EGFR mutation, smoking, and gender in advanced lung adenocarcinoma

Purpose In the current targeted therapy era, information on the effect of smoking in epidermal growth factor receptor (EGFR)-mutant lung cancer patients is scarce. Results In total, 11,678 adenocarcinoma patients were enrolled. Of these, 33.3% and 91.8% of male and female patients were non-smokers, respectively. An increased amount of smoking (P < 0.001 for trend), fewer smoke-free years (P < 0.001 for trend), and younger age of smoking initiation (P = 0.034 for trend) were all associated with significantly lower EGFR mutation rates. Smokers had a shorter median overall survival (OS) among both EGFR-mutant and EGFR-wild type patients (17.8 vs. 21.1 months, and 7.9 vs. 11.4 months respectively; both P < 0.001). Among patients with EGFR-mutant adenocarcinoma, younger smokers were associated with shorter OS (P = 0.047). In multivariate analysis, female gender was an independent prognostic factor for OS (hazard ratio: 0.86 [95% confidence interval {CI}: 0.80–0.93]; P < 0.001 in the EGFR-mutant group and 0.88 [95% CI: 0.81–0.96]; P = 0.004 in the EGFR-wild type group). Materials and Methods We reviewed the National Lung Cancer database (Taiwan) to assess the impact of smoking on the EGFR mutation rate and survival in advanced lung adenocarcinoma patients during 2011 and 2014 retrospectively. Conclusions Smoking was associated with lower incidence of EGFR mutation rate and reduced OS of advanced lung adenocarcinoma patients in a dose-dependent manner. In addition to EGFR mutation and smoking, gender also plays an important role in survival among these patients.


INTRODUCTION
Lung cancer is the leading cause of cancer-related death worldwide [1]. Cigarette smoking remains a major risk factor for lung cancer [2]. Moreover, smoking status influences the histological types [3], genotypes [4,5], and outcomes of lung cancer patients [6].
The epidermal growth factor receptor (EGFR) mutation is one of the most prevalent genetic alterations in lung cancer patients [4,7]. EGFR-tyrosine kinase inhibitors (TKIs) offer better efficacy and quality of life for lung cancer patients [8,9], and have hence emerged as an important frontline therapy for patients with EGFRmutant, non-small cell lung cancer [10].
Previous studies have identified smoking status as a poor prognostic factor in lung cancer [6,11]. However, in the current era of targeted therapy, limited information is available regarding whether the effects of smoking are similar in patients with different EGFR genotypes. Here, we assessed the National Lung Cancer database from the Taiwan Cancer Registry from 2011 to 2014 to investigate the impact of the smoking status on the EGFR mutation rate and the survival time of advanced lung adenocarcinoma patients.

Patients
Of 45,055 patients with newly diagnosed lung cancer in Taiwan from 2011 to 2014, 37,961 (84.3%) had detailed data for smoking history. Among these patients, 19,685 (51.9%) were non-smokers, and adenocarcinoma was the most common histological type (64.7%). In particular, 33.3% male adenocarcinoma and 91.8% female adenocarcinoma patients were non-smokers. Among the patients with lung adenocarcinoma, 14,654 (59.7%) exhibited advanced stage diseases, and 79.7% of these patients had available EGFR mutation data. A total of 11,678 patients were enrolled for survival analysis (Supplementary Figure 1).
The patient characteristics of the study population are shown in Table 1. In brief, 6,489 patients (55.6%) were female and 8,150 patients (69.8%) were non-smokers. The mean age was 65.9 ± 12.8 years. At the time of lung cancer diagnosis, 93.3% of the patients had stage IV disease and 76.8% of the patients had an ECOG performance status of 0-2. The overall EGFR mutation rate was 61.5%, and the median OS was 16.0 months (95% CI, 15.6-16.4).

