Med Oncol (2014) 31:203 DOI 10.1007/s12032-014-0203-5
ORIGINAL PAPER
Prognostic significance of microRNA expression in completely resected lung adenocarcinoma and the associated response to erlotinib Guanzhong Yan • Ruyong Yao • Dongfang Tang Tong Qiu • Yi Shen • Wenjie Jiao • Nan Ge • Yunpeng Xuan • Yongjie Wang
•
Received: 9 August 2014 / Accepted: 22 August 2014 Ó Springer Science+Business Media New York 2014
Abstract The mechanism of action of oncogenic or tumor suppressor microRNAs is not well understood. We examined the microRNA expression profile in completely resected lung adenocarcinoma and examined the associated response to erlotinib. The lung adenocarcinoma tissue and adjacent normal lung parenchyma of 226 stage IIB and IIIA patients who underwent complete resection were obtained for two separate retrospective cohorts. In cohort 1 (119 patients; 80 with epidermal growth factor receptor (EGFR) mutations and 39 without), miRNA microarrays were used to identify EGFR-related miRNAs and their association with survival. In cohort 2 (107 patients with EGFR mutations), the miRNAs and their association with survival and response to erlotinib were analyzed by qRTPCR. Cox proportional hazards regression was used to evaluate the effect of treatment on survival. As a result, erlotinib is associated with a significant improvement in overall survival (P = 0.0075, cohort 1; P = 0.0372, cohort 2) and disease progression (P = 0.6929, cohort 1; P = 0.3347, cohort 2) in patients with reduced miRNA-21 expression. Additionally, miRNA-145 is strongly associated with overall survival (P = 0.0008, cohort 1; P = 0.0131, cohort 2) and progression-free survival (P = 0.0198, cohort 1; P = 0.0269, cohort 2). Understanding the response rate to erlotinib relative to miRNA21 (77.3 vs. 41.7 %, P \ 0.01) and miRNA-145 (74.1 vs. G. Yan D. Tang T. Qiu Y. Shen W. Jiao Y. Xuan Y. Wang (&) Department of Thoracic Surgery, The Affiliated Hospital of Qingdao University, 16 Jiangsu Road, Qingdao 266003, China e-mail: [email protected] R. Yao N. Ge Central Laboratory, The Affiliated Hospital of Qingdao University, Qingdao 266003, China
42.6 %, P \ 0.01) expression is critical. The miRNA expression profiles differed significantly between patients with and without EGFR mutations. In conclusion, lung adenocarcinoma patients with reduced miRNA-21 expression exhibit longer overall survival and a poor response rate to erlotinib. Increased miRNA-145 levels can predict overall survival, progression-free survival and excellent response rate to erlotinib. Keywords microRNA Erlotinib Lung adenocarcinoma EGFR mutation Survival Disease progression
Introduction Lung cancer is the leading cause of cancer-related deaths worldwide, and non-small cell lung cancer (NSCLC) is the predominant pathological type, accounting for approximately 80 % of all cases of lung cancer [1, 2]. Despite the recent advances in diagnosis, chemotherapy and targeted therapy, the overall survival rate of NSCLC patients remains low, and the recurrence rate is very high, including in patients diagnosed early. The poor prognosis is closely related to our relatively limited understanding of the biology, late presentation and the heterogeneity of lung cancer. Patients with lung adenocarcinoma exhibit better survival than patients with other pathological subtypes of lung cancer. However, most patients are not completely cured, and the 5-year overall survival rate varies from 70 % in stage IA to 25 % in stage IIIA. The most effective treatment for lung adenocarcinoma is complete surgical resection combined with chemotherapy, radiation therapy or targeted therapy [3, 4]. The epidermal growth factor receptor (EGFR) regulates tumor cell proliferation, invasion, migration and apoptosis
123
203 Page 2 of 9
Med Oncol (2014) 31:203
Methods
China. Written informed consent was obtained from all patients. Patient data for all specimens were anonymized according to the legal standard. This study was approved by the ethical committee of our hospital. We used the amplification refractory mutation system (ARMS) to test for EGFR mutations in stage IIB and IIIA lung adenocarcinoma from patients receiving complete surgical resection. Patients with the T790 M mutation, in which threonine at amino acid 790 is replaced by methionine, were excluded from the study. The eligibility criteria for this study included an age of 75 years or younger (it remains uncertain whether patients [75 years old will benefit from platinum-based therapy) and an ECOG performance status of 0–3. All patients selected in the study received preoperative computed tomography scans and had no significant cardiac dysfunction, active infection or neurologic disorders. The inclusion criteria also required the patients to have undergone complete preoperative or intraoperative staging as well as intraoperative mediastinal lymph node examination or biopsy of nodes. All nodes larger than 1.5 cm were resected. Patients with sub-lobectomy (wedge resection or segmental resection), mixed histological features, diffuse lobar or multifocal bronchi alveolar carcinoma were ineligible for this study. Patients who underwent preoperative chemotherapy and patients who had breast cancer, renal cell carcinoma, gastric cancer, melanoma or other cancer treatment within the previous 5 years were ineligible. This study collected 226 pairs of lung adenocarcinoma tissues and adjacent noncancerous lung parenchyma within 4 cm of tumors that were resected by lobectomy between February 2010 and September 2013 at the Department of Thoracic Surgery at the Affiliated Hospital, Qingdao University, China. The patients who underwent radical resection and met the standards described previously were divided into two separate cohorts. Cohort 1 consisted of 119 patients (66 were positive for EGFR mutations, and 53 were negative). The miRNA microarray data were available and were analyzed to search for miRNAs associated with EGFR mutation and survival. Table 1 lists the detailed eligibility and exclusion criteria. Cohort 2 consisted of 207 lung adenocarcinoma patients with EGFR mutations. Erlotinib was administered as firstline adjuvant chemotherapy at a dose of 150 mg per day orally until the computed tomographic scan revealed disease progression or until the onset of intolerable toxic effects. In this cohort, we used qRT-PCR assays to investigate the relationship between miRNA expression and the response to erlotinib. Furthermore, the correlation between miRNA expression and disease progression was examined.
Patients
Sample processing
The study was conducted in the Department of Thoracic Surgery, The Affiliated Hospital, Qingdao University,
Primary lung adenocarcinoma tissue and paired adjacent normal tissue meeting the standards described preciously
[5]. Targeted therapies associated with epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) improve progression-free survival and overall survival in lung adenocarcinoma patients who have undergone complete surgical resection. EGFR-TKIs represent an important targeted therapy because these inhibitors block the complex EGFR transduction pathway involving the downstream Ras/MAPK cascade and phosphatidyl inositol 3-kinase (PI3 K) pathway [6]. Although there has been progress in treating NSCLC with EGFR-TKIs (gefitinib or erlotinib), the relationship between EGFR mutation and targeted therapy has not been elucidated. In addition, additional effective targeted therapies are urgently needed. MicroRNAs are a class of small, non-coding, singlestranded, endogenous RNAs that regulate gene expression. MicroRNAs bind to target mRNAs in the 30 untranslated region (30 -UTR) and cause mRNA cleavage or translational repression [7, 8]. The complementarity between the seed sequence of an individual microRNA and the target miRNA determines the mRNA specificity. Several proteomic studies have revealed that every miRNA has a board range of targets. It has been estimated that nearly half of all genes are regulated by miRNAs, and each miRNA can target several hundred transcripts, making miRNAs one of the largest families of gene regulators. Previous studies have demonstrated that the dysregulation of miRNAs may lead to alterations in differentiation, proliferation and apoptosis in cancer cells [9]. Thus, miRNA might serve as oncogenes or tumor suppressor genes. Several studies have demonstrated that specific microRNA profiles can predict prognosis and recurrence in NSCLC [10]. Additionally, microRNAs have been shown to play critical roles in the development of NSCLC. As a result, the dysregulation of microRNA expression may be a potential diagnostic and prognostic indicator [11]. Here, we investigated the expression of miRNAs in a test cohort of 119 patients, including 66 patients positive for EGFR mutations and 53 patients negative for EGFR mutations. Subsequently, we used quantitative real-time polymerase chain reaction (qRT-PCR) assays to assess miRNA profiles and their association with disease progression, survival and response to erlotinib in lung adenocarcinoma patients in a validation cohort, which consisted of 107 patients with EGFR mutations who had been treated with erlotinib.
