Next Generation Sequencing Technology in Lung Cancer Diagnosis

Simple Summary Lung cancer is still one of the most commonly diagnosed and deadliest cancers in the world. Its diagnosis at an early stage is highly necessary and will improve the standard of care of this disease. The aim of this article is to review the importance and applications of next generation sequencing in lung cancer diagnosis. As observed in many studies, next generation sequencing has been proven as a very helpful tool in the early detection of different types of cancers, including lung cancer, and has been used in the clinic, mainly due to its many advantages, such as low cost, speed, efficacy, low quantity usage of biological samples, and diversity. Abstract Lung cancer is still one of the most commonly diagnosed cancers, and one of the deadliest. The high death rate is mainly due to the late stage of diagnosis and low response rate to therapy. Previous and ongoing research studies have tried to discover new reliable and useful cbiomarkers for the diagnosis and prognosis of lung cancer. Next generation sequencing has become an essential tool in cancer diagnosis, prognosis, and evaluation of the treatment response. This article aims to review the leading research and clinical applications in lung cancer diagnosis using next generation sequencing. In this scope, we identified the most relevant articles that present the successful use of next generation sequencing in identifying biomarkers for early diagnosis correlated to lung cancer diagnosis and treatment. This technique can be used to evaluate a high number of biomarkers in a short period of time and from small biological samples, which makes NGS the preferred technique to develop clinical tests for personalized medicine using liquid biopsy, the new trend in oncology.


Lung Cancer
Lung cancer remains one of the most common cancers diagnosed in 2020, and one of the deadliest cancer types. In Europe, the lung cancer incidence rate is 97.6 (men) and 38.3 (women), with a mortality rate of 81.7 (men) and 29 (women), respectively [1]. In Romania, the incidence is lower in women (28.5) and higher in men (105.3) than the European incidence rate. The same trend is also observed in mortality, 24.8 (women) and 95.6 in men [1]. An increased number of deaths due to lung cancer is mainly due to late-stage diagnosis, mostly because this cancer shows no symptoms in its early stages.
There are several risk factors associated with lung cancer, such as smoking, air pollution, radon exposure, occupational exposure to different chemicals, heredity susceptibility,

Next Generation Sequencing
Next generation sequencing (NGS) is a comprehensive technology used for sequence (DNA) and gene expression (RNA species) analysis [16][17][18]. The NGS technique was developed to overcome the Sanger sequencing limitation, but it evolved into being used in all areas of genomic research, starting with DNA, RNA, miRNA, ChIP and methylation sequencing [19][20][21]. As with any technique, NGS has multiple advantages that have made it an essential tool in all areas of research and in the clinic [22]. However, even after over 15 years of development, this technique has some disadvantages, such as the need for powerful bioinformatics tools and specialized personnel for both experimental and data analysis [20]. Some advantages and disadvantages of NGS are presented in Table 1.
The data provided by NGS have proven valuable and reliable for both research and in the clinic to improve the diagnosis, prognosis, and treatment of several diseases [23][24][25][26], and are widely used in the oncology field [27,28]. In lung cancer, this technique has been used for early diagnosis biomarker identification, targeted treatment decisions, and identification of causative mutations [29][30][31][32][33][34].

