Radiol Oncol 2005; 39(1): 49-53.
Multiple primary malignancies in patients with lung cancer
Koichi Kurishima1, Hiroaki Satoh1, Shinsuke Homma1, Katsunori Kagohashi1, Hiroichi Ishikawa2, Morio Ohtsuka1, Kiyohisa Sekizawa1
1Division of Respiratory Medicine, Institute of Clinical Medicine, University of Tsukuba; 2Department of Internal Medicine, Tsukuba Medical Center Hospital, Japan
Background. To evaluate the incidence of multiple primary malignancies in lung cancer patients, we sum-marized our experience in lung cancer patients with multiple primary malignancies. Methods. A total of 1194 consecutive lung cancer patients, who were admitted to our division over a 29-year period up to August 2004, were retrospectively analyzed.
Results. Ninety-eight (8.2%) of 1194 lung cancer patients had multiple primary malignancies. Metachronous malignant disease comprised 77.6% and synchronous 21.4%. Multiple primary tumours in our patients were detected more frequent in the advanced stage of lung cancer (IIIA-IV 67.3%) than in the early stage (IA-IIB 32.7%). The histological examination of lung cancer revealed a preponderance of squa-mous cell carcinoma (40 patients, 40.8%). First primary tumours developed most commonly in gastroin-testinal tract, followed by lung and uterus. Fifty-seven (85.1%) of 67 patients with aerogastrointestinal and head and neck cancers had a smoking habit. In 98 patients with multiple primary cancers, forty (40.8%) pa-tients had stage IA-IIIA lung cancer, however, 26 (26.5%) had a surgical resection. Conclusions. Existing metachronous primary tumours proved to be a worse prognostic factor in non-small cell lung cancer patients (p=0.0480), while synchronous primary tumours were not, as well as there was not proven that multiple primary tumours were worse prognostic factors in patients with small cell lung cancer.
Key words: lung neoplasms; neoplasms, multiple primary
Introduction
Nearly one hundred years ago, Billroth1 first reported on synchronous cancers in various
Received 23 September 2004 Accepted 20 October 2004
Correspondence to: Hiroaki Satoh, M.D., Division of Respiratory Medicine, Institute of Clinical Medicine, University of Tsukuba, Tsukuba-City, Ibaraki, 305-8575, Japan; Fax: +81-29-853-3320; E-mail: hirosato@ md.tsukuba.ac.jp
organs. About 40 years later, Warren and Gated established the criteria for the diagno-sis of multiple primary tumours.2 According to the criteria, such tumours occurring at dif-ferent locations must be histologically malig-nant and separated by normal mucosa, and one tumour must not be a metastasis of an-other. Thereafter, a greater awareness, im-proved diagnostic techniques and facilities account for the observed increase in the inci-dence of metachronous and synchronous ma-lignancies. Of patients with lung cancer, 3.2 -
50
Kurishima K et al /Multiple malignancies in lung cancer
9.7% are considered to have a metachronous or synchronous malignancy in various sites at presentation.3-5 This article summarizes our experience in lung cancer patients with metachronous or synchronous malignancies at various sites.
Methods
A total of 1194 consecutive lung cancer pa-tients, who were admitted to our division over a 29-year period up to August 2004, were retrospectively analyzed. In all patients, the diagnosis of lung cancer was confirmed
Table 1. Characteristics of lung cancer patients with multiple primary malignancies (n=98)
	Number of patients
Gender	
Male	78
Female	20
Age in years	
Range; median	45-85; 70
Histology	
Adenocarcinoma	35
Squamous cell carcinoma	40
Small cell carcinoma	19
Large cell carcinoma	3
Others	1
Performance status (ECOG)	
0-1 / 2-3 / 4	76/20/1
Clinical stage	
IA-B/IIA-B/IIIA/IIIB/IV	20/12/8/24/34
Malignant diseases	
Synchronous/matachronous/both  21/76/1	
Gastrointestinal	49
Urogenital	15
Lung	9
Head and neck	9
Others	19
Treatment	
Chemotherapy	37
Surgery	26
Radiation	20
Best supportive care	15
pathologically. Patients were classified using the International System for Staging Lung Cancer.6 The following criteria have been used for the designation of synchronous can-cers: (1) the tumours were anatomically separate; (2) the tumours were histologically dif-ferent. If the tumours were histologically the same ones, the gross appearance of each tu-mour was strongly suggestive of the primary
cancer.7
The criteria that we have used to determine metachronous cancers were related: (1) the tumours were anatomically separated; (2) tumour-free interval was at least 2 years.
At the time of admission, the past medical history including malignancies was taken from all patients. Staging procedures includ-ing physical examination, brain magnetic resonance imaging (MRI), chest computed tomo-graphic (CT) scan, abdominal CT scan or ul-trasonography, and bone scintigraphy were performed in all patients.
A Cox's proportional hazard model8 was used for the multivariate regression analysis to clarify the independent prognostic impor-tance of the following variables: gender, age, stage, performance status (PS), resectability at surgery, and existing multiple primary tu-mours.
Results
Among the 1194 patients with lung cancer, 98 (8.2%) patients were diagnosed as having metachronous and synchronous malignan-cies. All relevant data relating to the patients who had metachronous and synchronous ma-lignancies are shown in Table 1. The mean age at presentation of lung cancer was 70 years (range 45-85). Eighty-six (87.8%) of pa-tients were 60 years of age or more. There were 78 men and 20 women. Among 98 pa-tients, 76 had metachronous malignancies, 21 had synchronous malignancies, and 1 had both. As metachronous or synchronous ma-
Radiol Oncol 2005; 39(1): 49-53.