123
Med Oncol (2014) 31:203
Page 3 of 9 203
Table 1 Clinical characteristics of the patients in training set Variate
Number
microRNA-21 expression level
microRNA-145 expression level
Mean ± SD
Mean ± SD
Tumor
Normal
Tumor
Normal
Age B60
68
2.64 ± 1.04
0.41 ± 0.23
0.57 ± 0.24
2.11 ± 0.69
[60 Gender
51
2.62 ± 0.99
0.47 ± 0.18
0.53 ± 0.27
2.03 ± 0.71
Male
57
2.67 ± 1.02
0.44 ± 0.22
0.58 ± 0.24
2.13 ± 0.71
Female
62
2.60 ± 1.07
0.46 ± 0.19
0.53 ± 0.28
2.07 ± 0.65
PS B2
83
2.61 ± 1.14
0.40 ± 0.26
0.56 ± 0.26
2.12 ± 0.77
[2
36
2.70 ± 0.81
0.28 ± 0.18
0.52 ± 0.20
2.03 ± 0.62
Smoking Never
65
2.78 ± 0.93
0.44 ± 0.23
0.60 ± 0.27
2.18 ± 0.81
Ever
54
2.59 ± 0.88
0.45 ± 0.20
0.51 ± 0.21
2.01 ± 0.60
Stage IIB
66
2.24 ± 0.89
0.44 ± 0.21
0.67 ± 0.17
2.14 ± 0.61
IIIA
53
3.22 ± 0.82
0.47 ± 0.22
0.46 ± 0.19
2.02 ± 0.63
Lymph node metastasis Y
56
2.68 ± 0.97
0.43 ± 0.23
0.66 ± 0.20
2.09 ± 0.62
N EGFR mutation
63
2.62 ± 1.16
0.41 ± 0.22
0.43 ± 0.22
2.04 ± 0.64
Y
80
3.02 ± 0.83
0.46 ± 0.21
0.50 ± 0.19
2.03 ± 0.61
N
39
2.30 ± 0.79
0.40 ± 0.18
0.64 ± 0.22
2.20 ± 0.66
were procured from patients at the time of surgery. All tissue samples were immediately flash frozen in liquid nitrogen and stored at -80 °C until use. We analyzed the histology of all of the samples in the specimen database of our department to assess the tumors and ensure the eligibility of materials. RNA isolation and real-time quantitative RT-PCR for microRNA Total RNA was extracted from grossly dissected fresh frozen tissue with Trizol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. All operating steps were performed in a laminar flow hood under RNase-free conditions. The RNA concentration was measured using the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies). To assess the purity of the RNA, the optical density (OD) was measured at 260 nm and 280 nm to determine the OD260/OD280 ratio. Only RNA with OD ratios exceeding 1.8 was used in this study. The isolated RNA was dissolved in RNA storage buffer (Ambion) and stored at -70 °C before use. Quantitative real-time PCR was performed using the Taqman PCR kit (Applied Biosystems) on a Bio-Rad IQ5
Multicolor Real-time PCR Detection System (Bio-Rad, Hercules, CA) to detect the expression levels of miRNAs in paired lung adenocarcinoma tissue and adjacent noncancerous lung parenchyma. The PCR mixture included RT product, 1 9 Taqman Universal PCR Master Mix, the corresponding primers and a Taqman probe for the target gene. The reaction mixture was incubated in a 94-well plate at 95 °C for 15 min, followed by 40 cycles of 95 °C for 15 s and then 60 °C for 60 s. The threshold cycle (Ct) was defined as the fractional cycle number at which the fluorescence passed the fixed threshold. The target miRNA expression was normalized to the Taqman endogenous control RUN6B. All assays were performed in triplicate. Follow-up All patients were followed after surgery. The chest radiograph examination and tumor maker assays were performed every 3 months for 3 years and then every 6 months thereafter. The follow-up evaluated progressionfree survival from the date of surgery to the date when disease progression was first observed by CT or when death occurred. The overall survival was evaluated from the date of surgery to the date of death. Experts who were
123
203 Page 4 of 9
Med Oncol (2014) 31:203
Table 2 Clinical characteristics of the patients in validation set Variate
Number
microRNA-21 expression level
microRNA-145 expression level
Mean ± SD
Mean ± SD
Tumor
Normal
Tumor
Normal
Age B60
58
2.79 ± 1.07
0.54 ± 0.32
0.53 ± 0.24
2.06 ± 0.61
[60 Gender
49
2.75 ± 1.26
0.56 ± 0.29
0.50 ± 0.22
2.01 ± 0.66
Male
42
2.82 ± 1.30
0.58 ± 0.33
0.55 ± 0.26
2.08 ± 0.72
Female
65
2.78 ± 1.12
0.59 ± 0.28
0.54 ± 0.25
1.99 ± 0.64
PS B2
74
2.72 ± 0.96
0.53 ± 0.31
0.54 ± 0.19
2.05 ± 0.56
[2
33
2.81 ± 1.33
0.60 ± 0.29
0.49 ± 0.26
1.98 ± 0.62
Smoking Never
68
2.88 ± 1.24
0.57 ± 0.29
0.53 ± 0.23
2.12 ± 0.77
Ever
39
2.72 ± 0.99
0.59 ± 0.27
0.42 ± 0.16
1.95 ± 0.61
Stage IIB
64
2.38 ± 0.87
0.55 ± 0.27
0.63 ± 0.27
2.08 ± 0.76
IIIA
43
3.40 ± 1.04
0.58 ± 0.31
0.41 ± 0.21
1.97 ± 0.62
Lymph node metastasis Y
40
2.80 ± 1.25
0.59 ± 0.32
0.65 ± 0.26
2.07 ± 0.68
N EGFR mutation
67
2.77 ± 1.02
0.57 ± 0.26
0.43 ± 0.29
2.02 ± 0.74
Y
77
3.14 ± 1.22
0.57 ± 0.26
0.52 ± 0.19
2.04 ± 0.61
blinded to the treatment assignment assessed the progression-free survival and treatment response by computed tomography scans. The toxic effects were evaluated on the basis of National Cancer Institute Common Terminology Criteria.
Data analysis The differential expression of miRNA-21 and miRNA-145 in lung adenocarcinoma tissue and adjacent normal parenchyma is presented as the mean ± SD. A Cox regression model was used to test the differences in progression-free survival and overall survival between the different subgroups. An unadjusted log-rank test and an exploratory, stratified Cox regression model analysis were performed to adjust for age, gender, performance status (PS 0, 1, 2 and 3) and clinical stage. All statistical analyses were performed using SAS 9.2 for Windows (SAS Institute, North Carolina, USA). Two-sided P-values are presented, and P-values less than 0.05 were considered statistically significant. Graph Pad Software (Graph Pad Software Inc., La Jolla, California, USA) was utilized to illustrate the distribution of miRNA-21 and miRNA-145 expression in each group.
123
Results From February 2010 to September 2013, our research center collected specimens from 226 patients who underwent radical resection surgery and met the inclusion criteria. The demographic and disease characteristics at baseline are presented in Tables 1 and 2. At the study endpoint (February 2014), the median follow-up period was 713 days (range 146–1,445). The median treatment with erlotinib was 524 days (range 79–1,410). Two patients received second-line treatment before they experienced RECIST-defined disease progression. The data on their progression-free survival were censored at the time of the last CT evaluation before they exhibited any evidence of disease progression (Fig. 1). Identification of miRNA expression associated with EGFR mutations in the training set Paired samples collected from the training set were profiled using miRNA microarrays to explore the relationship between miRNA expression and EGFR mutation. To avoid potential confounding factors, we established exclusion criteria, and the detailed characteristics of the patients were matched. The comparison of paired lung cancer tissue with
Med Oncol (2014) 31:203
Page 5 of 9 203
Fig. 1 Expression levels of microRNAs in training and validation set. a MiRNA-21 expression in training set associated with EGFR mutation in tumor tissue and paired adjacent normal tissue. b MiRNA-145 expression in training set associated with EGFR mutation in tumor tissue and paired adjacent normal tissue. A Expression of normal tissue with EGFR mutation. B Expression of tumor tissue with EGFR mutation. C Expression of tumor tissue without EGFR
mutation. D Expression of normal tissue without EGFR mutation. c MiRNA-21 expression in training and validation set for tumor tissue with EGFR mutation. d MiRNA-145 expression in training and validation set for tumor tissue with EGFR mutation. A Expression of tumor tissue with EGFR mutation in training set. B Expression of tumor tissue with EGFR mutation in validation set
adjacent normal lung parenchyma revealed 7 microRNAs (miRNA-21, miRNA-126, miRNA-143, miRNA-145, miRNA-182, miRNA-196 and miRNA-199) that were differentially expressed. MiRNA-21, miRNA-182, miRNA196 and miRNA-199 were expressed at significantly higher levels in cancer tissue. Conversely, miR-126, miR-143 and miR-145 were expressed at significantly lower levels. MiRNA-21 and miRNA-145 were analyzed in another research cohort as diagnostic biomarkers in our department. It noteworthy that the expression profile of miRNA-145 differed significantly between paired samples with and without EGFR mutations. Consequently, we selected miRNA-21 and miRNA-145 for further investigations.
in progression-free survival. Additionally, the patients with reduced miRNA-145 expression exhibited significantly greater progression-free survival (20.5 vs. 17.5 months, P = 0.0198) compared with the patients with elevated miRNA-145 levels. Patients with elevated miRNA-145 expression exhibited longer overall survival relative to patients with low levels of miRNA-145 (29.5 vs. 22.5 months, P = 0.0080) when overall survival was considered the end point.
