Circulating plasma lncRNAs as novel markers of EGFR mutation status and monitors of epidermal growth factor receptor‐tyrosine kinase inhibitor therapy

Background Epidermal growth factor receptor (EGFR) gene mutations predict tumor response to EGFR tyrosine kinase inhibitors (EGFR‐TKIs) in non‐small cell lung cancer (NSCLC). However, even patients with EGFR‐sensitive mutations in NSCLC have limited efficacy with EGFR‐TKI. Studies have shown that long noncoding RNA (lncRNA) is related to diagnosis and prognosis with NSCLC. This study aimed to explore the correlation between lncRNA in NSCLC patients with EGFR mutation status and EGFR‐TKI efficacy. Methods The amplification‐refractory mutation system method was used to test the EGFR mutation status in tumor tissues and pleural effusions of NSCLC patients. Three EGFR‐mutant patients and three EGFR wild‐type patients were selected. Differential lncRNA was performed on the pleural effusions of the two selected groups of patients using Clariom D Human chip technology. Five lncRNAs significantly associated with EGFR mutation status were screened by FC value and GO analysis, and then evaluated by real‐time quantitative polymerase chain reaction in NSCLC patients' pleural effusions. Three were further analyzed in NSCLC patients' plasma. Results There were 61 significant differences in lncRNA between EGFR mutation‐positive and wild‐type patients. Among them, SCARNA7, MALAT1, NONHSAT017369, NONHSAT051892, and FTH1P2 were significantly associated with EGFR mutation status. SCARNA7, MALAT1, and NONHSAT017369 showed consistent results with plasma in pleural effusions compared to EGFR wild‐type, all upregulated in the EGFR mutation group. Conclusion This study shows that lncRNAs can be used not only as potential biomarkers for predicting the mutation status of EGFR and the efficacy of EGFR‐TKI, but also for monitoring the efficacy of EGFR‐TKI.


Introduction
As one of the cancers with the highest morbidity, lung cancer ranks first among all tumors worldwide in terms of mortality. 1 Non-small cell lung cancer (NSCLC) accounts for approximately 80.0%-85.0% of all lung cancer cases. 2,3 Most NSCLC patients, especially those receiving traditional chemotherapy, are at an advanced stage at the time of diagnosis and have a poorer prognosis. According to the National for establishing personalized treatment for patients. However, tumor tissue specimens cannot be obtained from some patients. Moreover, rebiopsy during EGFR-TKI therapy in order to continuously monitor EGFR mutation status and analyze EGFR-TKI resistance remains a major challenge. In recent years, liquid biopsy has gained extensive attention as a noninvasive detection method. Its unique advantage is that it overcomes tumor heterogeneity to some extent, which can help clinicians adjust the treatment strategy for NSCLC patients in a timely manner. In fact, circulating tumor DNA (ctDNA) demonstrates high accuracy in the analysis of EGFR mutation status, but its sensitivity is relatively low. 3,[5][6][7] The sensitivity of ctDNA ranges between 70.0% and 75.0%, and a possible explanation is that the ability to detect gene mutations in plasma ctDNA is related to tumor load; for patients with a smaller tumor load, the false-negative rate will increase significantly. Another possible reason is that somatic mutations accumulated by aging cells restrict the utility of ctDNA. 7,8 At the same time, a number of studies have shown that the efficacy of EGFR-TKI in patients with EGFR-mutant NSCLC is only about 70%, which may be related to the activation of the bypass signaling pathway to cause primary resistance. [9][10][11] However, there is currently no clinically relevant marker to predict the efficacy of EGFR-TKI. Therefore, it is necessary to define a group of noninvasive, convenient, economical markers as a supplement for predicting EGFR mutation status and monitoring the efficacy of EGFR-TKI therapy in patients with NSCLC.
In previous studies, more attention was given to the relationship of the protein coding gene mutation with the tumor. With the continuous understanding of lncRNA, it is believed that the differential expression and mutation of lncRNA play an important role in the malignant proliferation of tumors. Over the past decade, lncRNA has been found to be stably expressed in human blood, and circulating lncRNAs can serve as novel molecular biomarkers for diseases and cancers. [12][13][14] Blood biomarkers have several significant advantages: they are easily obtained in a minimally invasive manner, can be measured repeatedly, and their expression levels can be compared longitudinally. However, the association between circulating plasma lncRNAs and EGFR mutation status and the role of lncRNAs in monitoring the efficacy of EGFR-TKI therapy have yet to be systematically studied. Our objective was to identify a group of circulating plasma lncRNAs that can predict EGFR mutation status and explore the potential of this group of lncRNAs to monitor the efficacy of EGFR-TKI therapy.
Compared to other chip technologies, Clariom D solutions offer unprecedented comprehensive transcriptome coverage, providing deeper, broader, and more sophisticated transcriptome information detection. Samples for different sources and properties, including whole blood and pleural effusion samples, are the tools of choice for transformation studies. A single experiment can accurately detect the difference in expression of coding RNA and noncoding RNA at both gene and exon levels, as well as enriching the comprehensive variable shear information. Plasma and pleural effusion samples were used in this study. A number of studies have shown that pleural effusions are more sensitive to detecting EGFR mutation status than blood samples. To improve the specificity and sensitivity of this study, we used Clariom D Human Chip technology to detect pleural effusions with different EGFR mutation status.

