- Email: [email protected]
Primary lung carcinoma arising from emphysematous bullae
Lung Cancer 38 (2002) 185 /191 www.elsevier.com/locate/lungcan
Primary lung carcinoma arising from emphysematous bullae Nobuharu Hanaoka, Fumihiro Tanaka, Yosuke Otake, Kazuhiro Yanagihara, Tatsuo Nakagawa, Yozo Kawano, Ryo Miyahara, Mio Li, Hiromi Wada * Department of Thoracic Surgery, Faculty of Medicine, Kyoto University, 54 Shogoin Sakyo-ku, Kyoto 606-8507, Japan Received 18 March 2002; received in revised form 23 May 2002; accepted 27 May 2002
Abstract To clarify clinical characteristics and biological features of primary lung carcinoma arising from emphysematous bullae (EB), a total of 50 patients (49 males and one female) among all 1478 patients who underwent operation for primary lung carcinoma cases were reviewed; biological features were examined in 31 patients whose resected specimens were available for immunohistochemical staining (IHS). Thirty-one patients (62.0%) had pathologic stage I disease, and 30 cases (60.0%) had poorly differentiated tumor, demonstrating earlier pathologic stages and poorer cell differentiation of lung carcinoma with EB as compared with that without EB. The mean proliferative index (PI) for carcinoma with EB was 64.0%, which was significantly higher than that for carcinoma without EB (47.2%, P /0.001); no significant difference in Apoptotic index (AI) was demonstrated. Aberrant p53 expression was less frequent in carcinoma with EB (29.0%) than in carcinoma without EB (47.9%, P /0.043). Five-year survival rates for carcinoma with and that without EB were 50.3 and 46.9%, respectively, showing no significant difference. Multivariate analysis did not demonstrate that association of EB was a significant prognostic factor. In conclusion, although with the poorer cell differentiation and accelerated proliferative activity of lung carcinoma arising from EB, this does not have a significantly different prognosis than primary lung carcinoma not associated with bullae. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Emphysematous bullae; Patients; Carcinoma
1. Introduction Bullous emphysema has been proved to be an important risk factor for development of primary lung carcinoma [1 /7]. Korol reported that lung carcinoma developed in 9% of patients with bullous emphysema whereas the frequency was only 1.5% in those without bullous emphysema [2]. Stoloff and coworkers also reported a 32 times higher risk for development of lung carcinoma in patients with bullous emphysema; the frequency of lung cancer developed in patients with bullous disease was 6.1%, whereas that in patients without bullous disease was only 0.19% [4]. Moreover,
* Corresponding author. Tel.: /81-75-751-3835; fax: /81-75-7514647 E-mail address: [email protected] (H. Wada).
Goldstein and coworkers documented 18 patients (4.4%) of lung carcinoma arising in emphysematous bulla among all 411 patients who underwent thoracotomy for primary lung carcinoma [3]. Tsutsui and coworkers also reported 11 patients (2.5%) of lung carcinoma arising in bullous emphysema among resected 443 lung carcinoma patients [6]. Although increased risk of development of primary lung carcinoma in patients with emphysematous bulla has been suggested in these studies, detailed biological and clinical features of lung carcinoma associated with emphysematous bulla have not been revealed. We have already conducted a preliminary review on lung carcinoma associated with emphysematous bulla, and reported that the postoperative prognosis was comparable to that of lung carcinoma not associated with emphysematous bulla [8]. In the present study, biological features as well as postoperative prognosis of lung carcinoma arising from emphysematous bulla were assessed.
0169-5002/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 9 - 5 0 0 2 ( 0 2 ) 0 0 1 8 6 - 1
186
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
2. Materials and methods 2.1. Clinical characteristics of patients A total of 1478 consecutive patients who underwent thoracotomy for primary lung carcinoma at Kyoto University Hospital between January 1976 and December 1998, were retrospectively reviewed. Among them, 50 patients (3.4%) had lung carcinoma arising from emphysematous bullae (EB) (Bulla group), and the others had lung carcinoma without coexisting EB (Control group). EB were defined as bullae associated with pulmonary emphysema; bullae were defined as emphysematous space of more than 1cm in the diameter in the infiltrated lung (CIBA symposium [9]), and pulmonary emphysema was diagnosed with a pulmonary function test and a pathological examination of resected specimens. Thus, any lung carcinoma arising from bulla in the emphysematous lung including a carcinoma arising from giant bullae was included in the study. Pathologic (p-) stage was re-evaluated and determined according to the current TNM classification as revised in 1997 [10]. Histological type and tumor-cell differentiation were determined according to the World Health Organization as revised in 1999. In analysis stratified by grade of tumor-cell differentiation, welldifferentiated squamous cell carcinoma (Sq) and adenocarcinoma (Ad) were classified into well-differentiated tumor; moderately differentiated Sq and Ad were classified into moderately differentiated tumor; large cell carcinoma (La) and small cell carcinoma (Sm) as well as poorly differentiated Sq and Ad were classified into poorly differentiated tumor; the other histologic types were excluded in the analysis. The inpatient and outpatient medical records, chest X-ray films, wholebody computed tomography films, bone and gallium scanning data, and operation records were reviewed without knowledge of the results of immunohistochemical staining (IHS) for all patients. Follow-up of the postoperative clinical course was conducted by outpatient medical records and by inquiries by telephone or letter. The day of thoracotomy was considered the starting day for counting postoperative survival days. In addition, to clarify biological features of primary lung carcinoma arising in bullous emphysema, p53 status and incidence of apoptotic cell death as well as cell-proliferation of tumor-cell were examined and compared between the Bulla group and the Control group. Since the preoperative therapy might influence on these biological features, ten patients in the Bulla group who underwent preoperative therapy (radiotherapy in four, chemotherapy in three, and chemo-radiotherapy in three patients) were excluded in the analysis. Tumor specimens could not be available in nine patients who underwent operation before 1985, and a total of 31