Identification of miRNA expression profiles associated with disease progression and survival in the training set For patients with positive EGFR mutations, erlotinib was administered at a dose of 150 mg per day orally to patients who were positive for EGFR mutations until recurrence or intolerable toxic effects were observed. Patients in whom miRNA-21 was expressed at low levels exhibited a significant improvement in overall survival relative to patients with upregulated expression (29.0 vs. 24.1 months, P = 0.0075). However, there were no obvious differences
Validation of miRNA with response to erlotinib, disease progression and survival To investigate the association between miRNA expression and the response to erlotinib, we analyzed the expression profiles of lung adenocarcinoma tissues in cohort 2. The demographic and disease characteristics for these patients are presented in Table 2. Compared with patients with high expression levels of microRNA-21, the patients with lower expression levels exhibited increased overall survival (25.3 vs. 21.4 months, P = 0.0372) after targeted therapy with erlotinib. However, patients with elevated miRNA-21 expression exhibited a better response rate than patients with reduced expression profiles (77.3 vs. 41.7 %, P \ 0.001). Therefore, miRNA-21 plays an important role
123
203 Page 6 of 9
Med Oncol (2014) 31:203
Fig. 2 A Survival curves of lung adenocarcinoma patients with overall survival and progression-free survival as end points in training set. a Association between miRNA-21 expression and overall survival in training set. b Association between miRNA-145 expression and overall survival in training set. c Association between miRNA-21 expression and progression-free survival in training set. d Association between miRNA-145 expression and progression-free survival in training set. B Survival curves of lung adenocarcinoma patients with
overall survival and progression-free survival as end points in validation set. a Association between miRNA-21 expression and overall survival in validation set. b Association between miRNA-145 expression and overall survival in validation set. c Association between miRNA-21 expression and progression-free survival in validation set. d Association between miRNA-145 expression and progression-free survival in validation set
in determining the response to erlotinib. We did not observe a significant association between miRNA-21 expression levels and progression-free survival in cohort 2 (Fig. 2 B c, P = 0.3341). Similarly, patients with increased miRNA-145 expression exhibited a significant improvement in overall survival after adjuvant therapy with
erlotinib (Fig. 2 B d, P = 0.0131) relative to patients with low miRNA-145 expression. The period of complete remission and stable disease was longer in patients with reduced miRNA-145 expression in cancer tissues. The response rate was associated with the expression of miRNA-145 (74.1 vs. 42.6 %, P \ 0.01).
123
Med Oncol (2014) 31:203
Treatment after protocol discontinuation The treatment protocols were designed by experts from the thoracic surgery and oncology department based on the patients’ conditions when the treatment protocol was suspended. Although any treatment was permitted, the protocol recommended second-line treatment during the intersection region. At the data cutoff points, 71 patients continued to take erlotinib as first-line targeted therapy, and 121 surviving patients who had stopped receiving erlotinib were given carboplatin–paclitaxel or carboplatin– pemetrexed as second-line treatment. The remaining patients received other therapies such as carboplatin– gemcitabine and necessary stereoscopic radiotherapy. Safety Predictable toxicity profiles with erlotinib were observed in the current study. Rash (79.1 %) and diarrhea (67.3 %) were the most common adverse events. However, the side effects were tolerable, and no patients stopped taking erlotinib due to side effects. Hematologic and neurologic effects were observed in several patients. The treatment caused elevated levels of aspartate aminotransferase or alanine aminotransferase in blood tests. There were 2 cases of severe interstitial lung disease, one of which was fatal.
Discussion MiRNAs that function as key post-transcription regulators of gene expression and tyrosine kinase inhibitors associated with EGFR have become key aspects of treating lung cancer. In our study, we analyzed the prognostic significance of miRNA expression in patients who underwent complete lung adenocarcinoma resection and evaluated the response of EGFR mutation-positive patients to erlotinib. We found that the expression profiles of miRNA-21 in lung tumor tissues were significantly elevated relative to adjacent noncancerous parenchyma. Conversely, the expression levels of miRNA-145 were significantly reduced. The patients with elevated miRNA-21 expression had a shorter overall survival, but there were no significant differences in progression-free survival. Patients with elevated miRNA21 expression exhibited a good response rate to erlotinib. Additionally, the patients with increased miRNA-145 expression had longer overall survival, progression-free survival and an excellent response rate to erlotinib treatment. Previous studies have demonstrated that EGFR is one of the most common proto-oncogenes in cancer [12]. In the last decade, many researchers have examined the EGFR singling pathway. The improvement in clinical outcomes in
Page 7 of 9 203
patients treated with EGFR-TKIs has promoted various targeted therapies for malignant tumors [13]. Maarten et al. [14] demonstrated that EGFR-independent activation of the PI3KS/Akt or Ras/ERK pathway contributed to targeted therapy in NSCLC cell lines. The use of an anti-EGFR treatment in combination with a specific inhibitor of these pathways might result in additional cytotoxic effects in NSCLC cell lines. By blocking signal transduction pathways in cancer cells, EGFR-TKIs promote the apoptosis of tumor cells with EGFR-activating mutations [15]. The mutation in exons 18–21 in the coding regions of EGFR modify the binding domain of intracellular ATP and enhance the binding capacity to EGFR-TKIs [16, 17]. There are more than 30 types of activating mutations associated with the utility of drugs (e.g., erlotinib and gefitinib). The leucine-to-arginine substitution at position 858 (L858R) in exon 21 and deletion mutants in exon 19 constitute the overwhelming majority of mutations that confer hypersensitivity to EGFR-TKIs [18, 19]. However, the subsequent acquired resistance and EGFR mutation rate has become a serious problem. In the research of Zhong et al. [20] hsa-miRNA-126 and hsa-miRNA-145 were evaluated for their abilities to overcome cellular resistance and enhance gefitinib cytotoxicity. The authors found that miRNAs are associated with the response rate to EGFRTKIs as well as with progression-free survival. Thus, previous research has demonstrated that miRNAs play an important role in the chemosensitivity of cancer cells. Many studies indicate that miRNAs have effects on the prognosis and invasion of lung cancer. Therefore, miRNAs have the potential to function as diagnostic biomarkers for cancer cells and could have effects on the prognosis and invasion of lung cancer [21]. Despite the research associated with miRNA-21 and miRNA-145, the mechanism of miRNA expression during carcinogenesis and apoptosis remains poorly understood. According to previous research, STAT3, EGFR and activator protein 1 regulate miRNA-21 expression in NSCLC cells, and miRNA-21 modulates thousands of downstream targets [22, 23]. Seik et al. [24] indicated that dysregulation of miRNA-21 was more remarkable in NSCLC cells with EGFR mutations and there was a significant positive correlation between miRNA-21 expression levels and p-EGFR levels in NSCLC cells. Additionally, EGFR-TKI treatment inhibited miRNA-21 expression levels in NSCLC cells with elevated p-EGFR. Thus, the discovery revealed that miRNA-21 is a significant factor for prognosis and diagnosis in lung cancer when combined with other factors. Previous studies from our department have suggested that plasma miRNA-21, miRNA-145 and miRNA-155 are strong potential biomarkers for early cancer detection [25]. Shen et al. [26] reported that NSCLC patients with elevated miRNA-21 expression levels had shorter overall survival but exhibited
123
203 Page 8 of 9
better responses to gefitinib. This finding formed the foundation for exploring the relationships between miRNA expression profiles and response rates to erlotinib. The functions of miRNAs in regulating cell growth and proliferation have also been explored in light of the ability of miRNAs to target genes involved in cell proliferation. MiRNA-145 can serve as a tumor suppressor targeting the c-Myc oncogene by directly binding to its 30 -UTR [27]. The activation of EGFR downregulates miRNA-145 expression though ERK1/2 in lung cancer cells [28]. Many miRNAs, such as miRNA-10b, are known to impact cell invasion and tumor metastasis. Similar to miRNA-10b, miRNA-145 is also expressed in microvasculature pericytes and inhibits the migration of microvascular cells in response to growth factor gradients by targeting the transcription factor Fli-1 [29]. William et al. [30] indicated that the restoration of miRNA-145 inhibited cancer cell growth in adenocarcinoma patients with EGFR mutations. Further studies will be needed to address these issues, and new modalities have been used to promote research focusing on miRNAs. In conclusion, our current study demonstrates that microRNA expression profiles in completely resected lung adenocarcinoma patients are prognostic and are associated with the response to erlotinib. Our study reported that lung adenocarcinoma patients with lower miRNA-21 expression have longer overall survival and exhibit a poor response rate to erlotinib. Additionally, increased miRNA-145 expression could serve as a significant predictive factor for increased overall survival, progression-free survival and response to erlotinib. There are several limitations in this study, including the limited number of miRNA candidates, the accuracy of the follow-up results and the use of only stage IIB and IIIA patients. In addition, due to the heterogeneity of lung cancer, the data from the patients who were enrolled exclusively in our department is insufficient to generalize to patients at other centers or with other stages of disease. The development of new insights into lung cancer biology and the development of novel therapies will improve treatment outcomes. We firmly support the idea that miRNAs can provide a new solution to refractory diseases and improve patient survival. Conflict of interest There is no conflict of interest for any of the authors in any aspects of the article.
