In this study, CTC counting, CTC next-generation sequencing, and LC-MS untargeted metabolomics were combined to characterize the potential gene mutation and energy metabolism disturbance characteristics of lung cancer, to provide a better detection method for the early screening and diagnosis of lung cancer patients. We used CellCollector in vivo CTC capture technology to detect CTC in early lung cancer patients. The CTC detection rate was 62.5%. Compared with in vitro technology, it has a higher detection rate, which is consistent with previous research results[18, 21]. The high CTC detection rate provides convenience for CTC molecular typing and CTC next-generation sequencing.
Through CTC NGS, this study found that more than 50% of early lung cancer patients have 4 common mutated genes, namely NOTCH1, IGF2, EGFR and PTCH1. Aslo, 47.37% of patients have ARID1A mutations. EGFR has the highest mutation rate in NSCLC[22]. NOTCH1, a member of the PCG gene family, was first discovered in mouse B-cell lymphoma and is regarded as a co-oncogene of C-MYC, closely related to cell proliferation, differentiation and apoptosis[23]. Studies have shown that the stimulation of the Notch signalling pathway by high NOTCH1 expression can induce BM-1 to mediate the production of related intracellular signals to accelerate the transmission efficiency of lung cancer, thereby inducing the occurrence, development, metastasis and invasion of lung cancer[24]. In our study, 68.42% of patients had a mutation in NOTCH1. Huang et al.[25] found that homozygous ARID1A was deleted at the 5' end of the lung adenocarcinoma cell line, strongly suggesting that ARID1A is a tumor suppressor gene. Imielinska et al. reported exons and genome sequences of 183 cases of lung adenocarcinoma, and the results showed that mutations of the ARID1A gene existed in lung adenocarcinoma cells[26]. This study is the first to find IGF2 and RTCH1 mutations in peripheral blood CTC NGS of early lung cancer. Whether these mutations can be combined with NOTCH1, EGFR and ARID1A mutations as tumor markers in the diagnosis of early lung cancer merits further investigation.
Studies have found that CTC already exists in the early stages of cancer[27], and disturbances of metabolism are produced in the body[28], including disorders of glucose and lipid metabolism[13, 19, 29, 30], and the homeostasis of the microenvironment of the body is disrupted. Through metabolomics analysis, we found 100 different metabolites, which mainly occurred in lipid metabolism, polysaccharide synthesis and metabolism, amino acid metabolism and other pathways and were dominated by lipid metabolism, being especially enriched in choline metabolism and glycerophospholipid metabolic pathways. Chen et al. found that abnormal sphingolipid metabolism is the most important metabolic change in lung cancer patients[31]. A study on lung adenocarcinoma in female non-smokers found that abnormal lipid metabolism may play role in the development of lung cancer[12]. High-lipid molecules, including phospholipids (e.g., glycerophospholipids and sphingomyelin), and cholesterol are the main component of cell membranes and participate in cell signalling and cell proliferation. Lipid metabolism changes cause abnormal cell signals and lead to tumor formation[32, 33].
Tumor growth requires the uptake of a large amount of energy in the blood. The body ensures the normal energy metabolism of other organs by increasing the "raw materials" in the aerobic oxidation pathway, resulting in increased glucose metabolism and decreased fat metabolism[34]. The purpose of this study was to discover a combination of serum metabolite biomarkers for the early detection of non-small cell lung cancer. In our study, the most obvious differences in the screened metabolites can be divided into four categories, namely phosphatidylethanolamine (PE), phosphatidylcholine (PC), lysophosphatidylcholine (LysoPC), L-isoleucine and L-palmitoylcarnitine. A large number of metabolomic studies have been undertaken to identify robust biomarkers for lung cancer diagnosis using plasma, serum, or urine. However, we found remarkably few metabolomic studies that specifically attempted to detect early-stage lung cancer.
In our study, the concentration of LysoPC was reduced in stage I/II NSCLC,which is similar to previous research[35]. Another targeted metabolomics study found and verified that β-hydroxybutyric acid, LysoPC 20:3, PC ae C40:6, citric acid, and fumaric acid differed significantly between healthy controls and stage I/II NSCLC. Robust predictive models with AUC > 0.9 were developed and validated using these metabolites and other, easily measured clinical data for detecting different stages of NSCLC[36].It has been observed and reported that in mouse and human models, the plasma concentration of total LysoPC is usually inversely related to the risk of various types of cancer[37–39]. In our study, we found that another member of the phosphatidylcholine family, PC(16:0/22:6(4Z,7Z,10Z,13Z,16Z,19Z)), PC(18:0/20:4(8Z,11Z,14Z,17Z)) and PC(16:0/20:4(5Z,8Z,11Z,14Z)), also appears to play a role in both stage I and stage II NSCLC. A study reported that PC levels were dysregulated in early-stage NSCLC patients[40]. Decreased lipid membrane unsaturation levels were observed to protect tumor cells from free radicals or chemotherapeutics and promote invasion and infiltration[41]. Clearly, more detailed lipidomic studies need to be conducted to investigate the biological significance of these PC alterations.
Lysophosphatidylethanolamine (LPE) is a group of lipids that has been recently shown to be related to breast cancer[42]. In addition, PE (16:0/18:1) is associated with the stage and prognosis of pancreatic cancer and may be a potential diagnostic marker[43]. Yang et al. found 25 different lipid metabolites, including PE, between malignant pleural effusion (MPE) and benign pleural effusion (NPE), indicating that lipid metabolites may be used to partition MBE and BPE[44]. In our study, PE (14:0/22:6(4Z,7Z,10Z,13Z,16Z,19Z)), PE (16:0/22:5(7Z,10Z,13Z,16Z, 19Z)), and PE (14:0/20:4(5Z,8Z,11Z,14Z)) were upregulated and effectively distinguished the control group, with specificity and sensitivity close to 100% being observed. In addition, we found that the level of amino acids (L-isoleucine) was significantly increased in the lung cancer model group compared with that of the control group, indicating disorder in amino acid metabolism in the cancer model group. Then, cell lesions occurred, causing diseases. Maeda et al. reported 6 significantly different amino acid metabolites, with AUCs of 0.817 and 0.801 (on their validation sets), for diagnosing stage I and stage II lung cancer[45]. One Study showed that L-palmitoylcarnitine is significantly reduced in advanced lung cancer patients[46]. In another study, the level of palmitoylcarnitine was lower in the hepatocellular carcinoma group than in the cancer-free control group, and blood acylcarnitine levels may be influenced by hepatic fatty acid metabolism, in other words, decreased acylcarnitine levels may reflect the decreased production of acyl groups in the liver or other tissues. Indeed, palmitoylcarnitine and palmitic acid are associated with fatty acid metabolism, and this group displayed an impact factor of 0.030 based on metabolic pathway analysis[47]. The decrease in L-palmitoylcarnitine in our study may also be related to a disorder of lipid metabolism in patients with lung cancer.
This study combines genomics and metabolomics to conduct a comprehensive assessment of early lung cancer in the search for biomarkers to enable early lung cancer screening and diagnosis and to improve the diagnosis rate of early lung cancer. However, there are still some shortcomings to this research. First, due to the small sample size in this study, a large sample study is needed to further verify the reliability of our research results. Second, the differential metabolites screened by untargeted metabolomics were not further verified by targeted metabolomics. In the end, we only performed CTC gene mutation detection but did not sequence the tissue. Whether there is a relationship between the mutations carried by CTCs and the changes in metabolic substances, which affect the microenvironment of patients, and whether there is a connection with the occurrence and development of lung cancer warrants further study.