patients were finally examined in the Bulla group. In the Control group, a total of consecutive 420 patients who underwent operation between 1986 and 1992 were analyzed. 2.2. Tissue preparation and detection of apoptotic cell, imuunohistochemistry Serial 4 mm sections were prepared from each formalin-fixed paraffin-embedded tissue of the primary tumor, and were served for routine hematoxylin and eosin (HE) staining, the terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling (TUNEL) staining to detect apoptotic cells, and IHS to detect PCNA expression and aberrant p53 expression. After the sections were dewaxed, the sections were digested with 20 mg/ml proteinase K (Boehringer Manheim, Manheim, Germany) for 20 min at 25 8C for the TNUEL staining. Sections were heated in microwave oven for 5 min three times each to retrieve their antigeneities for IHS. Endogenous peroxidese was inactivated by incubating the sections with 0.03% H2O2 in methanol (Nakalai Tesque, Kyoto, Japan) for 30 min at 25 8C. To reduce unspecific labeling, the sections were incubated with normal calf serum (DAKO Japan, Kyoto, Japan). The TUNEL staining was performed using In Situ Death Detection Kit, POD (Boehringer Manheim, Manheim, Germany) following the manufactured protocol, as described previously [11,12]. The specificity of the TUNEL staining of apoptotic cells was confirmed by making the negative and the positive control slides at every staining. As negative control slides, sections incubated with the TUNEL reaction mixture without TdT were used. As positive control slides, sections treated with Dnase I (0.7 mg/ml, Stratagene, La Jolla, CA) for 10 min at 25 8C before the TUNEL reaction. Apoptotic cells were determined with careful observation of TUNEL-staining sections and serial HE-staining sections. TUNEL-positive staining cells, if they did represent histological features of necrosis in HE-staining sections, were not considered to be apoptotic cells. In each case, a total of 10 000 tumor cells, 1000 tumor cells each in ten different fields, were evaluated at high magnification ( /400) by two authors independently (Fumihiro Tanaka and Yosuke Otake) without knowledge of patient’s characteristics or clinical data. Apoptotic index (AI) was defined as the number of apoptotic cells among 1000 cancer cells (AI). Procedure of IHS using streptoavidin /biotinylated horseradish peroxidase complex method (LSAB kit; DAKO JAPAN, Kyoto, Japan) was also described in the previous paper [13 /15]. Mouse anti-human PCNA, monoclonal antibody (MoAb) PC-10 (mouse IgG2a, kappa, 400 mg/ml, DAKO Japan) diluted at 1:50, antihuman p53 MoAb, clone DO-7 (mouse IgG2b, kappa,
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
250 mg/ml, DAKO Japan) diluted at 1:50 were used as the primary antibody. A total of 1000 tumor cells were counted for positive staining, and the percentages of positive cells were determined. When the percentage of the cells with positive staining exceeded 5%, the slide was judged to exhibit aberrant expression of p53. The fraction of proliferative cells was represented by the percentage of PCNA-positive cancer cells (proliferative index (PI)).
187
patients was only 3 (6.0%). In contrast, there were 597 (41.8%) patients with p-stage I, 214 (15.0%) with p-stage II, 495 (34.7%) with p-stage III, and 122 (8.5%) with pstage IV disease in the Control group. With respect to pT-factor, there was no significant difference between the Bulla group and the Control group (P /0.172), whereas pT4 disease seemed more frequent in the Control group (Table 1). 3.2. Biological features
2.3. Statistical methods Counts were compared by the chi-square test (x2), and trends in counts were analyzed by the x2-test for trends. Continuous data were compared using Student’s t -test if the distribution of samples was normal, or using Mann / Whitney U -test if the sample distribution was asymmetrical. Postoperative survival rate was analyzed by the Kaplan /Meier method, and the difference was assessed by the log /rank test. Multivariate analysis of prognostic factors was performed using Cox’s regression model. Differences were considered significant when P -value was less than 0.05. All statistical manipulations were performed using the SPSS for WINDOWS software system (SPSS Inc., Chicago, IL, USA, 1993).
3. Results 3.1. Clinical characteristics Clinical characteristics of patients in the Bulla group and those in the Control group were shown in Table 1. There was only one females (2.0%) among 50 patients in the Bulla group, whereas 414 females (29.0%) among 1428 patients in the Control group (P B/0.001). All patients in the Bulla group and about two-thirds patients in the Control group had a smoking-history with more than 20 pack-years, and the difference was significant (P B/0.001). In 18 patients (36.0%) in the Bulla group, the lung tumor was discovered with chest roentgenogram taken during follow-up of other disease (pulmonary emphysema in all patients); the rate was significantly higher than that in the Control group (14.5%, P B/0.001). Eleven patients (22.0%) among the Bulla group patients and 65 patients (4.6%) had La, showing a significant higher rate of La patients in the Bulla group (P B/0.001). About two-thirds patients (30 of 50 patients, 62.5%) in the Bulla group had a poorly-differentiated tumor, and only 16.7% patients had a welldifferentiated tumor; showing a significantly higher rate of patients with poorly-differentiated tumor as compared with the Control group (P B/0.001). In the Bulla group, there were 31 (62.0%) of p-stage I patients, whereas the number of p-stage III or IV
Aberrant expression of p53 was demonstrated in nine (29.0%) of 30 patients in the Bulla group and in 201 (47.9%) of 420 patients in the Control group, showing a significantly lower rate of patients with aberrant p53 in the Bulla group (P /0.043) (Table 2, Fig. 1A and B). The mean PI for the Bulla group was 64.0%, which was significantly higher than that for the Control group (47.2%, P /0.001) (Table 2, Fig. 1C and D). No significant difference in the AI was demonstrated between the Bulla and the Control groups (Table 2). 3.3. Postoperative survival For all patients, 5-year survival rates for the Bulla group and the Control group were 50.3 and 46.9%, showing a no significant difference (Fig. 2). No significant difference in the postoperative survival was demonstrated in any histologic type or p-stage between the Bulla group and the Control group (Table 3). Multivariative analysis confirmed that the prognosis of lung carcinoma arising from EB was not worse than that of lung carcinoma not associated with EB (Table 4).