References 1. Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics. CA Cancer J Clin. 2001;51:15–36. 2. Tanner NT, Mehta H, Silvestri GA. New testing for lung cancer screening. Oncology. 2012;26:176–82. 3. Pfister DG, Johnson DH, Azzoli CG, Sause W, Smith TJ, et al. American Society of Clinical Oncology treatment of unresectable
123
Med Oncol (2014) 31:203
4.
5.
6.
7. 8. 9. 10.
11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23.
non-small-cell lung cancer guideline: update 2003. J Clin Oncol. 2003;22:330–53. Yilmaz A, Damadoglu E, Salturk C, Okur E, Tuncer LY, et al. Delays in the diagnosis and treatment of primary lung cancer: are longer delays associated with advanced pathological stage? Ups J Med Sci. 2008;113:287–96. Yang G, Yao Y, Zhou J, Zhao Q. Effects of icotinib, a novel epidermal growth factor receptor tyrosine kinase inhibitor, in EGFR-mutated non-small cell lung cancer. Oncol Rep. 2012;6:2066–72. Harari PM, Allen GW, Bonner JA. Biology of interactions: antiepidermal growth factor receptor agents. J Clin Oncol. 2007;25:4057–65. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–97. Farazi TA, Spitzer JI, Morozov P, Tuschl T. MiRNAs in human cancer. J Pathol. 2011;223:102–15. Papagiannakopoulos T, Kosik KS. MicroRNAs: regulators of oncogenesis and stemness. BMC Med. 2008;6:6–15. Cheung TH, Man KN, Yu MY, Yim SF, Siu NS, et al. Dysregulated microRNAs in the pathogenesis and progression of cervical neoplasm. Cell Cycle. 2012;11:2876–84. Duncavage E, Goodgame B, Sezhiyan A, Govindan R, Pfeifer J. Use of microRNA expression levels to predict outcomes in resected stage I non-small cell lung cancer. J Thorac Oncol. 2010;11:1755–63. Yu HA, Arcila ME, Hellmann MD, Kris MG, Ladanyi M, et al. Poor response to erlotinib in patients with tumors containing baseline EGFR T790M mutations found by routine clinical molecular testing. Ann Oncol. 2014;25:423–8. Reungwetwattana T, Weroha SJ, Molina JR. Oncogenic pathways, molecularly targeted therapies, and highlighted clinical trials in non-small-cell lung cancer (NSCLC). Clin Lung Cancer. 2012;13:252–66. Maarten L, Janmaat ML, Rodriguez JA, Kruyt FA, Giaccone G. Enhanced cytotoxicity induced by gefitinib and specific inhibitors of the Ras or phosphatidyl inositol-3 kinase pathways in nonsmall cell lung cancer cells. Int J Cancer. 2006;118:209–14. Zhang Z, Stiegler AL, Boggon TJ, Kobayashi S, Halmos B. EGFR mutated lung cancer: a paradigm of molecular oncology. Oncotarget. 2010;1:497–514. Herbst RS, Fukuoka M, Baselga J. Gefitinib—a novel targeted approach to treating cancer. Nat Rev Cancer. 2004;12:956–65. Sharma SV, Bell DW, Settleman J, Haber DA. Epidermal growth factor receptor mutations in lung cancer. Nat Rev Cancer. 2007;7:169–81. Costa DB, Kobayashi S. Are exon 19 deletions and L858R EGFR mutations in non-small-cell lung cancer clinically different? Br J Cancer. 2007;96:399 (author reply 400). Smouse JH, Cibas ES, Janne PA, Joshi VA, Zou KH, Lindeman NI. EGFR mutations are detected comparably in cytologic and surgical pathology specimens of nonsmall cell lung cancer. Cancer. 2009;117:67–72. Zhong Miao, Ma Xin, Sun Caijun, et al. MicroRNAs reduce tumor growth and contribute to enhance cytotoxicity induced by gefitinib in non-small cell lung cancer. Chem Biol Interact. 2010;184:431–8. Zhang T, Wang Q, Zhao D, Chen L. The oncogenetic role of microRNA-31 as a potential biomarker in oesophageal squamous cell carcinoma. Clin Sci. 2011;121:437–47. Iliopoulos D, Jaeger SA, Hirsch HA, Bulyk ML, Struhl K. STAT3 activation of miR-21 and miR-181b-1 via PTEN and CYLD are part of the epigenetic switch linking inflammation to cancer. Mol Cell. 2010;39:493–506. Krichevsky AM, Gabriely G. MiR-21: a small multi-faceted RNA. J Cell Mol Med. 2009;13:39–53.
Med Oncol (2014) 31:203 24. Seike M, Goto A, Okano T, Bowman ED, Schetter AJ, et al. MiR21 is an EGFR-regulated anti-apoptotic factor in lung cancer in never-smokers. Proc Natl Acad Sci USA. 2009;106:12085–90. 25. Tang Dongfang, Shen Yi, Wang Mingzhao, Yang Ronghua, Wang Zizong, et al. Identification of plasma microRNAs as novel noninvasive biomarkers for early detection of lung cancer. Eur J Cancer Prev. 2013;6:540–8. 26. Shen Yi, Tang Dongfang, Yao Ruyong, Wang Mingzhao, Wang Yongjie, et al. microRNA expression profiles associated with survival, disease progression, and response to gefitinib in completely resected non-small-cell lung cancer with EGFR mutation. Med Oncol. 2013;30:750–8. 27. Cho WC, Chow AS, Au JS. Restoration of tumour suppressor hsa-miR-145 inhibits cancer cell growth in lung adenocarcinoma
Page 9 of 9 203 patients with epidermal growth factor receptor mutation. Eur J Cancer. 2009;45:2197–206. 28. Ma Z, Qi J, Meng S, Wen B, Zhang J. Swimming exercise training-induced left ventricular hypertrophy involves microRNAs and synergistic regulation of the PI3K/AKT/mTOR signaling pathway. Eur J Appl Physiol. 2013;113:2473–86. 29. Larsson E, Fredlund Fuchs P, Heldin J, Barkefors I, Bondjers C, et al. Discovery of microvascular miRNAs using public gene expression data: miR-145 is expressed in pericytes and is a regulator of Fli1. Genome Med 2009; 11:108.01-12. 30. Cho WCS, Chow ASC, Au JSK. Restoration of tumour suppressor hsa-miR-145 inhibits cancer cell growth in lung adenocarcinoma patients with epidermal growth factor receptor mutation. Eur J Cancer. 2009;45:2197–209.