Patient enrollment
This study was approved by the Ethics Committee of the Fifth Medical Center of the Chinese PLA General Hospital (No. 2012-11-171). All experiments were carried out in accordance with the Declaration of Helsinki. From December 2016 to April 2018, patients with NSCLC (n = 155) were recruited from the Department of Pulmonary Oncology of the Fifth Medical Center of the Chinese PLA General Hospital. Informed consent was obtained from all patients. Inclusion criteria for patients were as follows: (i) NSCLC diagnosis by histopathology or cytology (stage IIIb, IV); (ii) age ≥ 18 years old; (iii) Eastern Cooperative Oncology Group Performance Status, (ECOG PS) <2 points; (iv) no prior chemotherapy, radiotherapy, targeted therapy, and other oncology treatment; and (v) EGFR gene detection by histology. Patients with diseases of the heart, liver, kidney, and other important organs were excluded. The study flow chart is shown in Figure S1.

Plasma (pleural effusion) collection
A total of 5 mL of peripheral blood was collected between 0700 and 0800 hours (15 mL of pleural effusion was collected during thoracentesis). Blood samples were dynamically collected from 36 of the 72 patients with EGFR mutations at one, three, and five months of EGFR-TKI treatment. In addition, blood samples were collected from 16 patients at the time of progressive disease (PD). Plasma (pleural effusion) was separated by centrifugation at 4500 g at 4 C for 10 minutes within 30 minutes after collection and immediately stored at −80 C for further analysis. QIAzol lysis reagent (Cat. No. 217184; QIAGEN Inc., Valencia, CA, USA) was used to separate total RNA containing lncRNA from 200 μL of plasma (pleural effusion).

Analysis of lncRNA in pleural effusions
Clariom D Human chip (Guangdong Geneway Decoding Bio-Tech Co. Ltd., Guangzhou, China) analysis was performed on the pleural effusions from three EGFR-mutant patients and three EGFR wild-type patients with NSCLC to initially screen for lncRNAs that were differentially expressed in these patients.

