Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma

doi:10.4236/jct.2012.324056

Paper Menu >>

Journal Menu >>

Journal of Cancer Therapy, 2012, 3, 435-441

http://dx.doi.org/10.4236/jct.2012.324056 Published Online September 2012 (http://www.SciRP.org/journal/jct) 435

Smoking Cause Specific Lung Cancer—Evidence from

Non-Smoking Lung Adenocarcinoma

Ning Li, Bin Qu, Kang Shao, Zhaoli Chen, Fengwei Tan, Xiaogang Tan, Baozhong Li, Jiwen Wang,

Jie He*

Department of Thoracic Surgery, Cancer Hospital and Institute, Chinese Academy of Medical Sciences, Beijing, China.

Email: *[email protected]

Received July 13th, 2012; revised August 25th, 2012; accepted September 16th, 2012

ABSTRACT

Introduction: Smoking and non-smoking lung cancer have many differences in clinical feature. But those may be the

result of interference due to differences in pathological type, as most smoking patients suffer squmous cell lung cancer

and non-smokings tend to get adenocarcinoma. This study was conducted on the specific histological type—lung ade-

nocarcinoma—to avoid histological bias and to reveal the true effect of smoking. Methods: A total of 2222 patients

with lung adenocarcinoma confirmed by histological or cytological evidence were enrolled from January 1, 1999 to

December 31, 2004. Differences in clinical features and prognosis between non-smoking and smoking patients were

analyzed. Chi-square test was used for univariate comparisons. Univariate probability of survival was computed using

Kaplan-Meier estimate and compared to using the log-rank test. Cox proportional hazards regression analysis was used

to evaluate the risk of death. Results: There were 777 current smokers (34.96%), 197 former smokers (8.87%) and

1248 non-smoking patients (56.17%). 860 non-smoking patients (68.91%) were female, compared with 6.31% among

current smokers and 4.06% among former smokers (p < 0.001). Non-smoking patients had an earlier age at diagnosis (p

< 0.001) and a better response to chemotherapy (p < 0.001) compared to current smoking patients. Current smoking

correlated with lower cell differentiation (p < 0.001) and worse prognosis (p = 0.0024). After multivariate analysis,

smoking was identified as an independent negative prognostic factor (HR, 1.302; 95% CI, 1.011 - 1.6780, p = 0.041).

No difference in prognosis was observed according to smoking conditions in smoking patients. Conclusions: Signifi-

cent differences exist in clinical features and prognosis between non-smoking and smoking lung adenocarcinoma pa-

tients. There is a strong evidence that non-smoking lung adenocarcinoma should be regard as different disease.

Keywords: Lung Adenocarcinoma; Cigarette Smoke; Differentiation; Chemotherapy; Prognosis

1. Introduction

Lung cancer is the most lethal cancer in the world with

over 1 million deaths each year [1]. Cigarette smoking

was regarded as the most important carcinogenic factor

of lung cancer, which has been well established from

epidemiological evidence [2]. However, over 50 percent

of female and 15 percent of male lung cancer patients

never smoke cigarette [3]. The incidence of lung cancer

in non-smoker was even higher than several high inci-

dence cancers, such as: cervix, pancreas and prostate

cancer worldwide [3].

Smoking can not only increase the lung cancer inci-

dence but also affect the lung cancer patient’s clinical

feature and prognosis. Smoking lung cancer patients have

many differences in gender distribution, clinical features,

pathology result and molecular mechanism comparing to

non-smoking lung cancer [4,5]. Smoking lung cancer

patients tend to be diagnosed at older age, present with

early stage, usually be afflicted with squamous cell lung

cancer, and have a better prognosis [6-8]. Non-smoking

lung cancer have a higher rate of mutation in Epidermal

growth factor receptor (EGFR) and a better response to

the EGFR Tyrosine Kinase Inhibitor: Gifitinib and Er-

lotinib [9,10], while have less mutations in P53 com-

pared with smoking lung cancer [4,11]. Therefore, many

scholarssuggested that lung cancer in never smokers is a

“different” disease, with a different aetiology and a dif-

ferent natural history [5,6].

A small defect exists in the evidences of this hypothe-

sis. Previous researches about clinical features and sur-

vival analysis mostly focus on all categories of lung can-

cer [7] or non-small cell lung cancer (NSCLC) [6,8,12].

