BRIEF REPORT

Characteristics of Smoking Patients with Lung Cancer

with Emphysematous Bullae
Eiji Iwama, MD, PhD,a,b Isamu Okamoto, MD, PhD,b,* Hidetake Yabuuchi, MD, PhD,c
Koichi Takayama, MD, PhD,d Taishi Harada, MD, PhD,b Yoshio Matsuo, MD, PhD,e
Shoji Tokunaga, PhD,f Eishi Baba, MD, PhD,a Yoichi Nakanishi, MD, PhDb
a
Faculty of Medical Sciences, Department of Comprehensive Clinical Oncology, Kyushu University, Fukuoka, Japan
b
Research Institute for Diseases of the Chest, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
c
Department of Health Sciences, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
d
Department of Pulmonary Medicine, Kyoto Prefectural University of Medicine, Kyoto, Japan
e
Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
f
Medical Information Center, Kyushu University Hospital, Fukuoka, Japan

Received 25 February 2016; revised 28 April 2016; accepted 29 April 2016

Available online - 7 May 2016

ABSTRACT
2016 International Association for the Study of Lung

Introduction: Emphysema is thought to be a risk factor for
lung cancer in smokers, with emphysematous bullae (EBs),
which are believed to have the potential to give rise to lung
cancer. The clinical characteristics of patients with lung
cancer with EBs have remained incompletely defined,
however.
Methods: A total of 488 patients with primary lung cancer
with or without EBs as detected by computed tomography
were studied retrospectively, and the regional relationship
between EBs and the primary cancer was evaluated.
Results: EBs were detected in 45 of the 488 patients with
lung cancer (9.2%) (in 45 of 339 smokers [13.3%] versus in
0 of 149 never-smokers [0%]). The frequency of lung cancer
in an upper lobe was significantly higher in smokers with
EBs than in those without EBs (71.1% versus 47.3%, p ¼
0.0107). The lobar site of primary lung cancer in smokers
with EBs was significantly associated with that of the EBs
(p < 0.0001). Most primary lung cancers (86.7%) in such
patients were found in the area adjoining EBs. Smoking
patients with lung cancer with EBs were significantly
younger (63.6 versus 67.7 years, p ¼ 0.0179) and had tu-
mors with a lower frequency of epidermal growth factor
gene (EGFR) mutations (3.8% versus 24.2%, p ¼ 0.0184)
compared with those without EBs.
Conclusions: The clinical characteristics of smoking pa-
tients with lung cancer differ according to the absence or
presence of EBs, with patients with EBs being potentially
more susceptible to the carcinogenic effects of cigarette
smoke. Further analysis of genetic alterations is warranted
to elucidate the mechanism of carcinogenesis for lung
cancer associated with EBs.
Cancer. Published by Elsevier Inc. All rights reserved.
Keywords: Emphysema; Bulla; Lung cancer; Smoking; EGFR
mutation
Introduction
Cigarette smoking is a major risk factor for the
development of lung cancer.1,2 Emphysema is detectable
by computed tomography (CT) in approximately 30% of
smokers, and the presence of emphysema as detected by
CT is associated with an increased risk for lung cancer in
smokers.3–5 Emphysematous bullae (EBs), which are
defined as bullous destructive lesions of the lung pa-
renchyma on a background of emphysema,6 are closely
associated with cigarette smoking and are thought to
have the potential to give rise to lung cancer.7–9 The
clinical characteristics of patients with lung cancer with
EBs remain unknown, however. We have now evaluated
the presence of EBs in patients with lung cancer as well
as the characteristics of lung cancer with EBs.
*Corresponding author.
Disclosure: The authors declare no conflict of interest.
Address for correspondence: Isamu Okamoto, MD, PhD, Research
Institute for Diseases of the Chest, Graduate School of Medical
Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka
812-8582, Japan. E-mail: okamotoi@kokyu.med.kyushu-u.ac.jp
ª 2016 International Association for the Study of Lung Cancer.
Published by Elsevier Inc. All rights reserved.
ISSN: 1556-0864
http://dx.doi.org/10.1016/j.jtho.2016.04.024