qRT-PCR of selected lncRNAs
Reverse transcription was performed using the All-in-One cDNA Synthesis SuperMix kit (Bimake, Houston, USA) according to the manufacturer's instructions. Moreover, 2× SYBR Green qPCR Master Mix kit (QIAGEN) was used for the qRT-PCR analysis of the expression levels of the selected lncRNAs in pleural effusions and plasma. PCR was performed in a total reaction volume of 23 μL, including 10 μL 2× SYBR Green qPCR Master Mix, 5 μL cDNA template, 2 μL forward primer (5 mM), 2 μL reverse primer (5 mM), 0.4 μL ROX Reference Dye, and 3.6 μL deionized water. qRT-PCR was performed in triplicate using the PCR program of the Applied Biosystems 7500 PCR System (Applied Biosystems, Thermo Fisher Scientific Corp. 5, USA). Denaturation was performed at 95.0 C for 10 minutes, followed by 45 cycles of denaturation at 95.0 C for 15 seconds, 60.0 C for 30 seconds, and 72.0 C for 30 seconds. The original Ct data was normalized to the expression level of 18S rRNA using the 2 −Δ(ΔCt) method.

Statistical analysis
Statistical analysis was performed using SPSS 17.0 (IBM Corp., Armonk, NY, USA). Student's t-test was used to evaluate differences in the expression of the chosen lncRNAs in pleural effusion and plasma from the NSCLC patients. The specificity, sensitivity, and area under the curve (AUC) for the lncRNA levels were determined using ROC analysis. Using the binary outcome of EGFR(+) and EGFR(−) samples as dependent variables, a logistic regression model was established using the stepwise model selection method. All statistical tests were two-tailed, and a P-value < 0.05 was considered statistically significant.

Patient characteristics
From December 2016 to April 2018, a total of 490 patients were included with advanced NSCLC confirmed by tissue or cytopathology with a median age of 58.2 (range 37-82) years and a male-to-female ratio of 1:0.74. According to the inclusion criteria, we selected 77 patients with EGFR mutation status confirmed in the tumor tissue by histological examination from 218 patients with NSCLC with pleural effusion, and those diagnosed with tumor cells in pleural effusion for the study (35 EGFR mutants, 42 EGFR wild-type). The clinical features of EGFR wild-type patients are summarized in Table 1. A number of studies have shown that the specificity and sensitivity of ctDNA with EGFR mutation status detected in pleural effusion are close to, or even higher than, the tumor tissue. 15,16 Among 372 patients with EGFR mutation status confirmed by tumor tissue, 142 patients who did not receive any lung cancer treatment were enrolled in the study (72 EGFR mutants, 70 EGFR wild-type), and their clinical features are summarized in Table 1. No significant differences in clinical features were observed between the EGFR-mutant group and the EGFR wild-type group, except for their smoking status; there were fewer smokers in the EGFRmutant group (P < 0.05).

Efficacy analysis
Of the 72 patients with advanced NSCLC EGFR mutations, 68 received a first-generation EGFR-TKI treatment.

Gene chip analysis of candidate lncRNAs associated with EGFR mutation
Pleural effusion samples of three EGFR-mutant and three EGFR wild-type patients were selected for Clariom D Human microarray analysis to detect the differences in the expression of lncRNAs (Fig 1a). Compared with the EGFR wild-type group, 48 lncRNAs were upregulated and 13 lncRNAs downregulated in the EGFR-mutant group (fold change ≥2 or ≤0.5, P < 0.05; Table 2). Five of these lncRNAs were selected for further validation. Among the five lncRNAs, SCARNA7, MALAT1, NONHSAT017369, and NONHSAT051892 were significantly upregulated and FTH1P2 was significantly downregulated in EGFR-mutant patients.