123
ORIGINAL PAPER
Prognostic significance of microRNA expression in completely resected lung adenocarcinoma and the associated response to erlotinib Guanzhong Yan • Ruyong Yao • Dongfang Tang Tong Qiu • Yi Shen • Wenjie Jiao • Nan Ge • Yunpeng Xuan • Yongjie Wang
•
Received: 9 August 2014 / Accepted: 22 August 2014 Ó Springer Science+Business Media New York 2014
Abstract The mechanism of action of oncogenic or tumor suppressor microRNAs is not well understood. We examined the microRNA expression profile in completely resected lung adenocarcinoma and examined the associated response to erlotinib. The lung adenocarcinoma tissue and adjacent normal lung parenchyma of 226 stage IIB and IIIA patients who underwent complete resection were obtained for two separate retrospective cohorts. In cohort 1 (119 patients; 80 with epidermal growth factor receptor (EGFR) mutations and 39 without), miRNA microarrays were used to identify EGFR-related miRNAs and their association with survival. In cohort 2 (107 patients with EGFR mutations), the miRNAs and their association with survival and response to erlotinib were analyzed by qRTPCR. Cox proportional hazards regression was used to evaluate the effect of treatment on survival. As a result, erlotinib is associated with a significant improvement in overall survival (P = 0.0075, cohort 1; P = 0.0372, cohort 2) and disease progression (P = 0.6929, cohort 1; P = 0.3347, cohort 2) in patients with reduced miRNA-21 expression. Additionally, miRNA-145 is strongly associated with overall survival (P = 0.0008, cohort 1; P = 0.0131, cohort 2) and progression-free survival (P = 0.0198, cohort 1; P = 0.0269, cohort 2). Understanding the response rate to erlotinib relative to miRNA21 (77.3 vs. 41.7 %, P \ 0.01) and miRNA-145 (74.1 vs. G. Yan D. Tang T. Qiu Y. Shen W. Jiao Y. Xuan Y. Wang (&) Department of Thoracic Surgery, The Affiliated Hospital of Qingdao University, 16 Jiangsu Road, Qingdao 266003, China e-mail: [email protected] R. Yao N. Ge Central Laboratory, The Affiliated Hospital of Qingdao University, Qingdao 266003, China
42.6 %, P \ 0.01) expression is critical. The miRNA expression profiles differed significantly between patients with and without EGFR mutations. In conclusion, lung adenocarcinoma patients with reduced miRNA-21 expression exhibit longer overall survival and a poor response rate to erlotinib. Increased miRNA-145 levels can predict overall survival, progression-free survival and excellent response rate to erlotinib. Keywords microRNA Erlotinib Lung adenocarcinoma EGFR mutation Survival Disease progression
Introduction Lung cancer is the leading cause of cancer-related deaths worldwide, and non-small cell lung cancer (NSCLC) is the predominant pathological type, accounting for approximately 80 % of all cases of lung cancer [1, 2]. Despite the recent advances in diagnosis, chemotherapy and targeted therapy, the overall survival rate of NSCLC patients remains low, and the recurrence rate is very high, including in patients diagnosed early. The poor prognosis is closely related to our relatively limited understanding of the biology, late presentation and the heterogeneity of lung cancer. Patients with lung adenocarcinoma exhibit better survival than patients with other pathological subtypes of lung cancer. However, most patients are not completely cured, and the 5-year overall survival rate varies from 70 % in stage IA to 25 % in stage IIIA. The most effective treatment for lung adenocarcinoma is complete surgical resection combined with chemotherapy, radiation therapy or targeted therapy [3, 4]. The epidermal growth factor receptor (EGFR) regulates tumor cell proliferation, invasion, migration and apoptosis
123
203 Page 2 of 9
Med Oncol (2014) 31:203
Methods
China. Written informed consent was obtained from all patients. Patient data for all specimens were anonymized according to the legal standard. This study was approved by the ethical committee of our hospital. We used the amplification refractory mutation system (ARMS) to test for EGFR mutations in stage IIB and IIIA lung adenocarcinoma from patients receiving complete surgical resection. Patients with the T790 M mutation, in which threonine at amino acid 790 is replaced by methionine, were excluded from the study. The eligibility criteria for this study included an age of 75 years or younger (it remains uncertain whether patients [75 years old will benefit from platinum-based therapy) and an ECOG performance status of 0–3. All patients selected in the study received preoperative computed tomography scans and had no significant cardiac dysfunction, active infection or neurologic disorders. The inclusion criteria also required the patients to have undergone complete preoperative or intraoperative staging as well as intraoperative mediastinal lymph node examination or biopsy of nodes. All nodes larger than 1.5 cm were resected. Patients with sub-lobectomy (wedge resection or segmental resection), mixed histological features, diffuse lobar or multifocal bronchi alveolar carcinoma were ineligible for this study. Patients who underwent preoperative chemotherapy and patients who had breast cancer, renal cell carcinoma, gastric cancer, melanoma or other cancer treatment within the previous 5 years were ineligible. This study collected 226 pairs of lung adenocarcinoma tissues and adjacent noncancerous lung parenchyma within 4 cm of tumors that were resected by lobectomy between February 2010 and September 2013 at the Department of Thoracic Surgery at the Affiliated Hospital, Qingdao University, China. The patients who underwent radical resection and met the standards described previously were divided into two separate cohorts. Cohort 1 consisted of 119 patients (66 were positive for EGFR mutations, and 53 were negative). The miRNA microarray data were available and were analyzed to search for miRNAs associated with EGFR mutation and survival. Table 1 lists the detailed eligibility and exclusion criteria. Cohort 2 consisted of 207 lung adenocarcinoma patients with EGFR mutations. Erlotinib was administered as firstline adjuvant chemotherapy at a dose of 150 mg per day orally until the computed tomographic scan revealed disease progression or until the onset of intolerable toxic effects. In this cohort, we used qRT-PCR assays to investigate the relationship between miRNA expression and the response to erlotinib. Furthermore, the correlation between miRNA expression and disease progression was examined.
Patients
Sample processing
The study was conducted in the Department of Thoracic Surgery, The Affiliated Hospital, Qingdao University,
Primary lung adenocarcinoma tissue and paired adjacent normal tissue meeting the standards described preciously
[5]. Targeted therapies associated with epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) improve progression-free survival and overall survival in lung adenocarcinoma patients who have undergone complete surgical resection. EGFR-TKIs represent an important targeted therapy because these inhibitors block the complex EGFR transduction pathway involving the downstream Ras/MAPK cascade and phosphatidyl inositol 3-kinase (PI3 K) pathway [6]. Although there has been progress in treating NSCLC with EGFR-TKIs (gefitinib or erlotinib), the relationship between EGFR mutation and targeted therapy has not been elucidated. In addition, additional effective targeted therapies are urgently needed. MicroRNAs are a class of small, non-coding, singlestranded, endogenous RNAs that regulate gene expression. MicroRNAs bind to target mRNAs in the 30 untranslated region (30 -UTR) and cause mRNA cleavage or translational repression [7, 8]. The complementarity between the seed sequence of an individual microRNA and the target miRNA determines the mRNA specificity. Several proteomic studies have revealed that every miRNA has a board range of targets. It has been estimated that nearly half of all genes are regulated by miRNAs, and each miRNA can target several hundred transcripts, making miRNAs one of the largest families of gene regulators. Previous studies have demonstrated that the dysregulation of miRNAs may lead to alterations in differentiation, proliferation and apoptosis in cancer cells [9]. Thus, miRNA might serve as oncogenes or tumor suppressor genes. Several studies have demonstrated that specific microRNA profiles can predict prognosis and recurrence in NSCLC [10]. Additionally, microRNAs have been shown to play critical roles in the development of NSCLC. As a result, the dysregulation of microRNA expression may be a potential diagnostic and prognostic indicator [11]. Here, we investigated the expression of miRNAs in a test cohort of 119 patients, including 66 patients positive for EGFR mutations and 53 patients negative for EGFR mutations. Subsequently, we used quantitative real-time polymerase chain reaction (qRT-PCR) assays to assess miRNA profiles and their association with disease progression, survival and response to erlotinib in lung adenocarcinoma patients in a validation cohort, which consisted of 107 patients with EGFR mutations who had been treated with erlotinib.
