most frequent EGFR (42.42%), EGFR (42.42%), p = 0.046). In addition, in the smoking group, patients with EGFR (42.42%), p = 0.046). In addition, in the smoking group, patients with EGFR (42.42%), EGFR (42.42%), p = 0.046). In addition, in the smoking group, patients with Conclusions The observational study from real-world demonstrated that using NGS in routine clinical detection may be useful in guiding the therapy decisions and benefit more Chinese NSCLC patients.


Introduction
Non-small cell lung cancer (NSCLC) contributes to over 80% of all lung cancer cases and it is one of the leading causes of cancer-related deaths worldwide [1]. Although the application of surgery, chemotherapy, radiation and targeted therapy was beneficial for some patients, most of patients died of relapse, metastasis or even adverse effects by treatment [2,3]. Epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKI) have been used to treat NSCLC since 2003 with great success in patients with EGFR mutations [4][5][6], which made physicians pay more attention to "individualized treatment".
Unfortunately, acquired resistance to EGFR-TKI treatment occurs inevitably. It is well known that NSCLC patients with the EGFR exon 19 deletion or L858R mutation show initial responses to first-generation of TKI, such as gefitinib and erlotinib. A er about 9-14 months treatment, more than half of the patients relapsed [7]. Possible mechanisms for the acquired resistance may be the appearance of second-site hot spot mutations [8]. e most common and famous alteration is EGFR T790M mutation. e efficiency of the third-generation EGFR-TKI osimertinib in treating EGFR T790M mutated patients has been demonstrated in several studies [9][10][11]. However, more targeted and rare mutations and biomarkers are needed to use in clinical practice for prolonging survival of NSCLC patients.
With the wide use of next-generation sequencing (NGS), the genetic basis of various diseases, especially human cancers, have been disclosed. NGS is a high-throughput method that can detect numerous genetic variations, such as single nucleotide variants, insertions and deletions, copy number variations, and gene fusions over larger genomic regions. It is also noted for its high sensitivity and specificity [12]. Consequently, NGS may be a good tool for guiding the treatment of NSCLC. e aim of this study was to identify somatic mutations by NGS and analyze the value for treatment in NSCLC patients in a real-world clinical setting.

Study Population.
is study included 66 histologically confirmed NSCLC cases diagnosed in Affiliated Kunshan Hospital of Jiangsu University between January 2010 and September 2017. ey were all advanced unresectable NSCLC patients. Tissues were obtained by transbronchoscopic lung biopsy, lymph node biopsy, thoracentesis or lumbar puncture. e personal data of each participant about clinical characteristics and survival information was collected from clinical record or family contact. e overall survival (OS) was defined as time from the data of diagnosis to the data of death or last visit. e progression-free survival (PFS) was calculated from the time of diagnosis to the time of progression, relapse, death, or the last follow-up. is prospective observational study was reviewed by our institutional review board and written informed consent was provided by each patient.

Tissue DNA and Plasma
Cell-Free DNA Extraction. Tissues were stored at −80°C until DNA extraction. Genomic DNA was extracted using a QIAamp DNA FFPE tissue kit (Qiagen, Valencia, CA, USA) according to manufacturer's instructions. Almost 20 ml fluid sample was collected with EDTA (0.5 mol/L, pH 8.0) to reach a final concentration of 10 mmol/L EDTA. e mixture was then centrifuged at 268 g to separate the supernatant. Finally, the final supernatants of these three liquids were stored at −80°C. A peripheral blood sample of 10 ml was collected in an EDTA-containing tube for 23 patients relapsed a er first-line targeted therapy. A er isolated from the peripheral blood cells, plasma was subsequently frozen at −80°C. Circulating cell-free DNA (cfDNA) was purified from 4 to 5 ml of plasma using the QIAamp Circulating Nucleic Acid kit (Qiagen, Hilden, Germany).

