Clinical characteristics of non‐small cell lung cancer patients with EGFR mutations and ALK&ROS1 fusions

Abstract Objective To study the relationship between clinical characteristics and anaplastic lymphoma kinase (ALK) fusions, c‐ros oncogene 1, receptor tyrosine kinase (ROS1) gene fusions, and epidermic growth factor receptor (EGFR) mutations in non‐small cell lung cancer (NSCLC) patients to distinguish these different types. Methods Both ALK, ROS1 gene rearrangements and EGFR mutations testing were performed. The clinical characteristics and associated pulmonary abnormalities were investigated. Results Four hundred fifty‐three NSCLC patients were included for analysis. One hundred seventy (37.5%), 32 (7.1%), and 9 cases (2.0%) with EGFR mutations, ALK gene fusions, and ROS1 gene fusions were identified, respectively. The EGFR‐positive and ALK&ROS1‐positive were more common in female (χ 2 = 61.934, P < 0.001 and χ 2 = 28.152, P < 0.001), non‐smoking (χ 2 = 59.315, P < 0.001 and χ 2 = 11.080, P = 0.001), and adenocarcinoma (χ 2 = 44.864, P < 0.001 and χ 2 = 12.318, P = 0.002) patients; proportion of patients with emphysema was lower (χ 2 = 35.494, P < 0.001 and χ 2 = 15.770, P < 0.001) than the wild‐type patients. The results of logistic regression analysis indicated that female (adjusted odds ratio [OR] 1.834, 95% confidence interval [CI] 1.069–3.144, P = 0.028), non‐smoking (adjusted OR 2.504, 95% CI 1.456–4.306, P = 0.001), lung adenocarcinoma (adjusted OR 4.512, 95% CI 2.465–8.260, P < 0.001), stage III–IV (adjusted OR 2.232, 95% CI 1.066–4.676, P = 0.033), and no symptoms of emphysema (adjusted OR 2.139, 95% CI 1.221–3.747, P = 0.008) were independent variables associated with EGFR mutations. Young (adjusted OR 3.947, 95% CI 1.873–8.314, P < 0.001) and lung adenocarcinoma (adjusted OR 2.950, 95% CI 0.998–8.719, P = 0.050) were associated with ALK/ROS1 fusions. Conclusions EGFR mutations were more likely to occur in non‐smoking, stage III–IV, and female patients with lung adenocarcinoma, whereas ALK&ROS1 gene fusions were more likely to occur in young patients with lung adenocarcinoma. Emphysema was less common in patients with EGFR mutations.


MPHKSTP-20190102
Conclusions: EGFR mutations were more likely to occur in non-smoking, stage III-IV, and female patients with lung adenocarcinoma, whereas ALK&ROS1 gene fusions were more likely to occur in young patients with lung adenocarcinoma. Emphysema was less common in patients with EGFR mutations.
K E Y W O R D S ALK, clinical characteristics, EGFR, emphysema, ROS1