Over 70 percent smoking patients suffer from squamous

cell lung cancer and most non-smoking patients suffer

from lung adenocarcinoma [13]. Many distinctions exist

in clinical features between subtypes of lung cancer, and

*Corresponding autho

Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma

436

histology, such as squamous cell or adenocarcinoma

have been recognized as a predictive factor for response

to chemotherapy and prognosis [14]. So, differences be-

tween smoking and non-smoking lung cancer might be

confounded by the differences between squamous cell

cancer and adenocarcinoma.

To avoid any bias, we performed a retrospective study

of lung adenocarcinoma, which contains the most part of

non-smoking patients. We compared all detectable vari-

ance in clinical features and prognosis between non-

smoking and smoking patients, to investigate the true

effect of smoking to the lung adenocarcinoma.

2. Patients and Methods

Current respective study was conducted in the Cancer

hospital, Chinese Academy of Medical Sciences (CAMS).

Patients had primary lung adenocarcinoma diagnosed

between 1 January 199 9 a n d 3 1 December 2004.

All patients diagnosed consecutively with primary

lung adenocarcinoma, which was confirmed by histo-

logical or cytological examination. To avoid the influ-

ence from different histological subtype, adenosquamous

carcinoma and bronchiolo-alveolar carcinoma were ex-

cluded. To avoid the influence form different ethnic

groups, only Chinese patients were enrolled. Patients

were excluded if smoking history was not available in

case record. Individual case records were reviewed. In

addition to smoke history, age, gender, lung cancer fa-

mily history, American Joint Committee on Cancer

(AJCC) stage at diagnosis and type of treatment were

recorded. When surgery was performed as initial treat-

ment, the metastasis of lymph nodes (N stage), differen-

tiation of adenocarcinoma cells were collected as well.

When chemotherapy was performed as initial treatment

in advanced patients (Stage IIIB and IV), the clinical

response after four cycles to first-line chemotherapy were

also recorded as four levels: Complete Response (CR),

Partial Response (PR), Stable Disease (SD) and Progr ess

Disease (PD), according to RECIST critical. Survival

data were obtained from follow-up group in thoracic de-

partment cancer hospital CAMS. All patients were noti-

fied and understand that their clinical information might

be used for clinical research and analysis, before they got

the therapy. And all included in this study signed the

authorizations form. The study was approved by the me-

dical ethics committee of Cancer Institute and Hospital,

CAMS.

Smoking history contains smok ing condition s, such as:

how many years of smoking, how many cigarettes smoked

per day, and how many years since quitting smoke. Pa-

tients who were smokers during diagnosis or stop smok-

ing less than 1 year were collected in “current smoker”

group [5]. Patients who quit smoking more than one year

were classified as “former smoker”. Patients who have a

lifetime exposure of less than 100 cigarettes or never

smoked in the past were termed as “never smoker”.

According to smoking habits, chi-square test was used

for univariate comparisons of patients, disease and treat-

ment-related characteristics. Cox proportional hazards

regression analysis w as used to evaluate the risk of death

between current, former and never smoking patients,

while adjusting for othe r prognostic factors in cluding age

at diagnosis, gender, lung cancer family history, disease

stage at diagnosis and initial treatment. Univariate prob-

ability of survival was computed using Kaplan-Meier

estimate and compared using the log-rank test. P-Value

of 0.05 was considered significant for all analysis.

3. Results

A total of 2222 patients were enrolled and met study se-

lection criteria. 777 patients (34.96%) were current smo-

kers, 1248 patients (56.17%) were never smokers, and

197 patients (8.87%) who quit smoking for at least one

year were termed as “former smokers”. Demographic

and clinical features of each group were listed in Table

1. The overwhelming majority of current smokers and

former smokers were males, 93.69% in current smokers

and 95.94% in former smokers. On the other hand, only

31.09% patients in never smoker group were males.

Non-smoking patients were diagnosed at median age of

56 years old, which was younger than current and former

smokers (p < 0.001). Former smokers tend to present

with earlier disease than current and never smokers (p =

0.016). Operations were preformed in 503 current smo-

kers (64.74%), 153 former smokers (77.66%) and 794

never smokers (63.62%). Differentiation in adenocarci-

noma cells was evaluated in the tissue samples resects in

operations. More proportion of poor differentiation was

found in current smoking patients (56.39%) than former

(47.37%) and never (35.87%) smokers (p < 0.001). There

was a weak significant difference in N2 lymph nodes

metastasis (p = 0.146) between three groups. Complete

response and partial response were seen in 165 (41.88%)

non-smoking patients, but only in 41(23.56%) current

and 8 (26.67%) former smoking patients (p < 0.001).