4. Discussion In the present study, characteristics of primary lung cancer arising in emphysematous bulla were demonstrated as follows: (1) poorer differentiation, (2) higher proliferative activity, (3) less-frequent p53 abnormality. With respect to histologic type, Zulueta and coworkers reviewed reported cases, and demonstrated that all but one case had NSCLC [16,17]. However, there was no study on tumor differentiation of lung cancer arising in emphysematous bulla. In the present study, poorly differentiated tumor including La was more frequently developed in the Bulla group than in the Control group, whereas no difference in the histologic type was demonstrated between these groups. In addition, higher PI and less-frequent p53 aberrant expression was demonstrated in the Bulla group, whereas no difference in incidence of apoptotic cancer-cell death was demonstrated. These biological features, which have not been studied in previous reports, may be correlated with the nature of development of lung carcinoma in emphyse-
188
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
Table 1 Characteristics of patients with resected primary lung cancer associated with and without bulla Number of patients
%
Lung cancer associated with or without bulla With bulla (%)
Total Age (mean9S.D., years)
1478 62.0910.1
100 50 (3.4) 62.0910.1 61.799.9
Sex Male Female
1063 415
71.9 28.1
Performance status (PS ) 0 1 2 /4
1208 244 26
Smoking Pack-yearB 20 Pack-year] 20
Without bulla (%)
P -value (with vs. without bulla)
1428 (96.6) 62.0910.1
/ 0.793
49 (98.0) 1 (2.0)
1014 (71.0) 414 (29.0)
B 0.001 /
81.7 16.5 1.8
36 (72.0) 12 (24.0) 2 (4.0)
1172 (82.1) 232 (16.2) 24 (1.7)
0.195 /
502 976
34.0 66.0
0 (0.0) 50 (100)
502 (35.2) 926 (64.8)
B 0.001 /
Cause of discovery At mass screening During follow-up of other diseases Subjective complaints or symptoms
607 225 646
41.1 15.2 43.7
17 (34.0) 18 (36.0) 15 (30.0)
590 (41.3) 207 (14.5) 631 (44.2)
0.380 B 0.001 0.059
Histologic type Squamous cell (Sq) Adenocarcinoma (Ad) Large cell (La) Small cell (Sm) Others
499 773 76 75 55
33.8 52.3 5.1 5.1 3.7
13 21 11 3 2
486 752 65 72 53
(34.0) (52.7) (4.6) (5.0) (3.7)
0.287 0.151 B 0.001 0.738 /
Tumor cell differentiation * Well differentiated Moderately differentiated Poorly differentiated
473 499 376
35.1 37.0 27.9
8 (17.8) 10 (22.2) 27 (60.0)
465 (35.7) 489 (37.5) 349 (26.8)
/ B 0.001 /
Pathologic (p- ) stage I (A, B) II (A, B) III (A, B) IV
628 227 498 125
42.5 15.4 33.7 8.5
31 13 3 3
(62.0) (26.0) (6.0) (6.0)
597 214 495 122
(41.8) (15.0) (34.7) (8.5)
/ B 0.001 / /
Pathologic T-factor T0 T1 T2 T3 T4
3 452 544 242 237
0.2 30.6 36.8 16.4 16.0
0 15 23 10 2
(0.0) (30.0) (46.0) (20.0) (4.0)
3 437 521 232 235
(0.2) (30.6) (36.5) (16.2) (10.5)
/ / 0.172 / /
(26.0) (42.0) (22.0) (6.0) (4.0)
* Other histologic types and small cell carcinoma were excluded in the analysis.
Table 2 P53 status, AI and PI in primary lung cancer associated with and without bulla Number of patients
p53 status (Aberrant p53 expression) Negative (%) Positive (%)
241 210
%
Lung cancer associated with or without bulla With bulla
Without bulla
P -value (with vs without bulla)
53.4 46.6
22 (71.0) 9 (29.0)
219 (52.1) 201 (47.9)
0.043 /
AI Mean9SE (per 1000 cancer cells)
23.992.3
18.594.4
24.492.5
0.519
/
PI Mean9S.E. (%)
47.791.5
64.094.5
47.291.5
0.001
/
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
189
Fig. 1. IHS for evaluation of p53 status (A and B) and proliferative activity (C and D). (A) Almost all tumor cells were positive for p53 expression in moderately differentiated squamous cell carcinoma arising from EB. (B) No p53 expression was documented in moderately differentiated adenocarcinoma not associated with EB. (C) Almost all tumor cells were positive for proliferative cell nuclear antigen (PCNA) in the same tumor as in A. (D) Almost no PCNA-positive cells were documented in the same tumor as in B.
matous bulla; impaired ventilation of bulla may facilitate deposition of several carcinogens in bulla, which may cause metaplastic transformation of the innerlining epithelial cells [3,18]. Accumulated carcinogens may cause aggressive behavior of lung carcinoma arising in emphysematous bulla. Interestingly, p53 abnormality was less frequent in lung carcinoma with bulla. Examination of other gene status such as k-ras as well as
examination of p53 gene mutation, not aberrant expression, should be performed in further study. Considering these aggressive characteristics of lung carcinoma arising from emphysematous bulla, early detection of lung tumor in patients with emphysematous bulla should be preferred. In fact, as demonstrated in the present study, postoperative survival of lung carcinoma arising from bulla was comparable to that lung
Fig. 2. Postoperative survival of patients who underwent operation for primary lung cancer associated with bulla (Bulla group) and that without bulla (Control group).