| INTRODUCTION
With the increasing morbidity and mortality, lung cancer has become the first malignant tumor. Lung cancer has the highest morbidity and mortality among men in China. [1][2][3][4] The morbidity of lung cancer among women in China ranks second and the mortality rate of lung cancer among women is the first. 5 Non-small cell lung cancer (NSCLC) accounts for more than 80% of all patients with lung cancer. 6,7 In recent years, personalized molecular targeted therapy as the core has become a research hotspot in the treatment of lung cancer. [8][9][10] Currently, the most commonly used treatment of NSCLC is molecular targeted drugs targeting at mutations of epidermic growth factor receptor (EGFR), also known as small-molecule tyrosine kinase inhibitors (TKIs), which have obvious clinical efficacy on NSCLC patients with EGFR sensitive mutations. 11 EGFR is a transmembrane tyrosine kinase receptor, expressed in a variety of epithelial tumors, and regulates tumor cell growth, invasion, transformation, angiogenesis, and metastasis by activating downstream signal transduction proteins. 12 The main mechanism of EGFR-TKI is to selectively bind ATP binding sites in the tyrosine kinase domain of EGFR in cells, block the phosphorization and activation of tyrosine itself in EGFR molecules through the Akt-MAPK pathway, and inhibit RAS/RAF/MAPK, PI3K-Akt, and other downstream signaling pathways and lead to apoptosis of tumor cells. 13 Echinoderm microtubule associated protein-like 4 (EML4) and anaplastic lymphoma kinase (ALK) fusion gene EML4-ALK and proto-oncogene protein tyrosine kinase ROS (ROS1) were found after EGFR mutations in NSCLC. [14][15][16] ALK and ROS1 rearrangements define important molecular subgroups of NSCLC. After discovery of ALK rearrangements in NSCLC, it was recognized that these confer sensitivity to ALK inhibition. 17 For NSCLC patients with ALK gene fusion and ROS1 gene fusion, targeted therapy with Crizotinib could achieve better efficacy. 18 Therefore, the detection of EGFR, ALK, and ROS1 (ALK&ROS1) gene mutations before targeted therapy is of great significance for the prediction of the efficacy of targeted therapy and the appropriate patient screening.
The mutation status of targeted therapy driver gene in NSCLC is closely related to its pathological classification. The differences between different pathological subtypes of lung cancer are of great significance for clinical treatment and prognosis of lung cancer patients. The relationship between the mutation status of targeted therapeutically driven genes and clinicopathology in NSCLC has not been consistent. The purpose of this study was to evaluate the clinical characteristics of NSCLC patients in order to distinguish ALK&ROS1 gene rearrangements, EGFR mutations, and non-ALK&ROS1/EGFR (no mutations and rearrangements), so as to distinguish these different types, to assist clinicians to assess the NSCLC patients with these genetic abnormalities.

| Detection of ALK and ROS1 gene fusions by RT-PCR
The fusion genes of ALK and ROS1 were analyzed by RT-PCR according to the manufacturer's protocol of AmoyDx ® ALK Gene Fusions and ROS1 Gene Fusions Detection Kit (Amoy Diagnostics, Xiamen, China) with the LightCycler 480 real-time PCR system. The detection range included the fusions of ALK gene with EML4, KIF5B, TFG, and KLC1 genes, and the fusions of ROS1 gene with SLC34A2, CD74, SDC4, EZR, TPM3, LRIG3, and GOPC genes. Reverse transcription reaction system: reverse transcriptase 0.5 μl, RNA template 6 μl (total RNA 0.5-5.0 μg), 42 C for 1 h, and 95 C for 5 min. PCR amplification: 1.5 μl ALK&ROS1 mixed enzyme was respectively taken to ALK cDNA and ROS1 cDNA of the samples to be tested, 5 μl was successively transferred to the eight-tube strip PCR reaction system, and negative and positive controls were set up. The detection instrument and circulating conditions were the same as EGFR mutation detection, and Ct values were also interpreted.

| Statistical analysis
All analysis was conducted using SPSS statistical software Version 21.0. Fisher's exact test and the Student's t-test were performed in this study. EGFR-positive group, ALK&ROS1-positive group, and non-ALK&ROS1/EGFR group, pairwise comparisons were performed. The relationship between EGFR, ALK, and ROS1 genes mutations and clinical characteristics, various types of mutations in the EGFR gene, and clinical characteristics were analyzed. Logistic regression analysis was applied to assess the variables independently associated with EGFR and ALK/ROS1 genes mutations. P < 0.05 is considered statistically significant. Compared with the EGFR-positive group, the significant differences in ALK&ROS1-positive group were younger (P < 0.001). There were no significant differences in gender, smoking history, histologic type, clinical stage, and computed tomography characteristics (lymphangitis, lymphadenopathy, emphysema, fibrosis, and pleural effusion).

| The relationship between various types of mutations in the EGFR gene and clinical characteristics
Patients with mutations at 2 or more locations of the EGFR gene and with T790M resistance mutation were excluded from this analysis. Among the exon 19 deletion, L858R, L861Q, G719X mutations, and exon 20 insertion in EGFR, there were no significant differences in age and gender. The NSCLC patients with exon 19 deletion (73.3%) and L858R (85.0%) most were non-smokers, whereas patients with L861Q (60.0%) most were smokers ( Table 4). But the sample size of patients with L861Q, G719X mutations, and exon 20 insertion in our study is relatively small, and this result cannot represent the actual situation and we need a large sample size to analyze this problem.