Overall survival is shown in Figure 1. The 3-year

overall survival rates for current smoker, former smoker

and never smoker were 23.63%, 28.15% and 33.55%.

The 5-year overall survival rate were 19.85%, 23.06%

and 22.93%, respectively (p = 0.0024). Interestingly,

former smokers had similar survival as never smokers,

and much better than current smokers in patients with

early stage (stage I and II) (Figure 2).

As smoking patients had different smoking conditio ns,

additional analysis were performed out within the current

smoking group to examine the impact of smoking co ndi-

tions on prognosis of smoking patients. There were no

Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma

437

Table 1. The clinical features of lung adenocarcinoma patients.

Current Smokers (n = 777) Former Smokers (n = 197) Never Smokers (n = 1248)

Variable

No. % No. % No. %

p value

Gander

Male 728 93.69% 189 95.94% 388 31.09%

Female 49 6.31% 8 4.06% 860 68.91%

p < 0.001

Age at diagnosis

Median 59 66 56

Range 23 - 86 40 - 83 20 - 84

p < 0.001

AJCC stage at diagnosis (n = 2144)

I 200 26.74% 57 31.32% 278 22.90%

II 64 8.56% 17 9.34% 108 8.90%

III 176 23.53% 54 29.67% 342 28.17%

IV 308 41.18% 54 29.67% 486 40.03%

p = 0.016

Treatment

Operation 503 64.74% 153 77.66% 794 63.62%

Chemotherapy/Others 274 35.26% 44 22.34% 454 36.38%

p = 0.001

Differentiation of adenocarcinoma cells (n = 1323)

Low differentiation 256 56.39% 63 47.37% 264 35.87%

Middle differentiatio n 166 36.56% 57 42.86% 404 54.89%

High differentiation 32 7.05% 13 9.77% 68 9.24%

p < 0.001

Metastasis of lymph nodes (n = 1298)

N0 205 46.49% 59 45.74% 288 39.56%

N1 69 15.65% 17 13.18% 117 16.07%

N2 167 37.87% 53 41.09% 323 44.37%

p = 0.146

significant difference in prognosis of smokers according

to years of smoking, number of cigarettes smoked per

day and the smoking index (Figure 3).

Table 2 summarized the hazard ratios and significance

of the impact of clinical features. The risk of dying for

current smokers remained higher compared with never-

smokers after adjusting for gender, clinical stage, treat-

ment and family history (HR: 1.302; 95% CI: 1.011 -

1.678; p = 0.041). Current smoking status was an inde-

pendent variable for poor prognosis in lung adenocarci-

mona.

4. Discussion

Non-smoking lung related cancer has many differences

compared with smoking related lung cancer, from etio-

logical factors to biological behavior. Non-smoking lung

cancer might result from other risk factors, such as: pre-

existing lung disease, endocrine factors, family history,

radiation, air pollution and environmental tobacco smok-

ing [13]. Cancers induced by different carcinogenic

agents might be variables in biological behavior. With

different ability of proliferation, differentiation, migra-

Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma

438

Table 2. Cox proportional hazards regression analysis of factors associated with survival in patients with lung adenocarci-

noma.

Variable Cate

ories/UniteHazard Ratio 95% CI

value

Never smoke

Former smoker 1.289 0.894 - 1.858 0.175

Cigarette smoke

Current smoker 1.302 1.011 - 1.678 0.041

Female 1

Gender

Male 1.129 0.875 - 1.457 0.352

Negative 1

Family histor y

Positive 0.635 0.699 - 1.244 0.635

Operation 1

Treatment

Others 1.410 0 .925 - 2.151 0.110

Stage I 0.132 0.081 - 0.215 <0.001

Stage II 0.419 0 .246 - 0.713 0.001

Stage III 0.752 0 .490 - 1.154 0.192

TNM stage

Stage IV 1

Figure 1. Kaplan-Meier survival curves stratified by smok-

ing status in patients with lung adenocarcinoma.

tion and invasion, even though cancer cells had similar

appearance, they could be regarded as “different cancer”.

It is a feasible method to investigate the biological fea-

tures of non-smoking related lung cancer, to analyze the

clinical features of patients, and then compare with

smoking patients. Differences in age, gender, initial stage,

response to chemotherapy and prognosis between non-

smoking and smoking patients had been found in previ-

ous studies [5,6].