Journal of Thoracic Oncology Vol. 11 No. 9: 1586-1590

September 2016
Materials and Methods
Patients
Clinical records of 499 patients with primary lung
cancer who were treated at Kyushu University Hospital
between January 2009 and December 2011 were studied
retrospectively. Chest CT scans were reevaluated inde-
pendently by two radiologists (H.Y. and Y.M.) and one
pulmonologist (T.H.). We defined a bulla as an airspace
that had a diameter greater than 1 cm and was sharply
demarcated by a smooth wall with a thickness of 1 mm
or less, as previously described.6 Eleven patients were
excluded from the analysis either because their CT scans
were not available or because their primary cancers
were undetectable on CT scans as a result of pleural
effusion. The remaining 488 patients were analyzed in
this study. A total of 54 patients were initially selected
by the three observers as potential cases of lung cancer
with EBs, but this number was reduced to 45 definite
cases on further analysis by H. Y. and T. H. Given that
interstitial pneumonia has been implicated as a risk
factor for lung cancer,10 individuals with interstitial
pneumonia (23 patients) or with combined pulmonary
fibrosis and emphysema (three patients) were excluded
from the category of patients with EBs. Individuals with
solitary airspaces (seven patients) were also excluded
from this category. This study was conducted in accor-
dance with the amended Declaration of Helsinki and was
approved by the Ethics Committee of Kyushu University.
Evaluation of EB Volume
For determination of the volume of EBs, the volume
of areas of low attenuation on CT scans was calculated
with a threshold of –910 Hounsfield units, as previously
described,11 and the volume of such areas with a diam-
eter greater than 1 cm was then evaluated with cluster
analysis. The quantitative assessment was performed
with the use of a workstation (Synapse Vincent, Fuji
Film, Tokyo, Japan). A 5-mm collimation was used for
reconstitution of the volume data.
Statistical Analysis
Fisher’s exact test or the chi-square test was used to
compare categorical variables. Student’s unpaired t test
was performed to compare continuous variables. Overall
survival was analyzed by the Kaplan-Meier method, and
differences were assessed with the log-rank test. A p value
less than 0.05 was considered statistically significant.
Results
EB Frequency in Patients with Lung Cancer
EBs were detected in 45 of the 339 smokers
(13.3%) and in none of the 149 never-smokers among
Lung Cancer with Emphysematous Bullae 1587
the 488 patients with lung cancer, yielding an overall
prevalence of EBs of 9.2% (Table 1). The frequency of
EBs differed significantly between smokers and never-
smokers (p < 0.0001).
Characteristics of Smoking Patients with Lung
Cancer with or without EBs
Among the 339 smoking patients with lung cancer,
there was no significant difference in sex; extent of
smoking; prevalence of airway obstruction; tumor his-
tologic diagnosis; tumor, node, and metastasis stage; or
treatment for lung cancer between those with and
without EBs (Table 2). Smoking patients with lung can-
cer with EBs were significantly younger than those
without EBs (63.6 versus 67.7 years old, p ¼ 0.0179).
Epidermal growth factor receptor gene (EGFR) mutation
analysis was performed for 183 lung cancers of smokers,
with 39 cases (21.3%) found to be positive for such
mutations. The frequency of tumors positive for EGFR
mutations was significantly lower in the patients with
EBs than in those without EBs (one of 26 patients [3.8%]
versus 38 of 157 patients [24.2%], p ¼ 0.0184). The sites
of primary lung cancer in smokers were categorized as
upper lobe, middle lobe (lingular segment), or lower
lobe, with primary cancer sites in the upper lobe being
more frequent for patients with EBs than for those
without EBs (71.1% versus 47.3%, p ¼ 0.0107). The
present study included 109 patients with advanced lung
cancer who were smokers and treated with chemo-
therapy. Overall survival was slightly worse in patients
with EBs than in those without EBs (median of 311.5
versus 423.0 days, p ¼ 0.129) (Fig. 1).
Relationship between Sites of Primary Lung
Cancer and EBs
We evaluated the sites of EBs in the 45 smoking
patients with lung cancer with this condition. EBs were
localized to upper or lower lobes in 33 patients (73.3%)
and 12 patients (26.7%), respectively. Examination of
the relationship between the sites of primary lung cancer
and EBs in these patients revealed that lung cancer was
present in the upper lobe of 29 of the 33 patients with
Table 1. Frequency of EBs in Patients with Lung Cancer
according to Smoking Habit
Smokers Never-smokers
Total (n ¼ 339 (n ¼ 149 p
(N ¼ 488) [69.5%]) [30.5%]) Valuea
With EBs 45 (9.2%) 45 (13.3%) 0 (0%) <0.0001
Without 443 (90.8%) 294 (86.7%) 149 (100%)
EBs
a
Fisher’s exact test.
EBs, emphysematous bullae.
1588 Iwama et al Journal of Thoracic Oncology Vol. 11 No. 9