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
190
Table 3 Univariative analysis of postoperative survival in primary lung cancer associated with and without bulla Five-year survival rate for patients lung cancer associated with and without bulla With bulla (%)
Without bulla (%)
P -value (with vs. without bulla)
All patients
50.3
46.9
0.473
Stratified by histologic type Squamous cell (Sq) Adenocarcinoma (Ad) Large cell (La) Small cell (Sm)
48.5 40.5 74.1 0.0
40.0 52.1 48.6 35.5
0.797 0.323 0.063 0.487
Stratified by pathologic (p- ) stage I (A, B) 65.4 II (A, B) 42.3 III (A, B) 0.0 IV 0.0
71.7 50.4 25.3 4.1
0.355 0.915 0.958 0.344
Table 4 Multivariative analysis of prognostic factors in primary lung cancer Prognostic factors
b
P -value
Relative hazard (95% confidence interval)
Age Sex (male/female) Performance status (0/1//2/3) Histologic type (non-adenocarcinoma/adenocarcinoma) Association with bulla (EB) (no/yes)
1.015 0.215 0.536 0.210 0.095
0.001 0.047 B 0.001 0.021 0.728
0.015 0.807 1.709 0.811 0.910
carcinoma without bulla, if lung carcinoma was resected in earlier stages. Goldstein and coworkers recommended annual chest roentgenogram in patients with emphysematous bulla [3]. However, it is sometimes difficult to detect lung tumor in chest roentgenogram when underlying bullous shadow is present. Nickoladze and coworkers reviewed 21 cases of lung carcinoma arising from emphysematous bulla, and reported that lung tumor could not revealed by chest roentgenogram in any case; lung tumor was detected by CT in 20 cases and by autopsy in the one case [7]. Since it is apparent that CT is superior in detection of lung tumor, CT should be taken when some changes on chest roentgenogram are demonstrated during follow-up of patients with emphysematous bulla [6,16]. In conclusion, although with the poorer cell differentiation and accelerated proliferative activity of lung carcinoma arising from EB, this does not have a significantly different prognosis than primary lung carcinoma not associated with bullae.
References [1] Bass H, Singer E. Co-existing lobar adenocarcinoma and cystic disease of the lung. Ann Intern Med 1951;34:498 /507.
(1.006 /1.024) (0.654 /0.995) (1.456 /2.006) (0.679 /0.969) (0.534 /1.550)
[2] Korol EK. The correlation of carcinoma and congenital cystic emphysema of the lungs: report of ten cases. Dis Chest 1953;23:403 /11. [3] Goldstein MJ, Snider GL, Liberson M, Poske RM. Bronchogenic carcinoma and giant bullous disease. Am Rev Respir Dis 1968;97:1062 /70. [4] 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 /7. [5] Wats MA, Klayton RJ, Munzel TL. Bullous emphysema and carcinoma of the lung: case report. Mil Med 1982;147:320 /3. [6] Tsutsui M, Araki Y, Shirakusa T, Inutsuka S. Characteristics radiographic features of pulmonary carcinoma associated with large bulla. Ann Thorac Surg 1988;46:679 /83. [7] Nickoladze GD. Bullae and lung cancer. J Thorac Cardiovasc Surg 1993;106:186. [8] Tanaka F, Wada H, Hitomi S, Itoh M. Surgery for complications of chronic obstructive lung disease. In: Deslauriers J, LeBlanc, editors. Chest surgery clinics of North America: surgery for emphysema. Philadelphia: WB Saunders, 1995:797 /813. [9] Deslauriers J, Leblanc P. Bullous and bleb diseases of the lung. In: Shields TW, editor. General thoracic surgery, 4th ed. Marvern/ Philadelphia: Williams and Wilkins, 1994:907 /29. [10] Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710 /7. [11] Tanaka F, Kawano Y, Li M, et al. Prognostic significance of apoptotic index in completely resected non-small cell lung cancer. J Clin Oncol 1999;17:2728 /36. [12] Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992;119:493 /501.
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191 [13] Bravo R, Frank R, Blundell PA, MacDonald PH. Cyclin/PCNA is the auxilary protein of DNA polymerase @. Nature 1987;326:515 /7. [14] Tanaka F, Miyahara R, Ohtake Y, et al. Lewis Y antigen expression and postoperative survival in nonsmall cell lung cancer. Ann Thorac Surg 1998;66:1745 /50. [15] Tanaka F, Yanagihara K, Ohtake K, et al. p53 status predicts the efficacy of postoperative oral administration of tegafur for completely resected non-small cell lung cancer. Jpn J Cancer Res 1999;90:432 /8.
191
[16] Zulueta JJ, Stephan SM, Rozansky MI, White AC. Lung carcinoma in patients with bullous disease. Am J Respir Crit Care Med 1996;154:519 /22. [17] Aronberg DJ, Sagel SS, LeFrak S, Kuhn C, Susman N. Lung carcinoma associated with bullous lung disease in young men. Am J Roentgenol 1980;134:249 /52. [18] Womack NA, Graham EA. Epithelial metaplasia in congenital cystic disease of the lung: its possible relation to carcinoma of the bronchus. Am J Pathol 1941;17:645 /54.