However, several influencing factors should not be

ignored. Distinctions in pathology have been established

by previous studies. Non-smoking lung cancer patients

tend to suffer from adenocarcinoma. Patients with lung

adenocarcinoma have many differences in gender, age at

diagnosis, response to chemotherapy and prognosis com-

pared to those with squamous cell lung cancer [14]. The

ethnic difference is another important factor. Asian fe-

male had much lower smoking rate and higher lung can-

cer incidence compared to American and European fe-

male, and they are more sensitive to the EGFR Tyrosine

Kinase Inhibitor [5]. Japanese ethnicity is independent

favorable prognostic factors when compared with Cauca-

sian ethnicity in NSCLC [15]. Another factor is the

choice of the treatment: each institution or oncologist

chose method mainly based on local policies, cost or con-

venience of the therapeutic schedule, which could affect

the prognosis of the patients. Therefore, we conducted

current study in a specific histological pathology, in a

specific ethnic group and in a specific institute, to avoid

those influences.

Non-smoking patients were diagnosed earlier at me-

dian age of 56 years old compared with current and for-

mer smoking patients. Former studies from Asia coun-

tries had similar results [6], while studies from United

States and Europe did not [5]. Those facts indicate that

non-smoking lung cancer in Asia might not be resulted

from environment tobacco smoking, as patients do not

have enough exposure to tobacco smoking [16]. Diffe-

rences between those two cohorts, such as genetic factors

Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma 439

Figure 2. Kaplan-Meier survival curves stratified by smok-

ing status in patients with Stages I and II, III, and IV.

and environment factors are important in lung cancer

carcinogenesis. Most patients in never smoking group are

female. Similar results were obtained in many previous

studies [6], which indicated the female is more sensitive

to carcinogen among non-smokers. Direct evidence comes

from incidence investigation: incidence of female patients

with non-smoking lung cancer is 14.4 - 20.8/100000,

while 4.8 - 13.7/100000 in male [17,18].

Here we reported that histological differentiation of

cancer cells is different among patients with diverse

Figure 3. Kaplan-Meier survival curves stratified by smok-

ing conditions in smoking patients.

smoking conditions. Non-smoking patients usually had a

better differentiation than smokers, which was a direct

pathomorphological evidence for “different cancer”. The

smoke of cigarette contains more than 4500 chemical

compounds. Many of them affect the cell differentiation

in vitro and in vivo [19], which could explain poorer dif-

ferentia tion in smoking pati ents.

We demonstrated patients in never smoker group had

better response to front line chemotherapy than those in

the other two groups. Tsao et al. had similar results. They

analyzed 873 advanced non small cell lung cancer pa-

tients who received chemotherapy, and found never-

Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma

440

smokers had higher response rates and lower rates of

progressive disease than former and current smokers [8].

As chemotherapy drugs work directly on the cancer cells,

different responses indicate different intracellular mecha-

nisms [20]. Previous study shows, compared with people

who never smoke, people with lung adenocarcinoma who

smoked or stopped smoking less than 25 years ago have

significant less EGFR mutations in exons 19 and 21 [21].

A study of comparing gene expression profiles in never

smokers and smokers with lung adenocarcinoma by

Powell et al. reported that there might be different path-

ways of cell transformation and tumor formation in two

groups, with many differences in gene expression. [22]

Those are strong evidences for the “different cancers”.

Surviving time of patients is considered as the most

important characteristics in cancer research. Different

subtypes of cancer usually have dissimilar prognosis,

such as: lung squamous cell cancer and adenocarcinoma;

gastric signet-ring cell carcinoma and adenocarcinoma;

hepatic cellular cancer and hepatobiliary cancer. Non-

smoking lung adenocarcinoma patients had a better prog-

nosis than current and former smokers. Interestingly,

former smoking patients had similar survival with non-

smoking patients and much better than current smoking

patients in the early stage, which is a good message for

patients who quit smoking. Study from Nordquist et al.

reported 5 years survival were 16% for current smokers

and 23% for never-smokers in 654 patients with lung

adenocarcinoma [23]. Similar results were reported by

Tammemagi et al., they studied 1155 lung cancer pa-

tients and found cigarette smoking was an important in-

dependent pred ictor of shor tened lung can cer surviv al [7].