Table 2. Characteristics of Smoking Patients with Lung Cancer according to the Absence or Presence of EBs
Total With EBs Without EBs
Characteristic (n ¼ 339) (n ¼ 45 [13.3%]) (n ¼ 294 [86.7%]) p Value
Mean age, y (95% CI)
67.2 (66.0–68.3) 63.6 (60.5–66.8) 67.7 (66.5–68.9) 0.0179a
Sex
Male 298 (87.9%) 43 (95.6%) 255 (86.7%) 0.137b
Female 41 (12.1%) 2 (4.4%) 39 (13.3%)
Mean pack-years (95% CI)c
52.4 (49.0–55.8) 53.2 (44.0–62.4) 52.3 (48.7–55.9) 0.861a
FEV1.0d
<70% 136 (48.4%) 18 (45.0%) 118 (49.0%) 0.642e
70% 145 (51.6%) 22 (55.0%) 123 (51.0%)
Tumor histologic diagnosis
Ad 174 (51.3%) 23 (51.1%) 151 (51.4%) 0.655e
Sq 94 (27.7%) 13 (28.9%) 81 (27.6%)
Small cell 50 (14.8%) 4 (8.9%) 46 (15.7%)
Large cell 9 (2.7%) 2 (4.4%) 7 (2.4%)
NOS 8 (2.4%) 2 (4.4%) 6 (2.0%)
Unknown 4 (1.2%) 1 (2.2%) 3 (1.0%)
TNM stage
I 121 (35.7%) 16 (35.6%) 105 (35.7%) 0.921e
II 23 (6.8%) 2 (4.4%) 21 (7.1%)
III 70 (20.7%) 10 (22.2%) 60 (20.4%)
IV 125 (36.9%) 17 (37.8%) 108 (36.7%)
Treatment for lung cancerf
Surgery 117 (34.6%) 18 (40.0%) 99 (33.8%) 0.269e
Radical RTxg 77 (22.8%) 10 (22.2%) 67 (22.9%)
Chemotherapy 109 (32.3%) 16 (35.6%) 93 (31.7%)
BSC 35 (10.4%) 1 (2.2%) 34 (11.6%)
EGFR mutationsh
Positive 39 (21.3%) 1 (3.8%) 38 (24.2%) 0.0184b
Negative 144 (78.7%) 25 (96.2%) 119 (75.8%)
Primary cancer sitei
Upper lobe 163 (50.6%) 32 (71.1%) 131 (47.3%) 0.0107e
Middle lobe 22 (6.8%) 1 (2.2%) 21 (7.6%)
Lower lobe 137 (42.6%) 12 (26.7%) 125 (45.1%)
a
Unpaired t test.
b
Fisher’s exact test.
c
Smoking status (pack-years) was evaluable in 337 patients.
d
Pulmonary function test was evaluable in 281 patients.
e
Chi-square test.
f
Information on treatment for lung cancer was available for 338 patients.
g
Radical RTx includes radiotherapy alone and chemoradiotherapy.
h
EGFR mutation analysis was performed in 183 patients.
i
Seventeen patients were excluded from this analysis because primary lung cancer existed as multiple nodules or straddled multiple lobes.
EBs, emphysematous bullae; y, year; CI, confidence interval; FEV1.0, forced expiratory volume in 1 second; Ad, adenocarcinoma; Sq, squamous cell carcinoma;
NOS, not otherwise specified; TNM, tumor, node, and metastasis; RTx, radiotherapy; BSC, best supportive care; EGFR, epidermal growth factor receptor gene.