NGS in Lung Cancer Diagnosis
Lung cancer diagnosis is challenging in the early stages because patients do not present any symptoms, or symptoms are shared with other pulmonary diseases. In addition, classic techniques for lung cancer diagnosis have many false-negative results due to different reasons, such as quality and quantity of the samples or sensitivity of the test [35]. Here, NGS can be beneficial due to its high sensitivity and specificity, using low amounts of sample. Additionally, NGS can determine an increased number of alterations simultaneously from the same quantity of sample. Therefore, NGS has been applied with success in the identification of lung cancer-specific mutations in paraffin-embedded tissue samples, with a higher rate than standard PCR testing [36,37]. Recently, studies have shown that NGS can effectively be used to identify specific lung cancer mutations in circulating tumor DNA, in a liquid biopsy sample [38][39][40][41]. The main applications of NGS in the clinic are related to genomics, transcriptomics and epigenomics. When using whole genome, whole exome or targeted DNA analysis, specific information on point mutations, copy number alterations, small indels or structural variance alterations can be identified. RNA seq analysis can provide information related to gene fusions, alternative splicing, differential expression or RNA editing, while Bisuphite seq or ChIP seq are used for the identification of the methylation profile, histone modification or transcription factor binding alterations.
These important advantages demonstrated by NGS in the evaluation of the alterations related to lung cancer diagnosis have created a new opportunity for the development of commercial kits and assays specific to lung cancer. One such kit is the NextDaySeq-Lung panel, developed by Beijing ACCB Biotech (Beijing, China), with primers for the amplification of EGFR exon 18, 19, 20, 21, KRAS exon 2, 3, PIK3CA exon 9, 20, and BRAF exon 11, 15. The mutations in the KRAS gene can predict the efficiency of EGFRtyrosine kinase inhibitors [42]. Recently, it was observed that most patients that developed resistance to TKIs have different EGFR mutations [43,44]. Mutations in BRAF can be correlated to response to BRAF/MEK inhibitors in NSCLC patients [45][46][47], while PIK3CA mutations could render SCLC patients sensible to triciribine treatment [48]. In addition, there is a study that uses alpelisib, a PIK3CA inhibitor, for breast cancer PIK3CA mutated patients, who have shown better survival than that of other treatment [49], which could be implemented in lung cancer as well. The NextDaySeq-Lung panel has been used in several studies and has demonstrated better results than Sanger sequencing or qRT-PCR [50,51]. Other gene panels specific for lung cancer focus on fusion alterations, based on RNA sequencing, and evaluate translocations, chromosomal inversions or interstitial deletions. One such panel is the Ion Ampliseq RNA fusion lung cancer panel offered by ThermoFisher Scientific, Waltham, USA, which targets 70 known fusion transcripts of ALK, RET, ROS1, and NTRK. This panel has shown high sensitivity and good concordance with the typical methods used for fusion testing [52]. For fusion testing, RNA seq has proven to be more sensitive and is used in parallel with DNA seq for mutation evaluation [53,54]. In addition, NGS has successfully been used to identify lung cancer patients that had MET exon14 skipping alterations [55]. Some other NGS lung cancer panels are presented in Table 2. It is well known that cancer is considered a genetic disorder in which somatic mutations accumulate and give cancer cells the ability to over proliferate and avoid apoptosis [56,57]. Lung cancer is one of the cancers that exhibit a high degree of mutation burden and a high number of driven mutations [9]. Consequently, NGS is extremely useful, due to its many advantages, and the development of different NGS panels is implemented in the clinical setting. In lung cancer diagnosis, NGS is employed mainly in evaluating the gene alteration in key genes involved in the development of lung cancer. These genes are EGFR, BRAF, KRAS, HER2, ROS, ALK, PIK3CA, NTRK, RET and MET [58]. One example is using CGH NGS-based assay for assessing 51 FFPE samples of adenocarcinoma to evaluate its efficiency compared to standard mutation testing. The authors observed that 58% of wild-type patients presented alterations in one of these genes when using the NGS approach, making them suitable for targeted therapy [35]. In other studies, NGS was implemented for NSCLC diagnosis due to the small quantity of tissue samples, which is not suitable for traditional testing methods. Hagemman et al. successfully sequenced 209 samples of NSCLC using a 28 gene NGS panel and identified actionable mutations in 46% of the tested samples [36]. In the same line, Moskalev et al. used the 454 NGS system to evaluate EGFR and KRAS mutation in NSCLC samples with a low number of tumor cells. They were able to identify mutations with an allele frequency of 0.2-1.5%. When reevaluating 16 cases with low tumor cells that were wild type by Sanger, seven of them presented mutations in the EGFR gene at a frequency of 0.9-10% [37]. Another study compared an NGS panel, Sanger sequencing, and qRT-PCR in evaluating mutation in 138 NSCLC FFPE samples. The authors observed that NGS and qRT-PCR have a higher sensitivity than Sanger sequencing. NGS is better than qRT-PCR because it also provides information about the mutation sequence and allele frequency, and identifies mutations that are not in the hotspot area [50]. Liang et al. used a DNA methylation profile to develop a blood-based test for the early diagnosis of lung cancer. Their method presented a sensitivity of 75% for stage 1A and 85.7 for stage 1B lung cancer [59]. NGS has proven to be more sensitive and specific than FISH or IHC when analyzing fusion alterations in lung cancer, which are the main methods used for fusion detection. Lin et al. observed a positive rate of 92.7% for ALK rearrangement when using NGS, 82.4% for FISH and 94.5% for IHC, and a concordance of 87.3% of NGS results with IHC results. They also concluded that IHC fusion testing is better for screening, while NGS fusion testing is more accurate for predicting the clinical benefits of crizotinib treatment [60]. Another benefit of NGS is the fact that it also provides information on the exact fusion alteration, which is very important in evaluating the treatment and outcome of patients [61]. To overcome the problem of harvesting tissue samples from early-stage lung cancer sample patients, new challenges related to identifying novel non-invasive biomarkers are under investigation. One such example is the use of miRNA for the diagnosis of lung cancer. miRNA sequencing was used to identify specific miRNAs for adenocarcinoma and SCLC. Jin et al. were able to identify miR-181-5p, miR-30a-3p, miR-30e-3p and miR-361-5p as being specific for adenocarcinoma, and miR-10b-5p, miR-15b-5p and miR-320b for SCLC (small cell lung cancer) [62]. In addition, taking advantage of the many benefits of NGS, oncology researchers have developed liquid biopsy testing for lung cancer diagnosis [63]. Leighl et al. observed a very high concordance for NGS results from cfDNA and tissue DNA in untreated metastatic NSCLC [64]. The same was observed by Mack et al. when analyzing 8388 cases of NSCLC [65]. NGS testing was successfully recommended in lung cancer diagnosis by different expert panels [66] and oncology organizations [67]. Gray et al. performed a thorough survey of the relevant literature regarding liquid biopsy and observed that the advantages of NGS have helped to develop different assays using liquid biopsy samples for the early diagnosis, treatment selection, minimal disease detection, monitoring treatment efficacy and evaluation of tumor burden in lung cancer [63]. Sueoka-Aragane et al. observed that the analysis of ctDNA by NGS could be a promising tool for the evaluation of the efficacy of osimertinib in NSCLC with EGFR T790M mutation [68]. Table 3 presents studies correlated to the performance of NGS technology in the diagnosis and screening of lung cancer [69,70].  The research area using NGS for lung cancer diagnosis has been extensively developed and has made this technique valuable for different clinical trials on lung cancer. Data from cl inicaltrial.gov include 98 trials on lung cancer that use NGS (https://clinicaltrials.gov/ct2/r esults?cond=Lung+Cancer&term=next+generation+sequencing&cntry&state=&city=&dis t=, accessed on 25 July 2021). Some of these trials are already complete; others are recruiting or enrolling. There are clinical trials that evaluate the possibility of NGS to identify mutations in very small samples (NGS NCT02420405), or improve the personalized treatment (NCT02281214). Table 4 presents data on the clinical trials using NGS for lung cancer diagnosis.    In addition, recent studies have shown that by implementing NGS-based testing, clinics can reduce the cost required for evaluating biomarkers specifically for targeted treatments or agnostic therapy implementation. NGS-based testing can reduce total testing cost by EUR 30-1249 depending on how comprehensive the analysis is, when compared to RT-PCR technology [80].