Bryant et al. also found never smoker patients with early-

stage cancer have a significantly b etter survival rate than

smokers [24]. As expected, in current study, we found

smoking conditions such as the number of cigarettes

smoked per day, years of smoking and the smoking index

do not affect the prognosis. The comorbidity of smoking

such as chronic obstructive pulmonary disease, hyperten-

sion, atherosclerosis and aortic aneurysm are more seri-

ous in heavy smoker; which may indicate that the diffe-

rence in prognosis was not because of the comorbidities

but the nature of cancer. We can infer that lung cancer

caused by smoking is the same type of cancer, which has

no correlation in prognosis with cigarette smoke burden.

With difference in prognosis, non-smoking lung adeno-

carcinoma could be considered as a different cancer.

In summary, we demonstrate significant differences

between non-smoking and smoking patients with lung

adenocarcinoma in age at diagnosis, gender, initial stage

at diagnosis, cell differentiation, response to chemothe-

rapy and prognosis. Furthermore, we suggest non-smok-

ing and smoking lung adenocarcinoma could be consi-

dered as different cancers with many distinctions in clin i-

cal features.

REFERENCES

[1] A. Jemal, T. Murray, E. Ward, et al., “Cancer Statistics,

2005,” CA—A Cancer Journal for Clinicians, Vol. 55,

No. 1, 2005, pp. 10-30. doi:10.3322/canjclin.55.1.10

[2] E. L. Wynder and E. A. Graham, “Landmark Article May

27, 1950: Tobacco Smoking as a Possible Etiologic Fac-

tor in Bronchiogenic Carcinoma. A Study of Six Hundred

and Eighty-Four Proved Cases,” JAMA, Vol. 253, No. 20,

1985, pp. 2986-2994.

doi:10.1001/jama.1985.03350440064033

[3] D. M. Parkin, F. Bray, J. Ferlay and P. Pisani, “Global

Cancer Statistics, 2002,” CA—A Cancer Journal for Cli-

nicians, Vol. 55, No. 2, 2005, pp. 74-108.

doi:10.3322/canjclin.55.2.74

[4] R. Gealy, L. Zhang, J. M. Siegfried, J. D. Luketich and P.

Keohavong, “Comparison of Mutations in the p53 and

K-Ras Genes in Lung Carcinomas from Smoking and

Nonsmoking Women,” Cancer Epidemiology, Biomar-

kers & Prevention, Vol. 8, No. 4, 1999, pp. 297-302.

[5] S. Sun, J. H. Schiller and A. F. Gazdar, “Lung Cancer in

Never Smokers—A Different Disease,” Nature Reviews

Cancer, Vol. 7, 2007, pp. 778-790. doi:10.1038/nrc2190

[6] C. K. Toh, F. Gao, W. T. Lim, et al., “Never-Smokers

with Lung Cancer: Epidemiologic Evidence of a Distinct

Disease Entity,” Journal of Clinical Oncology, Vol. 24,

No. 34, 2006, pp. 2245-2251.

doi:10.1200/JCO.2005.04.8033

[7] C. M. Tammemagi, C. Neslund-Dudas, M. Simoff and P.

Kvale, “Smoking and Lung Cancer Survival: The Role of

Comorbidity and Treatment,” Chest, Vol. 125, No. 1,

2004, pp. 27-37. doi:10.1378/chest.125.1.27

[8] A. S. Tsao, D. Liu, J. J. Lee, M. Spitz and W. K. Hong,

“Smoking Affects Treatment Outcome in Patients with

Advanced Nonsmall Cell Lung Cancer,” Cancer, Vol.

106, No. 11, 2006, pp. 2428-2436.

doi:10.1002/cncr.21884

[9] T. J. Lynch, D. W. Bell, R. Sordella, et al., “Activating

Mutations in the Epidermal Growth Factor Receptor Un-

derlying Responsiveness of Non-Small-Cell Lung Cancer

to Gefitinib,” New England Journal of Medicine, Vol.

350, No. 21, 2004, pp. 2129-2139.

doi:10.1056/NEJMoa040938

[10] S. W. Han, T. Y. Kim, P. G. Hwang, et al., “Predictive

and Prognostic Impact of Epidermal Growth Factor Re-

ceptor Mutation in Non-Small-Cell Lung Cancer Patients

Treated with Gefitinib,” Journal of Clinical Oncology,

Vol. 23, No. 16, 2005, pp. 2493-2501.

doi:10.1200/JCO.2005.01.388

[11] S. N. Rodin and A. S. Rodin, “On the Excess of G --> T

Transversions in the p53 Gene in Lung Cancer Cell Lines.

Reply to Pfeifer and Hainaut,” Mutation Research, Vol.