NGS Library Preparation and Sequencing
. NGS was carried out on all DNA samples. Sequencing libraries were prepared using a KAPA Hyper Prep kit (KAPA Biosystems, Boston, MA) with an optimized manufacturer's protocol for different samples types. In brief, 250 ng −1 μg genomic DNA fragments or 10-250 ng cf DNA underwent end-repairing. A-tailing and ligation with indexed adapters sequentially, followed by size selection of genomic DNA using Agencourt AMPure XP beads (Beckman Coulter, Pasadena, CA). Finally, libraries were amplified by PCR and purified for target enrichment. Hybridization-based target enrichment was performed using GeneseeqOne™ 416-gene panel (Nanjing Geneseeq Technology Inc., Nanjing, China). Library fragment size was determined by an Agilent Technologies (Palo Alto, CA) 2100 Bioanalyzer. e target-enriched library was then sequenced on HiSeq4000 NGS platforms (Illumina) [13].

NGS Data Analysis.
e raw data were aligned to Human Genome version 19 (hg19) using Torrent Suite so ware (version 3.6.2; ermo Fisher Scientific, Inc.). e coverage analysis was performed using the Coverage Analysis plugin (version 3.6; ermo Fisher Scientific, Inc.). Cases for which the quality was <20% and/or the average base coverage was <500X reads and/or the frequency was <10% were considered noninformative. Mutations were detected using the Variant Caller plugin (version 3.6; ermo Fisher Scientific, Inc.). Each mutation was verified using the Integrative Genome Viewer (IGV) from the Broad Institute (www.broadinstitute.org) [14].
e NGS testing process took about five to seven days.

Statistical Analysis.
Comparisons between groups were performed using Chi-squared test. e Kaplan-Meier method and log-rank tests were used to compare survival curves. For all the analyses, a two-sided value of <0.05 was defined as significance. Statistical analyses were performed using SPSS version 16.0 (SPSS, Chicago, IL, USA). Table 1, 37 (56.06%) enrolled patients were male, and 29 (43.94%) were female. e median age was 63 years old (range 33-87 years). Of all the 66 patients, 55 (83.33%) were diagnosed with lung adenocarcinoma, and 11 (16.67%) were squamous cell carcinoma. More than half (57.58%) of the patients had a history of smoking and 26 patients (39.39%) were never smokers. Another 2 cases lost the information of smoking history.

Patient Characteristics. As shown in
ere were 23 (34.85%) patients have the family history of cancer.
We further analyzed the information of 23 adenocarcinoma patients received the first generation of EGFR-TKI treatment a er they were told EGFRmutated. ese patients all received liquid biopsy a er they relapsed or progressed. We found that 11 (47.83%) patients had mutation of EGFR T790M in exon 20, which was known as the most common mutation associated with acquired drug resistance. Other rare mutations included EGFR G719A (1.52%) and MET amplification (1.52%).

Clinicopathological Features and Genetic Mutations.
We analyzed the association between genetic mutations and clinicopathologic variables of the enrolled patients. As shown in Table 1, EGFR exon 19 deletion mutation was associated with sex (푝 = 0.045), smoking history (푝 = 0.041) and family history (푝 = 0.016) and pathological type (푝 = 0.032). In another word, EGFR 19-del mutation had an increased frequency in female, nonsmokers, who had family history and adenocarcinoma patients.However, EGFR L858R, PTEN, ALK, or KRAS were not associated with any clinicopathologic variables.

Utilization of NGS for Treatment Option and Prognosis
Prediction in NSCLC Patients. In this study, 55 (83.33%) patients received targeted therapy such as EGFR-TKI or crizotinib a er the detection of EGFR mutations or ALK gene rearrangements, and the others received chemotherapy. e median OS for target therapy patients were 24 months and the first-line PFS were 10.5 months. As shown in Figure 2, patients with EGFR exon 19 deletion had longer OS than those with exon 21 L858R mutation (37.0 months versus 19.0 months, 푝 = 0.01). Subsequently, we further conducted stratified analysis, finding that among the adenocarcinoma cases, patients with EGFR 19 deletion mutation have longer OS than the wide-type (36.0 months versus 19.0 months, 푝 = 0.046, Figure 3). In addition, in the smoking group, patients with EGFR 19 deletion mutation tended to have longer OS (38.0 months versus 16.5 months, 푝 < 0.01, Figure 4).
A er the first step of NGS detection, most of the EGFR mutated patients received first-generation EGFR-TKIs. e media PFS of these patients was 11 months. A er disease progression, 23 patients received liquid biopsy. 11 (47.83%) patients had mutation of T790M in EGFR exon 20. All these 11 patients received osimertinib as the second-line targeted therapy. It was also shown that, T790M positive patients have longer PFS than the negative ones in the second-line treatment (15 months versus 9.5 months, 푝 = 0.032, Figure 5).