Primary lung carcinoma arising from emphysematous bullae Nobuharu Hanaoka, Fumihiro Tanaka, Yosuke Otake, Kazuhiro Yanagihara, Tatsuo Nakagawa, Yozo Kawano, Ryo Miyahara, Mio Li, Hiromi Wada * Department of Thoracic Surgery, Faculty of Medicine, Kyoto University, 54 Shogoin Sakyo-ku, Kyoto 606-8507, Japan Received 18 March 2002; received in revised form 23 May 2002; accepted 27 May 2002
Abstract To clarify clinical characteristics and biological features of primary lung carcinoma arising from emphysematous bullae (EB), a total of 50 patients (49 males and one female) among all 1478 patients who underwent operation for primary lung carcinoma cases were reviewed; biological features were examined in 31 patients whose resected specimens were available for immunohistochemical staining (IHS). Thirty-one patients (62.0%) had pathologic stage I disease, and 30 cases (60.0%) had poorly differentiated tumor, demonstrating earlier pathologic stages and poorer cell differentiation of lung carcinoma with EB as compared with that without EB. The mean proliferative index (PI) for carcinoma with EB was 64.0%, which was significantly higher than that for carcinoma without EB (47.2%, P /0.001); no significant difference in Apoptotic index (AI) was demonstrated. Aberrant p53 expression was less frequent in carcinoma with EB (29.0%) than in carcinoma without EB (47.9%, P /0.043). Five-year survival rates for carcinoma with and that without EB were 50.3 and 46.9%, respectively, showing no significant difference. Multivariate analysis did not demonstrate that association of EB was a significant prognostic factor. In conclusion, although with the poorer cell differentiation and accelerated proliferative activity of lung carcinoma arising from EB, this does not have a significantly different prognosis than primary lung carcinoma not associated with bullae. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Emphysematous bullae; Patients; Carcinoma
1. Introduction Bullous emphysema has been proved to be an important risk factor for development of primary lung carcinoma [1 /7]. Korol reported that lung carcinoma developed in 9% of patients with bullous emphysema whereas the frequency was only 1.5% in those without bullous emphysema [2]. Stoloff and coworkers also reported a 32 times higher risk for development of lung carcinoma in patients with bullous emphysema; the frequency of lung cancer developed in patients with bullous disease was 6.1%, whereas that in patients without bullous disease was only 0.19% [4]. Moreover,
* Corresponding author. Tel.: /81-75-751-3835; fax: /81-75-7514647 E-mail address: [email protected] (H. Wada).
Goldstein and coworkers documented 18 patients (4.4%) of lung carcinoma arising in emphysematous bulla among all 411 patients who underwent thoracotomy for primary lung carcinoma [3]. Tsutsui and coworkers also reported 11 patients (2.5%) of lung carcinoma arising in bullous emphysema among resected 443 lung carcinoma patients [6]. Although increased risk of development of primary lung carcinoma in patients with emphysematous bulla has been suggested in these studies, detailed biological and clinical features of lung carcinoma associated with emphysematous bulla have not been revealed. We have already conducted a preliminary review on lung carcinoma associated with emphysematous bulla, and reported that the postoperative prognosis was comparable to that of lung carcinoma not associated with emphysematous bulla [8]. In the present study, biological features as well as postoperative prognosis of lung carcinoma arising from emphysematous bulla were assessed.
0169-5002/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 9 - 5 0 0 2 ( 0 2 ) 0 0 1 8 6 - 1
186
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
2. Materials and methods 2.1. Clinical characteristics of patients A total of 1478 consecutive patients who underwent thoracotomy for primary lung carcinoma at Kyoto University Hospital between January 1976 and December 1998, were retrospectively reviewed. Among them, 50 patients (3.4%) had lung carcinoma arising from emphysematous bullae (EB) (Bulla group), and the others had lung carcinoma without coexisting EB (Control group). EB were defined as bullae associated with pulmonary emphysema; bullae were defined as emphysematous space of more than 1cm in the diameter in the infiltrated lung (CIBA symposium [9]), and pulmonary emphysema was diagnosed with a pulmonary function test and a pathological examination of resected specimens. Thus, any lung carcinoma arising from bulla in the emphysematous lung including a carcinoma arising from giant bullae was included in the study. Pathologic (p-) stage was re-evaluated and determined according to the current TNM classification as revised in 1997 [10]. Histological type and tumor-cell differentiation were determined according to the World Health Organization as revised in 1999. In analysis stratified by grade of tumor-cell differentiation, welldifferentiated squamous cell carcinoma (Sq) and adenocarcinoma (Ad) were classified into well-differentiated tumor; moderately differentiated Sq and Ad were classified into moderately differentiated tumor; large cell carcinoma (La) and small cell carcinoma (Sm) as well as poorly differentiated Sq and Ad were classified into poorly differentiated tumor; the other histologic types were excluded in the analysis. The inpatient and outpatient medical records, chest X-ray films, wholebody computed tomography films, bone and gallium scanning data, and operation records were reviewed without knowledge of the results of immunohistochemical staining (IHS) for all patients. Follow-up of the postoperative clinical course was conducted by outpatient medical records and by inquiries by telephone or letter. The day of thoracotomy was considered the starting day for counting postoperative survival days. In addition, to clarify biological features of primary lung carcinoma arising in bullous emphysema, p53 status and incidence of apoptotic cell death as well as cell-proliferation of tumor-cell were examined and compared between the Bulla group and the Control group. Since the preoperative therapy might influence on these biological features, ten patients in the Bulla group who underwent preoperative therapy (radiotherapy in four, chemotherapy in three, and chemo-radiotherapy in three patients) were excluded in the analysis. Tumor specimens could not be available in nine patients who underwent operation before 1985, and a total of 31