545, No. 1-2, 2004, pp. 141-147.

doi:10.1016/j.mrfmmm.2003.09.008

[12] C. K. Toh, E. H. Wong, W. T. Lim, et al., “The Impact of

Smoking Status on the Behavior and Survival Outcome of

Smoking Cause Specific Lung Cancer—Evidence from Non-Smoking Lung Adenocarcinoma

441

Patients with Advanced Non-Small Cell Lung Cancer: A

Retrospective Analysis,” Chest, Vol. 126, No. 6, 2004, pp.

1750-1756. doi:10.1378/chest.126.6.1750

[13] J. Subramanian and R. Govindan, “Lung Cancer in Never

Smokers: A Review,” Journal of Clinical Oncology, Vol.

25, No. 5, 2007, pp. 561-570.

doi:10.1200/JCO.2006.06.8015

[14] G. Selvaggi and G. V. Scagliotti, “Histologic Subtype in

NSCLC: Does It Matter?” Oncology, Vol. 23, No. 13,

2009, pp. 1133-1140.

[15] T. Kawaguchi, A. Matsumura, S. Fukai, et al., “Japanese

Ethnicity Compared with Caucasian Ethnicity and Never-

Smoking Status Are Independent Favorable Prognostic

Factors for Overall Survival in Non-small Cell Lung

Cancer: A Collaborative Epidemiologic Study of the Na-

tional Hospital Organization Study Group for Lung Can-

cer (NHSGLC) in Japan and a Southern California Re-

gional Cancer Registry Databases,” Journal of Thoracic

Oncology, Vol. 5, No. 7, 2010, pp. 1001-1010.

[16] L. A. Loeb, V. L. Ernster, K. E. Warner, J. Abbotts and J.

Laszlo, “Smoking and Lung Cancer: An Overview,” Can-

cer Research, Vol. 44, No. 12, 1984, pp. 5940-5958.

[17] H. A. Wakelee, E. T. Chang, S. L. Gomez, et al., “Lung

Cancer Incidence in Never Smokers,” Journal of Clinical

Oncology, Vol. 25, No. 5, 2007, pp. 472-478.

doi:10.1200/JCO.2006.07.2983

[18] L. C. Koo and J. H. Ho, “Worldwide Epidemiological

Patterns of Lung Cancer in Nonsmokers,” International

Journal of Epidemiology, Vol. 19, Suppl. 1, 1990, pp.

S14-S23.

[19] L. S. Wistuba II, C. Behrens, et al., “Molecular Damage

in the Bronchial Epithelium of Current and Former Smo-

kers,” Journal of the National Cancer Institute, Vol. 89,

No. 18, 1997, pp. 1366-1373. doi:10.1093/jnci/89.18.1366

[20] L. Mao, J. S. Lee, J. M. Kurie, et al., “Clonal Genetic

Alterations in the Lungs of Current and Former Smok-

ers,” Journal of the National Cancer Institute, Vol. 89,

No. 12, 1997, pp. 857-862. doi:10.1093/jnci/89.12.857

[21] D. Pham, M. G. Kris, G. J. Riely, et al., “Use of Ciga-

rette-Smoking History to Estimate the Likelihood of Mu-

tations in Epidermal Growth Factor Receptor Gene Exons

19 and 21 in Lung Adenocarcinomas,” Journal of Clini-

cal Oncology, Vol. 24, 2006, pp. 1700-1704.

doi:10.1200/JCO.2005.04.3224

[22] C. A. Powell, A. Spira, A. Derti, et al., “Gene Expression

in Lung Adenocarcinomas of Smokers and Nonsmokers,”

American Journal of Respiratory Cell and Molecular Bi-

ology, Vol. 29, No. 2, 2003, pp. 157-162.

doi:10.1165/rcmb.2002-0183RC

[23] L. T. Nordquist, G. R. Simon, A. Cantor, W. M. Alberts

and G. Bepler, “Improved Survival in Never-Smokers vs

Current Smokers with Primary Adenocarcinoma of the

Lung,” Chest, Vol. 126, No. 2, 2004, pp. 347-351.

doi:10.1378/chest.126.2.347

[24] A. Bryant and R. J. Cerfolio, “Differences in Epidemiol-

ogy, Histology, and Survival between Cigarette Smokers

and Never-Smokers Who Develop Non-Small Cell Lung

Cancer,” Chest, Vol. 132, No. 1, 2007, pp. 185-192.

doi:10.1378/chest.07-0442