123
Med Oncol (2014) 31:203
Page 3 of 9 203
Table 1 Clinical characteristics of the patients in training set Variate
Number
microRNA-21 expression level
microRNA-145 expression level
Mean ± SD
Mean ± SD
Tumor
Normal
Tumor
Normal
Age B60
68
2.64 ± 1.04
0.41 ± 0.23
0.57 ± 0.24
2.11 ± 0.69
[60 Gender
51
2.62 ± 0.99
0.47 ± 0.18
0.53 ± 0.27
2.03 ± 0.71
Male
57
2.67 ± 1.02
0.44 ± 0.22
0.58 ± 0.24
2.13 ± 0.71
Female
62
2.60 ± 1.07
0.46 ± 0.19
0.53 ± 0.28
2.07 ± 0.65
PS B2
83
2.61 ± 1.14
0.40 ± 0.26
0.56 ± 0.26
2.12 ± 0.77
[2
36
2.70 ± 0.81
0.28 ± 0.18
0.52 ± 0.20
2.03 ± 0.62
Smoking Never
65
2.78 ± 0.93
0.44 ± 0.23
0.60 ± 0.27
2.18 ± 0.81
Ever
54
2.59 ± 0.88
0.45 ± 0.20
0.51 ± 0.21
2.01 ± 0.60
Stage IIB
66
2.24 ± 0.89
0.44 ± 0.21
0.67 ± 0.17
2.14 ± 0.61
IIIA
53
3.22 ± 0.82
0.47 ± 0.22
0.46 ± 0.19
2.02 ± 0.63
Lymph node metastasis Y
56
2.68 ± 0.97
0.43 ± 0.23
0.66 ± 0.20
2.09 ± 0.62
N EGFR mutation
63
2.62 ± 1.16
0.41 ± 0.22
0.43 ± 0.22
2.04 ± 0.64
Y
80
3.02 ± 0.83
0.46 ± 0.21
0.50 ± 0.19
2.03 ± 0.61
N
39
2.30 ± 0.79
0.40 ± 0.18
0.64 ± 0.22
2.20 ± 0.66
were procured from patients at the time of surgery. All tissue samples were immediately flash frozen in liquid nitrogen and stored at -80 °C until use. We analyzed the histology of all of the samples in the specimen database of our department to assess the tumors and ensure the eligibility of materials. RNA isolation and real-time quantitative RT-PCR for microRNA Total RNA was extracted from grossly dissected fresh frozen tissue with Trizol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. All operating steps were performed in a laminar flow hood under RNase-free conditions. The RNA concentration was measured using the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies). To assess the purity of the RNA, the optical density (OD) was measured at 260 nm and 280 nm to determine the OD260/OD280 ratio. Only RNA with OD ratios exceeding 1.8 was used in this study. The isolated RNA was dissolved in RNA storage buffer (Ambion) and stored at -70 °C before use. Quantitative real-time PCR was performed using the Taqman PCR kit (Applied Biosystems) on a Bio-Rad IQ5
Multicolor Real-time PCR Detection System (Bio-Rad, Hercules, CA) to detect the expression levels of miRNAs in paired lung adenocarcinoma tissue and adjacent noncancerous lung parenchyma. The PCR mixture included RT product, 1 9 Taqman Universal PCR Master Mix, the corresponding primers and a Taqman probe for the target gene. The reaction mixture was incubated in a 94-well plate at 95 °C for 15 min, followed by 40 cycles of 95 °C for 15 s and then 60 °C for 60 s. The threshold cycle (Ct) was defined as the fractional cycle number at which the fluorescence passed the fixed threshold. The target miRNA expression was normalized to the Taqman endogenous control RUN6B. All assays were performed in triplicate. Follow-up All patients were followed after surgery. The chest radiograph examination and tumor maker assays were performed every 3 months for 3 years and then every 6 months thereafter. The follow-up evaluated progressionfree survival from the date of surgery to the date when disease progression was first observed by CT or when death occurred. The overall survival was evaluated from the date of surgery to the date of death. Experts who were
123
203 Page 4 of 9
Med Oncol (2014) 31:203
Table 2 Clinical characteristics of the patients in validation set Variate
Number
microRNA-21 expression level
microRNA-145 expression level
Mean ± SD
Mean ± SD
Tumor
Normal
Tumor
Normal
Age B60
58
2.79 ± 1.07
0.54 ± 0.32
0.53 ± 0.24
2.06 ± 0.61
[60 Gender
49
2.75 ± 1.26
0.56 ± 0.29
0.50 ± 0.22
2.01 ± 0.66
Male
42
2.82 ± 1.30
0.58 ± 0.33
0.55 ± 0.26
2.08 ± 0.72
Female
65
2.78 ± 1.12
0.59 ± 0.28
0.54 ± 0.25
1.99 ± 0.64
PS B2
74
2.72 ± 0.96
0.53 ± 0.31
0.54 ± 0.19
2.05 ± 0.56
[2
33
2.81 ± 1.33
0.60 ± 0.29
0.49 ± 0.26
1.98 ± 0.62
Smoking Never
68
2.88 ± 1.24
0.57 ± 0.29
0.53 ± 0.23
2.12 ± 0.77
Ever
39
2.72 ± 0.99
0.59 ± 0.27
0.42 ± 0.16
1.95 ± 0.61
Stage IIB
64
2.38 ± 0.87
0.55 ± 0.27
0.63 ± 0.27
2.08 ± 0.76
IIIA
43
3.40 ± 1.04
0.58 ± 0.31
0.41 ± 0.21
1.97 ± 0.62
Lymph node metastasis Y
40
2.80 ± 1.25
0.59 ± 0.32
0.65 ± 0.26
2.07 ± 0.68
N EGFR mutation
67
2.77 ± 1.02
0.57 ± 0.26
0.43 ± 0.29
2.02 ± 0.74
Y
77
3.14 ± 1.22
0.57 ± 0.26
0.52 ± 0.19
2.04 ± 0.61
blinded to the treatment assignment assessed the progression-free survival and treatment response by computed tomography scans. The toxic effects were evaluated on the basis of National Cancer Institute Common Terminology Criteria.
Data analysis The differential expression of miRNA-21 and miRNA-145 in lung adenocarcinoma tissue and adjacent normal parenchyma is presented as the mean ± SD. A Cox regression model was used to test the differences in progression-free survival and overall survival between the different subgroups. An unadjusted log-rank test and an exploratory, stratified Cox regression model analysis were performed to adjust for age, gender, performance status (PS 0, 1, 2 and 3) and clinical stage. All statistical analyses were performed using SAS 9.2 for Windows (SAS Institute, North Carolina, USA). Two-sided P-values are presented, and P-values less than 0.05 were considered statistically significant. Graph Pad Software (Graph Pad Software Inc., La Jolla, California, USA) was utilized to illustrate the distribution of miRNA-21 and miRNA-145 expression in each group.
123
Results From February 2010 to September 2013, our research center collected specimens from 226 patients who underwent radical resection surgery and met the inclusion criteria. The demographic and disease characteristics at baseline are presented in Tables 1 and 2. At the study endpoint (February 2014), the median follow-up period was 713 days (range 146–1,445). The median treatment with erlotinib was 524 days (range 79–1,410). Two patients received second-line treatment before they experienced RECIST-defined disease progression. The data on their progression-free survival were censored at the time of the last CT evaluation before they exhibited any evidence of disease progression (Fig. 1). Identification of miRNA expression associated with EGFR mutations in the training set Paired samples collected from the training set were profiled using miRNA microarrays to explore the relationship between miRNA expression and EGFR mutation. To avoid potential confounding factors, we established exclusion criteria, and the detailed characteristics of the patients were matched. The comparison of paired lung cancer tissue with
Med Oncol (2014) 31:203
Page 5 of 9 203
Fig. 1 Expression levels of microRNAs in training and validation set. a MiRNA-21 expression in training set associated with EGFR mutation in tumor tissue and paired adjacent normal tissue. b MiRNA-145 expression in training set associated with EGFR mutation in tumor tissue and paired adjacent normal tissue. A Expression of normal tissue with EGFR mutation. B Expression of tumor tissue with EGFR mutation. C Expression of tumor tissue without EGFR
mutation. D Expression of normal tissue without EGFR mutation. c MiRNA-21 expression in training and validation set for tumor tissue with EGFR mutation. d MiRNA-145 expression in training and validation set for tumor tissue with EGFR mutation. A Expression of tumor tissue with EGFR mutation in training set. B Expression of tumor tissue with EGFR mutation in validation set
adjacent normal lung parenchyma revealed 7 microRNAs (miRNA-21, miRNA-126, miRNA-143, miRNA-145, miRNA-182, miRNA-196 and miRNA-199) that were differentially expressed. MiRNA-21, miRNA-182, miRNA196 and miRNA-199 were expressed at significantly higher levels in cancer tissue. Conversely, miR-126, miR-143 and miR-145 were expressed at significantly lower levels. MiRNA-21 and miRNA-145 were analyzed in another research cohort as diagnostic biomarkers in our department. It noteworthy that the expression profile of miRNA-145 differed significantly between paired samples with and without EGFR mutations. Consequently, we selected miRNA-21 and miRNA-145 for further investigations.
in progression-free survival. Additionally, the patients with reduced miRNA-145 expression exhibited significantly greater progression-free survival (20.5 vs. 17.5 months, P = 0.0198) compared with the patients with elevated miRNA-145 levels. Patients with elevated miRNA-145 expression exhibited longer overall survival relative to patients with low levels of miRNA-145 (29.5 vs. 22.5 months, P = 0.0080) when overall survival was considered the end point.