patients were finally examined in the Bulla group. In the Control group, a total of consecutive 420 patients who underwent operation between 1986 and 1992 were analyzed. 2.2. Tissue preparation and detection of apoptotic cell, imuunohistochemistry Serial 4 mm sections were prepared from each formalin-fixed paraffin-embedded tissue of the primary tumor, and were served for routine hematoxylin and eosin (HE) staining, the terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling (TUNEL) staining to detect apoptotic cells, and IHS to detect PCNA expression and aberrant p53 expression. After the sections were dewaxed, the sections were digested with 20 mg/ml proteinase K (Boehringer Manheim, Manheim, Germany) for 20 min at 25 8C for the TNUEL staining. Sections were heated in microwave oven for 5 min three times each to retrieve their antigeneities for IHS. Endogenous peroxidese was inactivated by incubating the sections with 0.03% H2O2 in methanol (Nakalai Tesque, Kyoto, Japan) for 30 min at 25 8C. To reduce unspecific labeling, the sections were incubated with normal calf serum (DAKO Japan, Kyoto, Japan). The TUNEL staining was performed using In Situ Death Detection Kit, POD (Boehringer Manheim, Manheim, Germany) following the manufactured protocol, as described previously [11,12]. The specificity of the TUNEL staining of apoptotic cells was confirmed by making the negative and the positive control slides at every staining. As negative control slides, sections incubated with the TUNEL reaction mixture without TdT were used. As positive control slides, sections treated with Dnase I (0.7 mg/ml, Stratagene, La Jolla, CA) for 10 min at 25 8C before the TUNEL reaction. Apoptotic cells were determined with careful observation of TUNEL-staining sections and serial HE-staining sections. TUNEL-positive staining cells, if they did represent histological features of necrosis in HE-staining sections, were not considered to be apoptotic cells. In each case, a total of 10 000 tumor cells, 1000 tumor cells each in ten different fields, were evaluated at high magnification ( /400) by two authors independently (Fumihiro Tanaka and Yosuke Otake) without knowledge of patient’s characteristics or clinical data. Apoptotic index (AI) was defined as the number of apoptotic cells among 1000 cancer cells (AI). Procedure of IHS using streptoavidin /biotinylated horseradish peroxidase complex method (LSAB kit; DAKO JAPAN, Kyoto, Japan) was also described in the previous paper [13 /15]. Mouse anti-human PCNA, monoclonal antibody (MoAb) PC-10 (mouse IgG2a, kappa, 400 mg/ml, DAKO Japan) diluted at 1:50, antihuman p53 MoAb, clone DO-7 (mouse IgG2b, kappa,
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
250 mg/ml, DAKO Japan) diluted at 1:50 were used as the primary antibody. A total of 1000 tumor cells were counted for positive staining, and the percentages of positive cells were determined. When the percentage of the cells with positive staining exceeded 5%, the slide was judged to exhibit aberrant expression of p53. The fraction of proliferative cells was represented by the percentage of PCNA-positive cancer cells (proliferative index (PI)).
187
patients was only 3 (6.0%). In contrast, there were 597 (41.8%) patients with p-stage I, 214 (15.0%) with p-stage II, 495 (34.7%) with p-stage III, and 122 (8.5%) with pstage IV disease in the Control group. With respect to pT-factor, there was no significant difference between the Bulla group and the Control group (P /0.172), whereas pT4 disease seemed more frequent in the Control group (Table 1). 3.2. Biological features
2.3. Statistical methods Counts were compared by the chi-square test (x2), and trends in counts were analyzed by the x2-test for trends. Continuous data were compared using Student’s t -test if the distribution of samples was normal, or using Mann / Whitney U -test if the sample distribution was asymmetrical. Postoperative survival rate was analyzed by the Kaplan /Meier method, and the difference was assessed by the log /rank test. Multivariate analysis of prognostic factors was performed using Cox’s regression model. Differences were considered significant when P -value was less than 0.05. All statistical manipulations were performed using the SPSS for WINDOWS software system (SPSS Inc., Chicago, IL, USA, 1993).
3. Results 3.1. Clinical characteristics Clinical characteristics of patients in the Bulla group and those in the Control group were shown in Table 1. There was only one females (2.0%) among 50 patients in the Bulla group, whereas 414 females (29.0%) among 1428 patients in the Control group (P B/0.001). All patients in the Bulla group and about two-thirds patients in the Control group had a smoking-history with more than 20 pack-years, and the difference was significant (P B/0.001). In 18 patients (36.0%) in the Bulla group, the lung tumor was discovered with chest roentgenogram taken during follow-up of other disease (pulmonary emphysema in all patients); the rate was significantly higher than that in the Control group (14.5%, P B/0.001). Eleven patients (22.0%) among the Bulla group patients and 65 patients (4.6%) had La, showing a significant higher rate of La patients in the Bulla group (P B/0.001). About two-thirds patients (30 of 50 patients, 62.5%) in the Bulla group had a poorly-differentiated tumor, and only 16.7% patients had a welldifferentiated tumor; showing a significantly higher rate of patients with poorly-differentiated tumor as compared with the Control group (P B/0.001). In the Bulla group, there were 31 (62.0%) of p-stage I patients, whereas the number of p-stage III or IV
Aberrant expression of p53 was demonstrated in nine (29.0%) of 30 patients in the Bulla group and in 201 (47.9%) of 420 patients in the Control group, showing a significantly lower rate of patients with aberrant p53 in the Bulla group (P /0.043) (Table 2, Fig. 1A and B). The mean PI for the Bulla group was 64.0%, which was significantly higher than that for the Control group (47.2%, P /0.001) (Table 2, Fig. 1C and D). No significant difference in the AI was demonstrated between the Bulla and the Control groups (Table 2). 3.3. Postoperative survival For all patients, 5-year survival rates for the Bulla group and the Control group were 50.3 and 46.9%, showing a no significant difference (Fig. 2). No significant difference in the postoperative survival was demonstrated in any histologic type or p-stage between the Bulla group and the Control group (Table 3). Multivariative analysis confirmed that the prognosis of lung carcinoma arising from EB was not worse than that of lung carcinoma not associated with EB (Table 4).