qRT-PCR validation of candidate lncRNAs
Based on the results of the pleural effusion chip, realtime quantitative polymerase chain reaction (qRT-PCR) was used to validate the expression of the five lncRNAs in the pleural effusions in 35 EGFR-mutant and 42 EGFR wild-type patients. The 18S ribosomal RNA (18S rRNA) was used as an internal control to normalize the original cycle quantification (Ct) values. The Ct of 18S rRNA was subtracted from the Ct value of the target lncRNA (ΔCt) and converted to 2 −ΔCt to evaluate the changes in the lncRNA expression between the groups. The fold change was calculated using the 2 −Δ(ΔCt) method, where show the relative expression levels of the five lncRNAs in the EGFR-mutant group and the EGFR wild-type group. The expression levels of both NONHSAT051892 and FTH1P2 were not significantly different between the EGFR-mutant and wild-type groups. Therefore, SCARNA7, MALAT1, and NONHSAT017369 were selected for further study.
Circulating plasma lncRNA for predicting EGFR mutation To further investigate the circulating plasma lncRNA in EGFR-mutant patients with NSCLC, qRT-PCR analysis of the plasma of EGFR-mutant and EGFR wild-type patients was performed using 18S rRNA as an internal control. We observed the same trend as we did in the pleural effusions,   Fig 2); no statistically significant differences were observed in the expression of the remaining two lncRNAs by EGFR mutation.
Diagnostic analysis of EGFR mutation status by detection of circulating plasma lncRNA in patients with NSCLC The potential of circulating plasma SCARNA7, MALAT1, and NONHSAT017369 as diagnostic markers for EGFR mutation status in NSCLC patients was evaluated using ROC curve analysis. Circulating SCARNA7 was significantly different between EGFR-mutant patients and EGFR wild-type patients, with a ROC-AUC of 0.76 (95% confidence interval [CI]: 0.68-0.84), sensitivity of 80.8%, and specificity of 67.1% (Fig 3a). No significant differences were found between the EGFR exon 19 deletion group and the EGFR p.L858R point mutation group. Circulating MALAT1 was also significantly different between EGFR-mutant patients and EGFR wild-type patients, with a ROC-AUC of 0.89 (95% CI: 0.84-0.95), sensitivity of 79.2%, and specificity of 85.7% (Fig 3b). MALAT1 was significantly upregulated in the EGFR exon 19 deletion group compared to the EGFR p.L858R point mutation group, with a ROC-AUC of 0.74 (95% CI: 0.64-0.85), sensitivity of 70.0%, and specificity of 69.0% (Fig 3c). NONHSAT017369 could only distinguish between the EGFR-mutant group and the EGFR wild-type group. The ROC-AUC was 0.76 (95.0% CI: 0.68-0.84), the sensitivity was 79.7%, and the specificity 63.6% (Fig 3d). To further evaluate the diagnostic potential of the combinations of SCARNA7, MALAT1, and NON-HSAT017369, a logistic regression analysis was conducted to combine the ROC curve analyses of these three lncRNAs. The AUC in the 3-lncRNA group was 0.920 (95% CI: 0.860-0.981) (Fig 3e). Furthermore, the characteristics of 2-lncRNA groups were examined (Fig 3f-h). Two lncRNAs were randomly selected from the three lncRNAs, and the AUC of all the combinations was lower than that in the 3-lncRNA group. The positive predictive value (PPV) and negative predictive value (NPV) of each model were also calculated to distinguish between EGFR-mutant patients and EGFR wild-type patients. As shown in Table 3, the PPV of the three individual lncRNAs was higher than 68.5%, and the NPV was higher than 74.5%. The combination of the three lncRNAs could distinguish between the EGFR-mutant patients and the EGFR wild-type patients, and the PPV and NPV were both higher than 80%. Overall, the results indicate that the plasma levels of the three lncRNAs have a high diagnostic value for distinguishing between EGFR-mutant and EGFR wild-type patients.