Smoking status and EGFR mutation prevalence
The impact of smoking on the EGFR mutation rate is shown in

ABSTRACT
Purpose: In the current targeted therapy era, information on the effect of smoking in epidermal growth factor receptor (EGFR)-mutant lung cancer patients is scarce.
Results: In total, 11,678 adenocarcinoma patients were enrolled. Of these, 33.3% and 91.8% of male and female patients were non-smokers, respectively. An increased amount of smoking (P < 0.001 for trend), fewer smoke-free years (P < 0.001 for trend), and younger age of smoking initiation (P = 0.034 for trend) were all associated with significantly lower EGFR mutation rates. Smokers had a shorter median overall survival (OS) among both EGFR-mutant and EGFR-wild type patients (17.8 vs. 21.1 months, and 7.9 vs. 11.4 months respectively; both P < 0.001). Among patients with EGFR-mutant adenocarcinoma, younger smokers were associated with shorter OS (P = 0.047). In multivariate analysis, female gender was an independent prognostic factor for OS (   Oncotarget 98387 www.impactjournals.com/oncotarget and 35.5%, respectively. Even in patients smoking for < 15 pack-years, the EGFR mutation rate was still significantly lower than that of the non-smokers (aOR, 0.54 [95% CI, 0.46-0.64]; P < 0.001). A significant relationship was observed between an increased amount of smoking and the EGFR mutation rate decline (P < 0.001 for trend). Moreover, similar trends were observed for the number of smoke-free years and the age of smoking initiation, wherein a larger number of smoke-free years and a later age of smoking initiation were associated with a higher EGFR mutation rate (P < 0.001 and 0.034 for trend, respectively).

Smoking status and its impact on survival
The adverse effects of smoking on the survival duration are shown in Figure 1 and  Oncotarget 98388 www.impactjournals.com/oncotarget P = 0.203). In contrast, EGFR-wild type patients smoking for < 15 pack-years had a significantly worse outcome than non-smokers (HR, 1.24 [95% CI, 1.08-1.42], P = 0.002). A significant relationship was observed between an increased amount of smoking and a reduced survival duration, in both the EGFR-mutant and EGFR-wild type populations (P = 0.007 and 0.009 for trend, respectively).
With regard to the smoke-free years, we found that, among EGFR-mutant patients, current smokers had a significantly worse outcome as compared to non-smokers (HR: 1.33 [95% CI, 1.20-1.47]; P < 0.001). However, former smokers had similar survival durations as compared to non-smokers, even among those who had quit smoking for < 5 years. In contrast, among the EGFR-wild type patients, only those who had quit smoking for > 15 years had comparable survival durations as those in non-smokers.
With regard to the age of smoking initiation, EGFRmutant patients who had only started smoking after 30 years of age had similar outcomes as non-smokers (HR, 1.03 [95% CI, 0.98-1.26], P = 0.089). However, within the EGFR-wild type group, patients who initiated smoking in all age groups had worse outcomes.

Interaction between gender and smoking status
An association has been reported between smoking behaviors and gender, particularly among Asians [17], which indicated that the smoking rate is significantly higher in men than in women. Therefore, we further analyzed the impact of gender-smoking behaviors and their interaction on the survival of patients with different EGFR mutation status. Figure 2 outlines the Kaplan-Meier curves plotting OS with regard to gender, smoking status, and EGFR mutation status. In general, EGFR-mutant patients survived longer than EGFR-wild type patients. Among patients with the same EGFR genotypes and smoking behavior, females experienced a better outcome than males. In both EGFRmutant and EGFR-wild type populations, female nonsmokers had the best outcome, whereas male smokers had the worst outcome. Of note, female smokers had similar outcomes as male non-smokers (P = 0.901 in the EGFRmutant group and 0.681 in the EGFR-wild type group). In the multivariate analysis, female gender was found to be an independent prognostic factor in both the EGFR-mutant and EGFR-wild type groups (HR: 0.86 [95% CI, 0.80-0.93]; P < 0.001 in the EGFR-mutant group and HR: 0.88 [95% CI, 0.81-0.96]; P = 0.004 in the EGFR-wild type group). All these observations suggest that gender may play an independent role in determining lung cancer outcome.