Identification of miRNA expression profiles associated with disease progression and survival in the training set For patients with positive EGFR mutations, erlotinib was administered at a dose of 150 mg per day orally to patients who were positive for EGFR mutations until recurrence or intolerable toxic effects were observed. Patients in whom miRNA-21 was expressed at low levels exhibited a significant improvement in overall survival relative to patients with upregulated expression (29.0 vs. 24.1 months, P = 0.0075). However, there were no obvious differences
Validation of miRNA with response to erlotinib, disease progression and survival To investigate the association between miRNA expression and the response to erlotinib, we analyzed the expression profiles of lung adenocarcinoma tissues in cohort 2. The demographic and disease characteristics for these patients are presented in Table 2. Compared with patients with high expression levels of microRNA-21, the patients with lower expression levels exhibited increased overall survival (25.3 vs. 21.4 months, P = 0.0372) after targeted therapy with erlotinib. However, patients with elevated miRNA-21 expression exhibited a better response rate than patients with reduced expression profiles (77.3 vs. 41.7 %, P \ 0.001). Therefore, miRNA-21 plays an important role
123
203 Page 6 of 9
Med Oncol (2014) 31:203
Fig. 2 A Survival curves of lung adenocarcinoma patients with overall survival and progression-free survival as end points in training set. a Association between miRNA-21 expression and overall survival in training set. b Association between miRNA-145 expression and overall survival in training set. c Association between miRNA-21 expression and progression-free survival in training set. d Association between miRNA-145 expression and progression-free survival in training set. B Survival curves of lung adenocarcinoma patients with
overall survival and progression-free survival as end points in validation set. a Association between miRNA-21 expression and overall survival in validation set. b Association between miRNA-145 expression and overall survival in validation set. c Association between miRNA-21 expression and progression-free survival in validation set. d Association between miRNA-145 expression and progression-free survival in validation set
in determining the response to erlotinib. We did not observe a significant association between miRNA-21 expression levels and progression-free survival in cohort 2 (Fig. 2 B c, P = 0.3341). Similarly, patients with increased miRNA-145 expression exhibited a significant improvement in overall survival after adjuvant therapy with
erlotinib (Fig. 2 B d, P = 0.0131) relative to patients with low miRNA-145 expression. The period of complete remission and stable disease was longer in patients with reduced miRNA-145 expression in cancer tissues. The response rate was associated with the expression of miRNA-145 (74.1 vs. 42.6 %, P \ 0.01).
123
Med Oncol (2014) 31:203
Treatment after protocol discontinuation The treatment protocols were designed by experts from the thoracic surgery and oncology department based on the patients’ conditions when the treatment protocol was suspended. Although any treatment was permitted, the protocol recommended second-line treatment during the intersection region. At the data cutoff points, 71 patients continued to take erlotinib as first-line targeted therapy, and 121 surviving patients who had stopped receiving erlotinib were given carboplatin–paclitaxel or carboplatin– pemetrexed as second-line treatment. The remaining patients received other therapies such as carboplatin– gemcitabine and necessary stereoscopic radiotherapy. Safety Predictable toxicity profiles with erlotinib were observed in the current study. Rash (79.1 %) and diarrhea (67.3 %) were the most common adverse events. However, the side effects were tolerable, and no patients stopped taking erlotinib due to side effects. Hematologic and neurologic effects were observed in several patients. The treatment caused elevated levels of aspartate aminotransferase or alanine aminotransferase in blood tests. There were 2 cases of severe interstitial lung disease, one of which was fatal.
Discussion MiRNAs that function as key post-transcription regulators of gene expression and tyrosine kinase inhibitors associated with EGFR have become key aspects of treating lung cancer. In our study, we analyzed the prognostic significance of miRNA expression in patients who underwent complete lung adenocarcinoma resection and evaluated the response of EGFR mutation-positive patients to erlotinib. We found that the expression profiles of miRNA-21 in lung tumor tissues were significantly elevated relative to adjacent noncancerous parenchyma. Conversely, the expression levels of miRNA-145 were significantly reduced. The patients with elevated miRNA-21 expression had a shorter overall survival, but there were no significant differences in progression-free survival. Patients with elevated miRNA21 expression exhibited a good response rate to erlotinib. Additionally, the patients with increased miRNA-145 expression had longer overall survival, progression-free survival and an excellent response rate to erlotinib treatment. Previous studies have demonstrated that EGFR is one of the most common proto-oncogenes in cancer [12]. In the last decade, many researchers have examined the EGFR singling pathway. The improvement in clinical outcomes in
Page 7 of 9 203
patients treated with EGFR-TKIs has promoted various targeted therapies for malignant tumors [13]. Maarten et al. [14] demonstrated that EGFR-independent activation of the PI3KS/Akt or Ras/ERK pathway contributed to targeted therapy in NSCLC cell lines. The use of an anti-EGFR treatment in combination with a specific inhibitor of these pathways might result in additional cytotoxic effects in NSCLC cell lines. By blocking signal transduction pathways in cancer cells, EGFR-TKIs promote the apoptosis of tumor cells with EGFR-activating mutations [15]. The mutation in exons 18–21 in the coding regions of EGFR modify the binding domain of intracellular ATP and enhance the binding capacity to EGFR-TKIs [16, 17]. There are more than 30 types of activating mutations associated with the utility of drugs (e.g., erlotinib and gefitinib). The leucine-to-arginine substitution at position 858 (L858R) in exon 21 and deletion mutants in exon 19 constitute the overwhelming majority of mutations that confer hypersensitivity to EGFR-TKIs [18, 19]. However, the subsequent acquired resistance and EGFR mutation rate has become a serious problem. In the research of Zhong et al. [20] hsa-miRNA-126 and hsa-miRNA-145 were evaluated for their abilities to overcome cellular resistance and enhance gefitinib cytotoxicity. The authors found that miRNAs are associated with the response rate to EGFRTKIs as well as with progression-free survival. Thus, previous research has demonstrated that miRNAs play an important role in the chemosensitivity of cancer cells. Many studies indicate that miRNAs have effects on the prognosis and invasion of lung cancer. Therefore, miRNAs have the potential to function as diagnostic biomarkers for cancer cells and could have effects on the prognosis and invasion of lung cancer [21]. Despite the research associated with miRNA-21 and miRNA-145, the mechanism of miRNA expression during carcinogenesis and apoptosis remains poorly understood. According to previous research, STAT3, EGFR and activator protein 1 regulate miRNA-21 expression in NSCLC cells, and miRNA-21 modulates thousands of downstream targets [22, 23]. Seik et al. [24] indicated that dysregulation of miRNA-21 was more remarkable in NSCLC cells with EGFR mutations and there was a significant positive correlation between miRNA-21 expression levels and p-EGFR levels in NSCLC cells. Additionally, EGFR-TKI treatment inhibited miRNA-21 expression levels in NSCLC cells with elevated p-EGFR. Thus, the discovery revealed that miRNA-21 is a significant factor for prognosis and diagnosis in lung cancer when combined with other factors. Previous studies from our department have suggested that plasma miRNA-21, miRNA-145 and miRNA-155 are strong potential biomarkers for early cancer detection [25]. Shen et al. [26] reported that NSCLC patients with elevated miRNA-21 expression levels had shorter overall survival but exhibited
123
203 Page 8 of 9
better responses to gefitinib. This finding formed the foundation for exploring the relationships between miRNA expression profiles and response rates to erlotinib. The functions of miRNAs in regulating cell growth and proliferation have also been explored in light of the ability of miRNAs to target genes involved in cell proliferation. MiRNA-145 can serve as a tumor suppressor targeting the c-Myc oncogene by directly binding to its 30 -UTR [27]. The activation of EGFR downregulates miRNA-145 expression though ERK1/2 in lung cancer cells [28]. Many miRNAs, such as miRNA-10b, are known to impact cell invasion and tumor metastasis. Similar to miRNA-10b, miRNA-145 is also expressed in microvasculature pericytes and inhibits the migration of microvascular cells in response to growth factor gradients by targeting the transcription factor Fli-1 [29]. William et al. [30] indicated that the restoration of miRNA-145 inhibited cancer cell growth in adenocarcinoma patients with EGFR mutations. Further studies will be needed to address these issues, and new modalities have been used to promote research focusing on miRNAs. In conclusion, our current study demonstrates that microRNA expression profiles in completely resected lung adenocarcinoma patients are prognostic and are associated with the response to erlotinib. Our study reported that lung adenocarcinoma patients with lower miRNA-21 expression have longer overall survival and exhibit a poor response rate to erlotinib. Additionally, increased miRNA-145 expression could serve as a significant predictive factor for increased overall survival, progression-free survival and response to erlotinib. There are several limitations in this study, including the limited number of miRNA candidates, the accuracy of the follow-up results and the use of only stage IIB and IIIA patients. In addition, due to the heterogeneity of lung cancer, the data from the patients who were enrolled exclusively in our department is insufficient to generalize to patients at other centers or with other stages of disease. The development of new insights into lung cancer biology and the development of novel therapies will improve treatment outcomes. We firmly support the idea that miRNAs can provide a new solution to refractory diseases and improve patient survival. Conflict of interest There is no conflict of interest for any of the authors in any aspects of the article.