Relationship between the initial level of circulating plasma lncRNAs and progression-free survival in patients treated with EGFR-TKIs
Of the 72 EGFR-mutant patients with NSCLC, 68 received EGFR-TKI therapy, and 59 developed resistance to EGFR-TKIs and displayed disease progression. Kaplan-Meier analysis was performed for progression-free survival (PFS) based on clinicopathological factors, as well as the initial (before EGFR-TKI) plasma levels of SCARNA7, MALAT1, and NONHSAT017369. As 55 of the 59 patients had an Eastern Cooperative Oncology Group (ECOG) score of one, and only four patients had an ECOG score of 0, we did not perform Kaplan-Meier analysis based on ECOG score. The plasma expression levels of the three lncRNAs were stratified by the median value. The PFS of patients with high lncRNA expression levels (≥the median) was compared to that of patients with low lncRNA expression levels (<the median). As shown in Figure 4, the median PFS of the 59 patients was 10 months (range: 1-24 months). The initial expression of SCARNA7 and NONHSAT017369 had no significant correlation with PFS (P = 0.67 and P = 0.855, respectively). However, the pretreatment expression of MALAT1 was significantly correlated with PFS (P = 0.020).  In patients with high and low expression of MALAT1, the median PFS were 11.6 and 8.2 months, respectively (hazard ratio: 0.431; 95% CI: 0.236-0.787). As indicated in Table 4, multifactor analysis showed smoking history was also significantly correlated with PFS (hazard ratio: 2.33; 95% CI, 1.28-4.27; P = 0.006). These results indicated that both MALAT1 and smoking history were important predictors of the efficacy of EGFR-TKIs in patients with NSCLC.
Circulating plasma lncRNAs as a potential marker for monitoring the efficacy of EGFR-TKI therapy To determine whether the plasma levels of SCARNA7, MALAT1, and NONHSAT017369 had predictive value for the efficacy of EGFR-TKI therapy, qRT-PCR was used to compare the plasma levels of the three lncRNAs in 36 patients before and after EGFR-TKI treatment. A total of 78 plasma samples were collected from the 36 patients before and after EGFR-TKI treatment. Initial blood samples were collected two weeks (AE two days) before the treatment and then every two months (AE one week). At the same time, chest computed tomography (CT) was performed to evaluate the response to treatment. Of the 36 patients, 23 achieved PR, nine had SD, and four had PD. The plasma lncRNA levels before and after treatment are shown in Figure 5, and they were analyzed by Wilcoxon matched pairs signed rank sum test. Of the 32 patients achieving PR and SD after EGFR-TKI treatment, 27 exhibited decreased plasma MALAT1 levels (P < 0.05), and 23 displayed decreased SCARNA7 levels (P < 0.05) after treatment. However, no similar trend was observed in patients with PD. There was no statistically significant difference in the plasma levels of NONHSAT017369 after EGFR- TKI treatment (P > 0.05). Overall, the study on the three lncRNAs showed that the changes in the plasma levels of SCARNA7 and MALAT1 from before to after EGFR-TKI treatment had a satisfactory predictive value regarding the efficacy of EGFR-TKI therapy.