EBs in the upper lobe (87.9%), and it was present in the
lower lobe of nine of the 12 patients with EBs in
the lower lobe (75.0%). Overall, lung cancer was thus
present in the same lobe as the EBs in 38 of the 45 pa-
tients with EBs (84.4%), with the site of primary lung
cancer showing a significant association with that of EBs
(p < 0.0001).
We examined whether primary lung cancer was
present in the area adjoining EBs (Ca-ADJ) or apart from
EBs (Ca-AP). Representative CT images of Ca-ADJ and
Ca-AP are shown in Figure 2. Most of the primary lung
cancers in smokers with EBs (86.7% [39 of 45 cases])
presented as Ca-ADJ. Given the possibility that lung
cancer might tend to present as Ca-ADJ when most of the
lung is occupied by EBs, the ratio of the volume of EBs to
the volume of the entire lung (EBs per whole lung) was
evaluated. There was no difference in this ratio between
the patients presenting with Ca-ADJ and those present-
ing with Ca-AP (16.0% versus 17.8%, respectively, p ¼
0.822), suggesting that the higher frequency of Ca-ADJ
than of Ca-AP in patients with EBs was not due to
increased occupation of the lung with EBs.
September 2016
Figure 1. Overall survival according to the absence or
presence of emphysematous bullae in patients with
advanced lung cancer who were smokers and treated with
chemotherapy. MST, median survival time.
Discussion
We have here determined that the frequency of EBs
as detected by CT was 13.3% in smoking patients with
lung cancer. Lung cancer was present in the same lobe as
EBs in 38 of the 45 patients with EBs (84.4%), with the
lobar site of primary lung cancer thus showing a signif-
icant association with that of EBs. In addition, most of
the primary lung cancers in patients with EBs (86.7%)
presented as Ca-ADJ. Together, these results suggest that
the region surrounding EBs is a preferred site for the
generation of lung cancer.
Oxidants in cigarette smoke are thought to be respon-
sible for damage to DNA that eventually results in
the development of lung cancer.12 The accumulation of
multiple genetic alterations is thought to underlie smoking-
induced lung carcinogenesis.13 On the other hand,
Figure 2. Representative computed tomography images of primary lung cancer in areas adjoining emphysematous bullae
(EBs) (A) or apart from EBs (B). Arrowhead and arrows indicate primary lung cancer and EBs, respectively.
Lung Cancer with Emphysematous Bullae 1589
activating mutations of EGFR have been identified as
exclusive genetic changes that present more frequently in
lung cancer of never-smokers than in that of smokers.14
EGFR mutations have also been detected more frequently
in adenocarcinoma than in lung cancer with other histologic
diagnoses. Although the extent of smoking and the fre-
quency of adenocarcinoma did not differ between smokers
with or without EBs, the frequency of EGFR mutations was
significantly lower in the tumors of smokers with EBs than
in those of smokers without EBs. This finding suggests that
lung cancer with EBs is closely associated with smoking-
induced DNA damage. In addition, the smoking patients
with lung cancer with EBs were significantly younger
than those without EBs. Together, these results suggest
that patients with EBs have a higher susceptibility to
smoking-induced DNA damage. Constitutional factors such
as the accumulation of carcinogens and recurring inflam-