Conclusions
NGS has successfully been used both in research and in the clinic, and has become one of the main tools in lung cancer diagnosis, showing better results that standard techniques used for lung cancer diagnosis, and being able to identify lung cancer-specific alteration in a variety of biological samples such as blood, plasma, fresh frozen or FFPE tissue, urine or other bodily fluids, even where the nucleic acid content is limited and where classic methods fail. In addition, the cost of NGS is lower than that of standard testing methods, which makes this technique appealing for the implementation of different agnostic therapies, targeted therapies and immune checkpoint inhibitor therapies. Its success was demonstrated in different clinical studies that were developed to obtain better methods for lung cancer diagnosis. The NGS technique has become the primary tool for investigating different types of samples and different subtypes of lung cancer, being implemented in mutation evaluation and fusion alteration identification, due to its great advantage over FISH and IHC, which are techniques that can have inconsistent results due to the expertise of the pathologist, and cannot be used on other types of samples, except tissue. As with any technique, NGS still has its limitations, mostly related to the amount of data obtained, and the need for a big data storage capacity and a good bioinformatics team.
Nevertheless, the advantages of NGS make it ideal to be used for evaluating a high number of biomarkers in a short period of time, from small biological samples, and at a low price. Therefore, NGS should be the preferred technique to develop clinical tests for personalized medicine using liquid biopsy, the new trend in oncology.

Acknowledgments:
The authors would also like to thank Sanogenetic (Voluntari, Romania) for their help with this manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.