References 1. Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics. CA Cancer J Clin. 2001;51:15–36. 2. Tanner NT, Mehta H, Silvestri GA. New testing for lung cancer screening. Oncology. 2012;26:176–82. 3. Pfister DG, Johnson DH, Azzoli CG, Sause W, Smith TJ, et al. American Society of Clinical Oncology treatment of unresectable
123
Med Oncol (2014) 31:203
4.
5.
6.
7. 8. 9. 10.
11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23.
non-small-cell lung cancer guideline: update 2003. J Clin Oncol. 2003;22:330–53. Yilmaz A, Damadoglu E, Salturk C, Okur E, Tuncer LY, et al. Delays in the diagnosis and treatment of primary lung cancer: are longer delays associated with advanced pathological stage? Ups J Med Sci. 2008;113:287–96. Yang G, Yao Y, Zhou J, Zhao Q. Effects of icotinib, a novel epidermal growth factor receptor tyrosine kinase inhibitor, in EGFR-mutated non-small cell lung cancer. Oncol Rep. 2012;6:2066–72. Harari PM, Allen GW, Bonner JA. Biology of interactions: antiepidermal growth factor receptor agents. J Clin Oncol. 2007;25:4057–65. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–97. Farazi TA, Spitzer JI, Morozov P, Tuschl T. MiRNAs in human cancer. J Pathol. 2011;223:102–15. Papagiannakopoulos T, Kosik KS. MicroRNAs: regulators of oncogenesis and stemness. BMC Med. 2008;6:6–15. Cheung TH, Man KN, Yu MY, Yim SF, Siu NS, et al. Dysregulated microRNAs in the pathogenesis and progression of cervical neoplasm. Cell Cycle. 2012;11:2876–84. Duncavage E, Goodgame B, Sezhiyan A, Govindan R, Pfeifer J. Use of microRNA expression levels to predict outcomes in resected stage I non-small cell lung cancer. J Thorac Oncol. 2010;11:1755–63. Yu HA, Arcila ME, Hellmann MD, Kris MG, Ladanyi M, et al. Poor response to erlotinib in patients with tumors containing baseline EGFR T790M mutations found by routine clinical molecular testing. Ann Oncol. 2014;25:423–8. Reungwetwattana T, Weroha SJ, Molina JR. Oncogenic pathways, molecularly targeted therapies, and highlighted clinical trials in non-small-cell lung cancer (NSCLC). Clin Lung Cancer. 2012;13:252–66. Maarten L, Janmaat ML, Rodriguez JA, Kruyt FA, Giaccone G. Enhanced cytotoxicity induced by gefitinib and specific inhibitors of the Ras or phosphatidyl inositol-3 kinase pathways in nonsmall cell lung cancer cells. Int J Cancer. 2006;118:209–14. Zhang Z, Stiegler AL, Boggon TJ, Kobayashi S, Halmos B. EGFR mutated lung cancer: a paradigm of molecular oncology. Oncotarget. 2010;1:497–514. Herbst RS, Fukuoka M, Baselga J. Gefitinib—a novel targeted approach to treating cancer. Nat Rev Cancer. 2004;12:956–65. Sharma SV, Bell DW, Settleman J, Haber DA. Epidermal growth factor receptor mutations in lung cancer. Nat Rev Cancer. 2007;7:169–81. Costa DB, Kobayashi S. Are exon 19 deletions and L858R EGFR mutations in non-small-cell lung cancer clinically different? Br J Cancer. 2007;96:399 (author reply 400). Smouse JH, Cibas ES, Janne PA, Joshi VA, Zou KH, Lindeman NI. EGFR mutations are detected comparably in cytologic and surgical pathology specimens of nonsmall cell lung cancer. Cancer. 2009;117:67–72. Zhong Miao, Ma Xin, Sun Caijun, et al. MicroRNAs reduce tumor growth and contribute to enhance cytotoxicity induced by gefitinib in non-small cell lung cancer. Chem Biol Interact. 2010;184:431–8. Zhang T, Wang Q, Zhao D, Chen L. The oncogenetic role of microRNA-31 as a potential biomarker in oesophageal squamous cell carcinoma. Clin Sci. 2011;121:437–47. Iliopoulos D, Jaeger SA, Hirsch HA, Bulyk ML, Struhl K. STAT3 activation of miR-21 and miR-181b-1 via PTEN and CYLD are part of the epigenetic switch linking inflammation to cancer. Mol Cell. 2010;39:493–506. Krichevsky AM, Gabriely G. MiR-21: a small multi-faceted RNA. J Cell Mol Med. 2009;13:39–53.
Med Oncol (2014) 31:203 24. Seike M, Goto A, Okano T, Bowman ED, Schetter AJ, et al. MiR21 is an EGFR-regulated anti-apoptotic factor in lung cancer in never-smokers. Proc Natl Acad Sci USA. 2009;106:12085–90. 25. Tang Dongfang, Shen Yi, Wang Mingzhao, Yang Ronghua, Wang Zizong, et al. Identification of plasma microRNAs as novel noninvasive biomarkers for early detection of lung cancer. Eur J Cancer Prev. 2013;6:540–8. 26. Shen Yi, Tang Dongfang, Yao Ruyong, Wang Mingzhao, Wang Yongjie, et al. microRNA expression profiles associated with survival, disease progression, and response to gefitinib in completely resected non-small-cell lung cancer with EGFR mutation. Med Oncol. 2013;30:750–8. 27. Cho WC, Chow AS, Au JS. Restoration of tumour suppressor hsa-miR-145 inhibits cancer cell growth in lung adenocarcinoma
Page 9 of 9 203 patients with epidermal growth factor receptor mutation. Eur J Cancer. 2009;45:2197–206. 28. Ma Z, Qi J, Meng S, Wen B, Zhang J. Swimming exercise training-induced left ventricular hypertrophy involves microRNAs and synergistic regulation of the PI3K/AKT/mTOR signaling pathway. Eur J Appl Physiol. 2013;113:2473–86. 29. Larsson E, Fredlund Fuchs P, Heldin J, Barkefors I, Bondjers C, et al. Discovery of microvascular miRNAs using public gene expression data: miR-145 is expressed in pericytes and is a regulator of Fli1. Genome Med 2009; 11:108.01-12. 30. Cho WCS, Chow ASC, Au JSK. Restoration of tumour suppressor hsa-miR-145 inhibits cancer cell growth in lung adenocarcinoma patients with epidermal growth factor receptor mutation. Eur J Cancer. 2009;45:2197–209.
123