mation in EBs due to impaired ventilation and elimination
of cigarette smoke are thought to be responsible for the
predisposition to lung cancer in smokers with EBs.8,9
A previous study found that 26 of 595 lung cancers
(4.4%) were present in the area adjoining cystic air-
spaces, with most of these tumors (88.5% [23 of 26])
being adenocarcinoma.15 The frequency of adenocarci-
noma in patients with EBs in the present study (51.1%)
was lower than that in the patients with tumors
adjoining cystic airspaces in the previous study, which
might reflect the corresponding difference in the fre-
quency of emphysema (100% versus 73%).
In conclusion, we have found that the clinical features
of smoking patients with lung cancer differ according
to the absence or presence of EBs, with the presence
of EBs appearing to confer an increased suscep-
tibility to smoking-induced DNA damage. Analysis of a
broader range of genetic alterations in patients with EBs
1590 Iwama et al
compared with in those without EBs may provide
further insight into the mechanisms of carcinogenesis in
EB-associated lung cancer.
References
1. Parkin DM, Pisani P, Lopez AD, et al. At least one in seven
cases of cancer is caused by smoking. Global estimates
for 1985. Int J Cancer. 1994;59:494–504.
2. Jha P. Avoidable global cancer deaths and total deaths
from smoking. Nat Rev Cancer. 2009;9:655–664.
3. de Torres JP, Bastarrika G, Wisnivesky JP, et al. Assessing
the relationship between lung cancer risk and emphy-
sema detected on low-dose CT of the chest. Chest.
2007;132:1932–1938.
4. Zulueta JJ, Wisnivesky JP, Henschke CI, et al. Emphy-
sema scores predict death from COPD and lung cancer.
Chest. 2012;141:1216–1223.
5. Henschke CI, Yip R, Boffetta P, et al. CT screening for
lung cancer: importance of emphysema for never
smokers and smokers. Lung Cancer. 2015;88:42–47.
6. Hansell DM, Bankier AA, MacMahon H, et al. Fleischner
Society: glossary of terms for thoracic imaging. Radi-
ology. 2008;246:697–722.
7. Stoloff IL, Kanofsky P, Magilner L. The risk of lung cancer
in males with bullous disease of the lung. Arch Environ
Health. 1971;22:163–167.
Journal of Thoracic Oncology Vol. 11 No. 9
8. Goldstein MJ, Snider GL, Liberson M, et al. Bronchogenic
carcinoma and giant bullous disease. Am Rev Respir Dis.
1968;97:1062–1070.
9. Hanaoka N, Tanaka F, Otake Y, et al. Primary lung
carcinoma arising from emphysematous bullae. Lung
Cancer. 2002;38:185–191.
10. Matsushita H, Tanaka S, Saiki Y, et al. Lung cancer
associated with usual interstitial pneumonia. Pathol Int.
1995;45:925–932.
11. Muller NL, Staples CA, Miller RR, et al. “Density mask”.
An objective method to quantitate emphysema using
computed tomography. Chest. 1988;94:782–787.
12. Church DF, Pryor WA. Free-radical chemistry of cigarette
smoke and its toxicological implications. Environ Health
Perspect. 1985;64:111–126.
13. Osada H, Takahashi T. Genetic alterations of multiple
tumor suppressors and oncogenes in the carcinogenesis
and progression of lung cancer. Oncogene. 2002;21:
7421–7434.
14. Pao W, Miller V, Zakowski M, et al. EGF receptor gene
mutations are common in lung cancers from “never
smokers” and are associated with sensitivity of tumors
to gefitinib and erlotinib. Proc Natl Acad Sci U S A.
2004;101:13306–13311.
15. Farooqi AO, Cham M, Zhang L, et al. Lung cancer asso-
ciated with cystic airspaces. AJR Am J Roentgenol.
2012;199:781–786.