DISCUSSION
An increasing number of studies have been assessing the effects of smoking, particularly on the survival of Oncotarget 98389 www.impactjournals.com/oncotarget lung cancer patients, and all these studies suggested that smoking is a poor prognostic factor [6,11,[13][14][15]. However, most of these studies were heterogeneous with regard to tumor stage and histology, and the data were not obtained from the era of EGFR-targeted therapy. In the present study, we enrolled 11,678 Taiwanese patients with a pure adenocarcinoma histology, advanced stage disease, detailed smoking data, and known EGFR mutation status. We assessed the different aspects of smoking behaviors, and our results suggest that smoking is not only an independent predictor of a lower EGFR mutation rate, but also reduces the survival duration in both EGFR-mutant and EGFR-wild type patients.
Smoking is known as a negative predictor for the EGFR mutation. In a meta-analysis of 26 studies including 3,688 NSCLC patients, Ren et al. confirmed that nonsmokers were associated with a significantly higher EGFR mutation rate [16]. Pham et al. further analyzed the influence of smoking in terms of pack-and smoke-free years on the prevalence of EGFR mutation, and found that the EGFR mutation rate was similar between non-smokers and patients who had smoked < 15 pack-years or those who had already quit smoking over the long-term [17]. In addition to smoking, both ethnicity and histology influenced the EGFR mutation rate [18]. Previous studies comprised patients with different ethnicities and histology, which may have led to different results. In the present study, we found that the EGFR mutation rate was significantly lower in smokers, even in those who had smoked for < 15 pack-years or those who had already quit smoking for > 15 years. Moreover, patients who had started smoking at a younger age also had a lower EGFR mutation rate.
The OS duration within our cohort was 16.0 months, which was longer than those of other nationwide Kaplan-Meier estimates of the duration of overall survival with regard to EGFR mutation status, smoking, and gender are shown. CI, confidence interval.
Oncotarget 98390 www.impactjournals.com/oncotarget studies [6,11,19]. This may be due to the inclusion of patients with pure adenocarcinoma, a greater number of non-smokers, a higher EGFR mutation rate among the Taiwanese patients, and the integration of EGFR-TKI treatment. In contrast, the OS duration of EGFR-wild type patients was 9.6 months, which was similar to the results of studies conducted prior to the EGFR-targeted therapy era. In the present study, smoking was associated with a worse OS in both EGFR-mutant and EGFR-wild type patients, and the survival duration of patients was inversely related to their levels of smoking. Moreover, we found a positive correlation between smoke-free years and outcomes, which could serve as valuable evidence for the recommendation of cigarette abstinence.
Large scale screening of oncogenic drivers of lung cancer has shown that patients with actionable genetic alterations would experience a better outcome [20,21]. EGFR mutation is the most common oncogenic driver in Asian NSCLC patients [4,22]. In Taiwan, the detection of EGFR mutations in advanced lung adenocarcinoma patients has become a routine procedure. Moreover, the National Health Insurance Administration has been reimbursing advanced EGFR-mutant lung adenocarcinoma patients receiving EGFR-TKI as the first-line of treatment since 2011; hence, every EGFR-mutant patient in the present study had an equal chance to receive EGFRtargeted therapy. Previous studies have identified smoking as a predictor of shorter progression-free survival among EGFR-mutant patients receiving EGFR-TKI, although its impact on OS was not consistent [23][24][25]. We found that smoking negatively affected OS in EGFR-mutant patients in a dose-dependent manner, despite the use of highly effective EGFR-TKI. As cigarette smoking negatively affected the outcomes to a greater extent in the EGFRwild type population, we suggest that the lower mutation burden in EGFR-mutant tumors [26] and the high efficacy of EGFR-TKI might at least in part attenuate the negative effects of smoking among EGFR-mutant patients.
Previous studies found that DNA adduct levels were inversely associated with the age at smoking initiation among former smokers, thus indicating that young smokers are more susceptible to DNA damage and persistence of genetic alterations than those who began smoking at an older age [27]. Our study found that younger smokers had a lower EGFR mutation rate, which may be due to the dilution of the effect of more smoking-related lung cancer [28]. Furthermore, DNA damage and persistence of genetic alterations may negatively affect the survival rates of young smokers among EGFR-mutant lung cancer patients, which may consequently have substantial implications on the need for preventing adolescent smoking, particularly in this era of targeted therapy.
A relationship between gender and smoking behaviors has been observed, particularly in the Asian population [12]. Gender is known to be an important factor in determining the pathological characteristics and driver mutations of lung cancer, as female gender is associated with a greater toxicity and efficacy of treatment [29]. Our results suggest that the harboring of EGFR mutations remains the most important prognostic factor. This may be related to the high efficacy of EGFR-TKI therapy, irrespective of whether it is administered as first-line or subsequent therapies [30]. However, among the patients with the same EGFR mutation and smoking status, females usually had a better outcome than males. Of note, female non-smokers had the best outcome, which suggests that these patients would benefit more from subsequent lung cancer treatment, such as chemotherapy [29,31]. Thus, our results imply that gender may influence the outcome of lung cancer patients through mechanisms independent of smoking.
There are 2 major limitations of the present study. First, the data were obtained from a registry database, and hence, the efficacy of a particular regimen cannot be determined. Nevertheless, we assessed OS as the primary endpoint, which is unambiguous and can be represented as a standard clinical outcome. Moreover, our patients had an equal chance for treatment owing to the fact that patients are reimbursed by the Taiwan National Health Insurance Administration for novel lung cancer therapies, such as EGFR-TKIs and pemetrexed, received over the study period. In addition, data from this period would not have been affected by the use of anaplastic lymphoma kinase (ALK) inhibitors, 3rd-generation EGFR-TKI, and immunotherapy; hence, this may accurately reflect the actual conditions in the EGFR-targeted therapy era. Second, our database did not register the detailed EGFR mutation spectrum. Based on our previous studies, primary resistant subtypes only accounted for a small portion of all EGFR mutations; [4,32] hence, it was less likely to influence the overall results.
In conclusion, smoking reduced both the EGFR mutation rate and survival duration in advanced lung adenocarcinoma patients in a dose-dependent manner, particularly among those who started smoking at a young age. In addition to EGFR mutation and smoking, gender also played an important role in the survival of these patients. This information may be valuable when recommending cigarette abstinence. Nevertheless, the impact of gender should be further elucidated in future studies.