Discussion
The assessment of the EGFR gene mutation status in lung cancer tissues has important predictive value for the efficacy of EGFR-TKI. However, lung cancer tissue in patients with advanced NSCLC is sometimes difficult to obtain. Although clinically circulating ctDNA detection of EGFR mutation status has been recommended by the NCCN and Chinese Society of Clinical Oncology (CSCO) guidelines as complementary means to obtain tumor tissue. But unfortunately, different detection methods show that the sensitivity is only 50.0%-81.8%. Even though next-generation sequencing has improved the sensitivity of detection of ctDNA EGFR mutation status, the detection efficiency and quality cannot be guaranteed due to its high cost and nonstandard operation. 17,18 In addition, several studies have shown that the efficiencies of EGFR-TKI in patients with EGFR-mutant NSCLC can only reach about 70%. Therefore, predictive markers are urgently needed to predict EGFR mutation status and EGFR-TKI efficacy in NSCLC patients. Several studies have explored the difference between cancer tissue and paracancerous tissues. lncRNA is used to diagnose different cancers, indicating the potential of lncRNA as a diagnostic biomarker. In a study of lung squamous cell carcinoma, Luo et al. used a bioinformatics tool to analyze the difference between lncRNA in 450 patients with primary lung squamous cell carcinoma and normal tissues, and showed that AC064853.2, AC090044.2, CTD-3099C6.9, and KB-1836B5.4 can be used as biomarkers for the diagnosis of lung squamous cell carcinoma. 19 Wang et al. analyzed the difference between EGFR 19 exon mutation and EGFR wild-type lncRNA in advanced NSCLC patients by lncRNA microarray and qRT-PCR. Finally, five lncRNAs were found to play a key role in EGFR 19 exon mutation. 20 However, few studies have focused on the expression of lncRNA in blood samples associated with EGFR mutation status in patients with advanced NSCLC. Recent studies have shown that lncRNAs are present in stable forms in serum, plasma, and other body fluids, independent of endogenous RNase, making them noninvasive suitable biomarkers. In this study, we systematically evaluated the ability of circulating lncRNAs as potential markers for predicting EGFR mutation status, EGFR-TKI efficacy, and monitoring the efficacy of EGFR-TKI in patients with CI, confidence interval. In the univariate analysis the progression-free survival was analyzed using Kaplan-Meier method, and hazard ratios were estimated using log rank test. In multivariate analysis, hazard ratios were estimated using the multivariate Cox proportional hazards regression model. EGFR-positive NSCLC. To the best of our knowledge, this study is the first to describe a prospective analysis of circulating lncRNA as a diagnostic or prognostic biomarker for EGFR-TKI in the treatment of EGFR-positive NSCLC. In recent years, several studies have shown that the specificity and sensitivity of pleural effusion detection of ctDNA EGFR mutation status are close to, or even higher than, that of tumor tissue. Lee et al. used pleural effusion supernatants from 50 patients with lung adenocarcinoma to test for EGFR mutation status. That study found that the results of pleural effusion were 100% matched with tumor tissue, and three patients with positive EGFR mutations were found among 19 EGFR wildtype patients. The reason for this may be that pleural effusion overcomes tumor heterogeneity to some extent. 16 The present study systematically analyzed the differential expression levels of lncRNAs in the pleural effusions of three EGFR-mutant patients and three EGFR wild-type patients with NSCLC, detected the relative expression levels in pleural effusions and plasma through qRT-PCR, and comprehensively evaluated the potential of SCARNA7, MALAT1, and NONHSAT017369 as prospective markers for EGFR mutation status in patients with NSCLC. As observed in a study comparing the clinical efficacy of icotinib and gefitinib in previously treated patients with advanced NSCLC, the therapeutic efficacy of TKIs is better in patients harboring the EGFR exon 19 deletion than in patients harboring the p.L858R mutation. Therefore, the identification of potential markers delineating EGFR mutant subtypes (mainly the EGFR exon 19 deletion and p.L858R point mutation) is also crucial in clinical studies. In this study, three circulating plasma lncRNAs (SCARNA7, MALAT1, and NONHSAT017369) were significantly upregulated in EGFRmutant patients as compared to EGFR wild-type patients. Their expression was able to distinguish between EGFR-mutant and wild-type NSCLC patients with high specificity and sensitivity. In the 3-lncRNA combination, the PPV and NPV both exceeded 80%. In addition, because EGFR exon 19 deletion has a better therapeutic response to EGFR-TKI than EGFR21 p. L858R point mutation, further analysis of EGFR mutant subtypes shows that plasma MALAT1 is significantly associated with it. Compared with EGFR21 p.L858R point mutation NSCLC patients, MALAT1 was significantly upregulated in plasma of patients with EGFR 19 exon deletion NSCLC.