4. Discussion In the present study, characteristics of primary lung cancer arising in emphysematous bulla were demonstrated as follows: (1) poorer differentiation, (2) higher proliferative activity, (3) less-frequent p53 abnormality. With respect to histologic type, Zulueta and coworkers reviewed reported cases, and demonstrated that all but one case had NSCLC [16,17]. However, there was no study on tumor differentiation of lung cancer arising in emphysematous bulla. In the present study, poorly differentiated tumor including La was more frequently developed in the Bulla group than in the Control group, whereas no difference in the histologic type was demonstrated between these groups. In addition, higher PI and less-frequent p53 aberrant expression was demonstrated in the Bulla group, whereas no difference in incidence of apoptotic cancer-cell death was demonstrated. These biological features, which have not been studied in previous reports, may be correlated with the nature of development of lung carcinoma in emphyse-
188
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
Table 1 Characteristics of patients with resected primary lung cancer associated with and without bulla Number of patients
%
Lung cancer associated with or without bulla With bulla (%)
Total Age (mean9S.D., years)
1478 62.0910.1
100 50 (3.4) 62.0910.1 61.799.9
Sex Male Female
1063 415
71.9 28.1
Performance status (PS ) 0 1 2 /4
1208 244 26
Smoking Pack-yearB 20 Pack-year] 20
Without bulla (%)
P -value (with vs. without bulla)
1428 (96.6) 62.0910.1
/ 0.793
49 (98.0) 1 (2.0)
1014 (71.0) 414 (29.0)
B 0.001 /
81.7 16.5 1.8
36 (72.0) 12 (24.0) 2 (4.0)
1172 (82.1) 232 (16.2) 24 (1.7)
0.195 /
502 976
34.0 66.0
0 (0.0) 50 (100)
502 (35.2) 926 (64.8)
B 0.001 /
Cause of discovery At mass screening During follow-up of other diseases Subjective complaints or symptoms
607 225 646
41.1 15.2 43.7
17 (34.0) 18 (36.0) 15 (30.0)
590 (41.3) 207 (14.5) 631 (44.2)
0.380 B 0.001 0.059
Histologic type Squamous cell (Sq) Adenocarcinoma (Ad) Large cell (La) Small cell (Sm) Others
499 773 76 75 55
33.8 52.3 5.1 5.1 3.7
13 21 11 3 2
486 752 65 72 53
(34.0) (52.7) (4.6) (5.0) (3.7)
0.287 0.151 B 0.001 0.738 /
Tumor cell differentiation * Well differentiated Moderately differentiated Poorly differentiated
473 499 376
35.1 37.0 27.9
8 (17.8) 10 (22.2) 27 (60.0)
465 (35.7) 489 (37.5) 349 (26.8)
/ B 0.001 /
Pathologic (p- ) stage I (A, B) II (A, B) III (A, B) IV
628 227 498 125
42.5 15.4 33.7 8.5
31 13 3 3
(62.0) (26.0) (6.0) (6.0)
597 214 495 122
(41.8) (15.0) (34.7) (8.5)
/ B 0.001 / /
Pathologic T-factor T0 T1 T2 T3 T4
3 452 544 242 237
0.2 30.6 36.8 16.4 16.0
0 15 23 10 2
(0.0) (30.0) (46.0) (20.0) (4.0)
3 437 521 232 235
(0.2) (30.6) (36.5) (16.2) (10.5)
/ / 0.172 / /
(26.0) (42.0) (22.0) (6.0) (4.0)
* Other histologic types and small cell carcinoma were excluded in the analysis.
Table 2 P53 status, AI and PI in primary lung cancer associated with and without bulla Number of patients
p53 status (Aberrant p53 expression) Negative (%) Positive (%)
241 210
%
Lung cancer associated with or without bulla With bulla
Without bulla
P -value (with vs without bulla)
53.4 46.6
22 (71.0) 9 (29.0)
219 (52.1) 201 (47.9)
0.043 /
AI Mean9SE (per 1000 cancer cells)
23.992.3
18.594.4
24.492.5
0.519
/
PI Mean9S.E. (%)
47.791.5
64.094.5
47.291.5
0.001
/
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
189
Fig. 1. IHS for evaluation of p53 status (A and B) and proliferative activity (C and D). (A) Almost all tumor cells were positive for p53 expression in moderately differentiated squamous cell carcinoma arising from EB. (B) No p53 expression was documented in moderately differentiated adenocarcinoma not associated with EB. (C) Almost all tumor cells were positive for proliferative cell nuclear antigen (PCNA) in the same tumor as in A. (D) Almost no PCNA-positive cells were documented in the same tumor as in B.
matous bulla; impaired ventilation of bulla may facilitate deposition of several carcinogens in bulla, which may cause metaplastic transformation of the innerlining epithelial cells [3,18]. Accumulated carcinogens may cause aggressive behavior of lung carcinoma arising in emphysematous bulla. Interestingly, p53 abnormality was less frequent in lung carcinoma with bulla. Examination of other gene status such as k-ras as well as
examination of p53 gene mutation, not aberrant expression, should be performed in further study. Considering these aggressive characteristics of lung carcinoma arising from emphysematous bulla, early detection of lung tumor in patients with emphysematous bulla should be preferred. In fact, as demonstrated in the present study, postoperative survival of lung carcinoma arising from bulla was comparable to that lung
Fig. 2. Postoperative survival of patients who underwent operation for primary lung cancer associated with bulla (Bulla group) and that without bulla (Control group).