Patient population
The detailed smoking history and EGFR mutation status of lung cancer patients have been routinely recorded since 2011 in the Taiwan Cancer Registry [33]. In the present study, we reviewed the information in the lung cancer database from 2011 to 2014. To be eligible for the study, patients were required to have cytologically or pathologically confirmed lung adenocarcinoma, stage www.impactjournals.com/oncotarget IIIB or IV disease, and available follow-up survival data. Patients were excluded if they had unclear smoking or EGFR mutation information.
Clinical data for analysis included patient age, gender, histological types, tumor stage, smoking status, the Eastern Cooperative Oncology Group performance status (ECOG PS), EGFR mutation status, and overall survival (OS). Non-smokers were defined as patients who had smoked < 100 cigarettes in their lifetime, whereas others were defined as smokers. Smokers were further stratified by smoking pack-years, smoke-free years, and age of smoking initiation. Lung cancer TNM (tumor, node, and metastases) staging was conducted according to the 7th edition of American Joint Committee on Cancer (AJCC) staging system [34]. This study was approved by the Institutional Review Board of Taiwan's National Health Research Institutes Research Ethics Committee (IRB No. CE17068B).

EGFR mutation testing
Several molecular tests are used for EGFR mutation analysis in Taiwan, including direct sequencing, protein nucleic acid-locked nucleic acid polymerase chain reaction (PNA-LNA PCR) clamp, scorpions amplification refractory mutation system (ARMS) (EGFR RGQ PCR Kit), Cobas EGFR Mutation Test, and matrixassisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) [32,35]. All are valid methods for EGFR mutation detection [36], although their use depended on the available laboratory facilities at each hospital. In Taiwan, the National Health Insurance Administration has been reimbursing patients receiving EGFR-TKI as the first-line treatment for advanced EGFRmutant lung adenocarcinoma since 2011.

Statistical methods
Multivariate logistic regression was performed to analyze the correlation of EGFR mutation with age, gender, stage, and smoking status. For survival analysis, the survival status was evaluated according to the National Death Certificate database, maintained by the Department of Statistics, Ministry of Health and Welfare, Taiwan, and was followed up until December 31, 2015. The survival duration for each patient was defined as the time from the date of initial diagnosis to the date of death, or the date of follow-up termination. The OS was estimated using the Kaplan-Meier method, whereas the between-group differences in the OS were assessed using a stratified logrank test. Hazard ratios (HRs) and the associated 95% confidence intervals (CIs) were estimated using the Cox proportional hazards model. The correlations between smoking intensity and EGFR mutation prevalence and survival were analyzed using the Cochran-Mantel-Haenszel test. All analyses were performed using SAS version 9.4 statistical software (SAS Institute, Cary, NC, USA).

CONFLICTS OF INTEREST
There are no conflicts of interest to declare.

FUNDING
The study was supported by a grant from the Health and Welfare Surcharge of Tobacco products provided by the Health Promotion Administration, Ministry of Health and Welfare, Taiwan (R.O.C.).