In recent years, in the NSCLC study, several potential markers have been reported to predict the efficacy of EGFR-TKI and to monitor the efficacy of EGFR-TKI. However, most studies are based on tissue or cell line levels, and there are relatively few studies related to serum or plasma levels. Salmon et al. used MALDI-TOF-MS to test the serum of 13 patients with pre-NSCLC treated with gefitinib or erlotinib, and an EGFR-TKI efficacy model with 8 characteristic peaks was established. Then, the groups of 67 patients treated with gefitinib, and 96 treated with erlotinib, and the control group without EGFR-TKI treatment were tested to predict whether EGFR-TKI was effective or had a poor efficacy. The model predicted a median survival of 207 days and 92 days for the good and poor patients in the gefitinib group, respectively. The median survival of the good and poor patients in the erlotinib-treated group was 306 and 107 days, respectively, while the control group could not be used to predict the results. 21 However, there are few studies on the prediction and monitoring of EGFR-TKI efficacy at the plasma lncRNA level. To explore the relevance of lncRNA to EGFR-TKI in patients with EGFR mutation-positive NSCLC, we performed a Kaplan-Meier analysis of PFS in 59 patients and found that patients with high plasma MALAT1 (≥median) expression after treatment with EGFR-TKI had longer PFS indicating that plasma MALAT1 has the potential to predict the efficacy of EGFR mutation-positive EGFR-TKI. Dynamic plasma analysis of patients with EGFR-TKI response to PR or SD showed a decrease in plasma SCARNA7 and MALAT1 levels, and a lower trend in PR patients than in SD patients. Plasma SCARNA7 and MALAT1 returned to baseline levels as the disease progressed.
To date, no report has described the correlation between SCARNA7 and NONHSAT017369 and lung cancer or EGFR signaling. Studies on SCARNA7 expression levels in other cancers have shown inconsistent results. 22,23 Most of the previous reports have confirmed that SCARNA7 is upregulated in cancer tissues, although one report indicated that SCARNA7 was downregulated in chronic lymphocytic leukemia. 24 In our study, the plasma SCARNA7 level was upregulated in EGFR-mutant patients with NSCLC compared to those with wild-type EGFR. There are few reports on NONHSAT017369, and regulatory network analysis found that NONHSAT017369 can be targeted to IL6R. IL6R is involved in the JAK-signal transducer and activator of transcription (STAT) signaling pathway, which is commonly known as the signaling pathway downstream of EGFR. 25,26 Therefore, NONHSAT017369 may affect EGFR status through STAT activation. MALAT1 is highly expressed in most tumors; siRNA-induced knockdown reduces the ability of tumors to invade and metastasize. 7 MALAT1 is more highly expressed in metastatic NSCLC than in nonmetastatic NSCLC, making it the first lncRNA confirmed to be able to trigger tumor metastasis and clearly determine the prognosis of patients. 27 Chen et al. found that MALAT1 plays an important role in tumor progression and the development of resistance to chemotherapeutic drugs. 28 Yang et al. inhibited the activity of gefitinib-resistant NSCLC and induced apoptosis by regulating PPI to downregulate MALAT1 and inhibit STAT3 phosphorylation. 29 Schmidt et al. found that STAT3 may play an important role in gefitinib resistance, and MALAT1 may be a potential target for reversing resistance to EGFR-TKIs. 30,31 In this study, MALAT1 levels were upregulated in EGFR-mutant patients with NSCLC and could distinguish between EGFR mutant subtypes. Moreover, serum MALAT1 decreased during EGFR-TKI treatment, but gradually increased during the development of drug resistance, which was also consistent with the results of previous studies. This indicated that MALAT1 can serve not only as a diagnostic marker for EGFR-mutant lung cancer, but also as an important potential marker for monitoring the efficacy of EGFR-TKI therapy.
This study had several limitations. First, the samples were collected from a single lung cancer center; further external validation of our results requires sample collection from multiple centers. Second, the number of dynamic plasma samples in patients with NSCLC treated with EGFR-TKIs was relatively small; more samples are required in future work to ensure the reliability of the predictive model. Third, this study failed to identify the expression of lncRNAs in the plasma and tissues of the same individual; future studies require matching plasma and tissue samples from the same individual to verify whether the expression levels of lncRNAs in these samples are consistent. Fourth, cytological experiments on EGFR signal transduction pathways are required to elucidate the specific functions of these lncRNAs. In conclusion, this exploratory study established a group of plasma lncRNAs for predicting EGFR mutation status, and this model could serve as a new noninvasive biomarker for the diagnosis of EGFR-mutant patients with NSCLC and an important indicator for predicting the efficacy of EGFR-TKI therapy.