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191
190
Table 3 Univariative analysis of postoperative survival in primary lung cancer associated with and without bulla Five-year survival rate for patients lung cancer associated with and without bulla With bulla (%)
Without bulla (%)
P -value (with vs. without bulla)
All patients
50.3
46.9
0.473
Stratified by histologic type Squamous cell (Sq) Adenocarcinoma (Ad) Large cell (La) Small cell (Sm)
48.5 40.5 74.1 0.0
40.0 52.1 48.6 35.5
0.797 0.323 0.063 0.487
Stratified by pathologic (p- ) stage I (A, B) 65.4 II (A, B) 42.3 III (A, B) 0.0 IV 0.0
71.7 50.4 25.3 4.1
0.355 0.915 0.958 0.344
Table 4 Multivariative analysis of prognostic factors in primary lung cancer Prognostic factors
b
P -value
Relative hazard (95% confidence interval)
Age Sex (male/female) Performance status (0/1//2/3) Histologic type (non-adenocarcinoma/adenocarcinoma) Association with bulla (EB) (no/yes)
1.015 0.215 0.536 0.210 0.095
0.001 0.047 B 0.001 0.021 0.728
0.015 0.807 1.709 0.811 0.910
carcinoma without bulla, if lung carcinoma was resected in earlier stages. Goldstein and coworkers recommended annual chest roentgenogram in patients with emphysematous bulla [3]. However, it is sometimes difficult to detect lung tumor in chest roentgenogram when underlying bullous shadow is present. Nickoladze and coworkers reviewed 21 cases of lung carcinoma arising from emphysematous bulla, and reported that lung tumor could not revealed by chest roentgenogram in any case; lung tumor was detected by CT in 20 cases and by autopsy in the one case [7]. Since it is apparent that CT is superior in detection of lung tumor, CT should be taken when some changes on chest roentgenogram are demonstrated during follow-up of patients with emphysematous bulla [6,16]. In conclusion, although with the poorer cell differentiation and accelerated proliferative activity of lung carcinoma arising from EB, this does not have a significantly different prognosis than primary lung carcinoma not associated with bullae.
References [1] Bass H, Singer E. Co-existing lobar adenocarcinoma and cystic disease of the lung. Ann Intern Med 1951;34:498 /507.
(1.006 /1.024) (0.654 /0.995) (1.456 /2.006) (0.679 /0.969) (0.534 /1.550)
[2] Korol EK. The correlation of carcinoma and congenital cystic emphysema of the lungs: report of ten cases. Dis Chest 1953;23:403 /11. [3] Goldstein MJ, Snider GL, Liberson M, Poske RM. Bronchogenic carcinoma and giant bullous disease. Am Rev Respir Dis 1968;97:1062 /70. [4] 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 /7. [5] Wats MA, Klayton RJ, Munzel TL. Bullous emphysema and carcinoma of the lung: case report. Mil Med 1982;147:320 /3. [6] Tsutsui M, Araki Y, Shirakusa T, Inutsuka S. Characteristics radiographic features of pulmonary carcinoma associated with large bulla. Ann Thorac Surg 1988;46:679 /83. [7] Nickoladze GD. Bullae and lung cancer. J Thorac Cardiovasc Surg 1993;106:186. [8] Tanaka F, Wada H, Hitomi S, Itoh M. Surgery for complications of chronic obstructive lung disease. In: Deslauriers J, LeBlanc, editors. Chest surgery clinics of North America: surgery for emphysema. Philadelphia: WB Saunders, 1995:797 /813. [9] Deslauriers J, Leblanc P. Bullous and bleb diseases of the lung. In: Shields TW, editor. General thoracic surgery, 4th ed. Marvern/ Philadelphia: Williams and Wilkins, 1994:907 /29. [10] Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710 /7. [11] Tanaka F, Kawano Y, Li M, et al. Prognostic significance of apoptotic index in completely resected non-small cell lung cancer. J Clin Oncol 1999;17:2728 /36. [12] Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992;119:493 /501.
N. Hanaoka et al. / Lung Cancer 38 (2002) 185 /191 [13] Bravo R, Frank R, Blundell PA, MacDonald PH. Cyclin/PCNA is the auxilary protein of DNA polymerase @. Nature 1987;326:515 /7. [14] Tanaka F, Miyahara R, Ohtake Y, et al. Lewis Y antigen expression and postoperative survival in nonsmall cell lung cancer. Ann Thorac Surg 1998;66:1745 /50. [15] Tanaka F, Yanagihara K, Ohtake K, et al. p53 status predicts the efficacy of postoperative oral administration of tegafur for completely resected non-small cell lung cancer. Jpn J Cancer Res 1999;90:432 /8.
191
[16] Zulueta JJ, Stephan SM, Rozansky MI, White AC. Lung carcinoma in patients with bullous disease. Am J Respir Crit Care Med 1996;154:519 /22. [17] Aronberg DJ, Sagel SS, LeFrak S, Kuhn C, Susman N. Lung carcinoma associated with bullous lung disease in young men. Am J Roentgenol 1980;134:249 /52. [18] Womack NA, Graham EA. Epithelial metaplasia in congenital cystic disease of the lung: its possible relation to carcinoma of the bronchus. Am J Pathol 1941;17:645 /54.