Discussion
In the present study, we demonstrated the real-world data of mutations detected by NGS in Chinese NSCLC patients. Most of the patients benefited from the targeted therapy. We identified that NGS can be applied to guide treatment and predict prognosis in NSCLC patients. patients who carried driver mutations showed longer OS and PFS. Several studies have reported the use of NGS to detect the oncogenic driver mutations could guide therapy decisions and thus prolong the survival of NSCLC patients [16,17]. We believe that our real-world data may add new evidence for it. In our study, 23 EGFR mutated patients received second-line liquid biopsy a er the failure of first-generation of Since its introduction in 2007, NGS technology has already made extraordinary advances, making the detection of genetic alterations to guide therapy to be more feasible. And as NGS becomes faster and less expensive, it's sure to be used more frequently and with greater benefit for Chinese NSCLC patients [15]. In our real-world data, 83.33% patients chose NGS-guided targeted therapy in first-line and/or second-line therapy. ese patients, especially adenocarcinoma T790M-positive [11]. We supposed that repeated NGS in relapse and metastasis NSCLC patients was essential and it also could guide the following treatment. It was reported that, EGFR mutation rate varies in different countries, and the mutation rate was much higher in Asian people [18,19]. Similar to the study conducted in southern China [20], we showed that EGFR was the most frequent mutation in NSCLC patients. e majority of EGFR mutations were exon 19 deletion (60.71%), L858R in exon 21 (46.43%). EGFR double mutation is not rare in Asia [21]. Analyzed for the 23 patients received second-line liquid biopsy, double mutation rate was 47.83% (11 cases), of which exon 19 deletion combined with T790M account for 81.82% (9/11). On the other hand, several studies conducted for Chinese patients reported L858R combined with T790M occurred more o en [20,22]. e different incidences of EGFR mutation pattern may cause by limited sample sizes and ethnic differences.
Besides EGFR, the mutation rate of other oncogenic driver mutations in our study such as PTEN, KRAS, ALK and BRAF were consistent with previous studies [23,24]. Although driver genes mutations were reported to be mutually exclusive in NSCLC [25], we found one case carried both EGFR L858R mutation and ALK rearrangement. e patient received firstline therapy of gefitinib, and the PFS was 8 months. With one cycle of crizotinib as second-line therapy, his tumor progressed again. e OS was 15 months which was shorter than the average. is result was similar to other study. Yang et al. showed the median PFS of patients with concurrent EGFR/ALK mutations treated with EGFR-TKI ranged from 5.0 to 11.2 months, relatively lower than patients harboring only EGFR mutation [26]. e use of NGS in this present study revealed that females, non-smokers, family history of cancer and adenocarcinoma patients had greater EGFR 19 deletion mutation rate, consistent with most previous reports [23,27]. e difference of mutation rate between males and females may be caused by the higher smoking rate in males. It was also found from the stratified analysis that among the smokers, patients with EGFR 19 deletion mutation tended to have longer OS. We think it may be due to that the mutated patients were more likely to choose the targeted therapies. Patients with EGFR exon 19 deletion were reported to have longer survival than those with exon 21 mutation [28,29], though the detailed mechanism remain unknown. Our results added new evidences for this conclusion.

Conclusions
In summary, we applied NGS in NSCLC tumor tissue at the moment of diagnosis and in liquid biopsy at the moment of progression in a subset of EGFR mutant patients. It was demonstrated that using NGS in routine clinical detection allows selecting a better treatment for patients and even improving PFS and OS. Further larger-scale studies focus on the prognostic value of NGS in NSCLC and other tumors are needed to confirm the advantages of this tool. TKI. 11 patients (47.83%) were found EGFR T790M mutation. All 11 patients received osimertinib therapy. Not surprising that T790M positive patients have longer PFS than the negative ones in the second-line treatment. e median OS of these patients was 36 months, which was much longer than the median 24 months of all patients. e result was consistent with the randomized phase III AURA trial that osimertinib showed a significant survival benefit in patients with advanced NSCLC who progressed to prior EGFR-TKIs and were