| DISCUSSION
The number of lung cancer patients and death cases in China has ranked first among all malignant tumors. 3,19,20 In the past decade, as the development of tumor molecular diagnosis and the continuous discovery of targeted drugs, the treatment of NSCLC has entered an era of individualized molecular targeted therapy. With the discovery of therapeutic targets EGFR, ALK, and ROS1 and the advent of corresponding targeted drugs, targeted therapy has become a very effective way to treat NSCLC clinically. 21 Studies have shown that NSCLC patients with EGFR sensitive mutations and ALK&ROS1 gene fusions can benefit from corresponding targeted therapy. 10,22,23 Therefore, detection of EGFR, ALK&ROS1 gene mutations have important clinical significance in screening patients suitable for targeted therapy. In the present study, the clinical characteristics of NSCLC with ALK&ROS1 gene rearrangement and EGFR mutations were investigated. Distinguishing the clinical characteristics of different molecular subtypes will be beneficial to the diagnosis and treatment of lung cancer. EGFR is a transmembrane receptor tyrosine kinase, and the activation or phosphorylation of this region is of great significance for the signaling of proliferation and growth of cancer cells. EGFR mutation mainly includes four types: deletion mutations in exon 19, point mutations in exon 21, point mutations in exon 18, and insertion mutations in exon 20, among which exon 19 deletion mutation and exon 21 L858R are the most common mutations sensitive to EGFR-TKI therapy. 24 In concordance with previous reports, EGFR mutations were mainly 19 exon deletion mutations and exon 21 L858R mutation in 453 NSCLC patients of this study. A number of researches have shown that the incidence of EGFR mutation is higher in women, non-smokers, and adenocarcinoma. 25,26 This study found that the incidence of EGFR mutation in women, adenocarcinoma, and nonsmokers is significantly higher than that in men, squamous cell carcinoma, and smokers, and the differences between these groups are statistically significant. Since Soda et al. 27 first discovered a new EML4-ALK gene fusion in NSCLC patients in 2007, many scholars have conducted studies on it. Regarding the mutation rate of EML4-ALK in NSCLC, different literatures reported slight differences. Domestic and foreign research data showed that the incidence of ALK gene fusion in NSCLC patients was 3%-7%. [28][29][30] The positive rate of ROS1 gene fusion in NSCLC was 1.0%-3.4%, 31 and the clinical characteristics of ALK and ROS1 gene fusion lung cancer were also very similar. The results of this study showed that the positive rate of ALK and ROS1 genes fusions was 9.1%, and the incidence of ALK and ROS1 gene fusions were relatively high in female patients and those less than 60 years old. ALK gene fusions were identified in 32 cases (33/453, 7.3%) and ROS1 gene fusions were identified in 9 cases (9/453, 2.0%). Our work also confirms the low incidence of the ALK&ROS1 fusion among unselected NSCLC patients.
In this study, compared with non-ALK&ROS1/EGFR mutations in NSCLC patients, patients with EGFR mutation had a lower incidence of pulmonary emphysema. And it reflected that most of NSCLC patients with EGFR mutation had a history of non-smoking. A plausible reason for this is that EGFR mutation status have a stronger association with non-emphysema status. Among the deletions in exon 19, L858R, L861Q, G719X, S768I mutations, and insertions in exon 20 of EGFR, there were no significant differences in age and gender. The NSCLC patients with deletions in exon 19 (74.7%) and L858R (86.4%) most were non-smokers, whereas patients with L861Q (62.5%) and G719X (66.7%) most were smokers. But the sample size of patients with L861Q, G719X, S768I mutations, and insertions in exon 20 in our study is relatively small, and this result cannot represent the actual situation and we need a large sample size to analyze this problem.
According to reports, the incidence of ALK gene rearrangement in NSCLC patients is about 3%-7%, 10 whereas the incidence of EGFR mutation is 40%-80% 32 ; the sample size of patients with ALK&ROS1 gene rearrangement is small in our study. This is one of the limitations in this study. The purpose of this study was to evaluate the clinical characteristics of NSCLC patients to distinguish between ALK&ROS1 gene rearrangement, EGFR mutation, and non-ALK&ROS1/EGFR mutation. These results may assist clinicians to assess the NSCLC patients with these genetic abnormalities.

| CONCLUSIONS
In conclusion, this study suggests that EGFR mutations were more likely to occur in non-smoking, stage III-IV, and female patients with lung adenocarcinoma, whereas ALK&ROS1 gene rearrangements were more likely to occur in young patients with lung adenocarcinoma. Emphysema was less common in patients with EGFR mutations.