Kurishima K et al /Multiple malignancies in lung cancer
51
lignancy, gastric cancer found in 32 patients, colon and rectal cancers in 10, lung cancer in 9, uterus cancer in 6, and prostate cancer in 5, respectively.
A histological examination of lung cancer revealed a preponderance of squamous cell carcinoma (40 patients, 40.8%). There were 35 patients (35.7%) with adenocarcinoma, 19 pa-tients (19.4%) with small cell carcinoma and 3 patients (3.1%) with large cell carcinoma. This distribution of pathologic types is in striking contrast to that found in patients without multiple primary tumours, in whom there was a lower frequency of squamous cell car-cinomas (29.5%) and a higher frequency of adenocarcinomas (47.1%).
With regard to tobacco smoking, 77.6% of patients with multiple primary tumours were habitual smokers including 68 patients (69.4%) with 30 pack year or more history of smoking. Fifty-seven (85.1%) of 67 patients with aerogastrointestinal and head and neck cancers were smokers. On the other hand, 76.3% of patients without multiple primary tumours were smokers including 65.1% of pa-tients with 30-pack year or more history of smoking. There was no significant difference in smoking habit between the two groups (p = 0.3510). However, there was a significant difference in 30 pack year or more history of smoking in the two groups (p=0.0005).
In 98 patients with multiple primary can-cers, forty (40.8%) patients had stage IA-IIIA lung cancer, however, 26 (26.5%) had a surgi-cal resection. Twenty (20.4%) received chest irradiation, and 15 (15.3%) had only the best supportive care.
A Cox's proportional hazard model was used for the multivariate regression analysis to clarify the independent prognostic impor-tance of the following variables: gender, age, stage, PS, resectability at surgery, and exist-ing multiple primary tumours. For each variable, the proportional hazard assumption was examined graphically. In 1194 patients with lung cancer, gender (p<0.0001), stage (p<0.0001), PS (p<0.0001), and resectability at surgery (p<0.0001) were significantly related to the survival. In these patients, the existing multiple primary tumours were not proved to be a worse prognostic factor for the survival (p=0.1919). However, in 1018 patients with non-small cell lung cancer, the existing metachronous primary tumours were a sig-nificant prognostic factor for survival (p=0.0480) (Table 2).
Discussion
The criteria of double primary cancers enun-ciated by Warren and Gates2 are now gener-ally accepted. Of patients with lung cancer, an incidence for multiple primary tumours of 3.2 - 9.7% has been reported in the litera-ture.3-5 Cahan reported the incidence of sec-ond malignancies was 3.2% of patients with lung cancer.3 Antakli et al4 demonstrated 4.1% incidence of the second primary lung cancer in 1572 cases of cancer patients, the metachronous cancer comprised 60% and the synchronous are 40%. Reynolds et al5 found an unusually high association of another pri-mary cancer (9.7%) in cases of lung cancer,
Table 2. Prognostic factors in 1018 patients with non-small cell lung cancer determined by the Cox proportional hazard model
Variables	Coefficient	Standard error	p-value
Age	- 0.095	0.236	0.6884
Performance status	0. 752	0.100	<0.0001
Stage	- 0.633	0.122	<0.0001
Existing metachronous primary tumours	- 0.381	0.193	0.0480
Resectability at surgery	1.019	0.122	<0.0001
Radiol Oncol 2005; 39(1): 49-53.
52
Kurishima K et al /Multiple malignancies in lung cancer
and the analysis by the stage of lung cancer as the second primary cancer showed a higher incidence of associated malignancies in the early stage lung cancer than in the advanced stage. In our 1194 lung cancer patients, 98 (8.2%) of them had multiple primary malig-nancies; the metachronous malignant disease comprised 77.6% and the synchronous one 21.4%. Inconsistent with the report by Reynolds et al,5 more multiple primary tu-mours in our patients were detected in the advanced stage of lung cancer (in stage IIIA-IV 67.3%) than in the early stage(in IA-IIB 32.7%). We cannot explain the reason why the difference in incidence with regard to the stage was derived from. The precise inci-dence varies from series to series and might also be related to the length of the follow-up. Moreover, greater awareness and improved diagnostic techniques account for the appar-ent increase in incidence.
All retrospective studies published so far have revealed that multiple primary tumours occur more frequently than it would be ex-pected by chance in patients who have pri-mary lung cancer. Several environmental fac-tors are thought to play a role in the development of primary and second malignancies. The role of tobacco and/or alcohol use in sus-ceptible tissues of the upper aerodigestive tract is of major importance in multiple pri-mary tumours including lung cancer. In our patients, 85.1% of patients with aerogastroin-testinal and head and neck cancers had a smoking habit. The possible roles of immuni-ty, heredity, nutrition are not completely known. The development of the second pri-mary cancer has been well known following the chemotherapy,9-11 but in this series no pa-tients received chemotherapy for the first pri-mary cancers.
There have been reported that a grater per-centage of multiple primary cancers occur in the same organ or in the organs of the same system than in the unrelated organs.12 The as-sociation between unrelated organs, such as
Radiol Oncol 2005; 39(1): 49-53.
lung, stomach and kidney may indicate the circulation of carcinogenic metabolites.13 Our study showed that the first primary tumours occurred most commonly in gastrointestinal tract, followed by lung and uterus. Our re-sults also revealed that the existing multiple primary tumours were not proved to be a worse prognostic factor for the survival in 1194 patients with lung cancer. However, in 1018 patients with non-small cell lung cancer, the existing metachronous primary tumours were a significant prognostic factor for the survival.
The second primary cancers may develop as much as 30 years later, and therefore, the possibility of a second cancer should be kept in mind in all cancer cases although the con-cept of a 5-year cure may be valid for the first cancer in most cases.14 Recognizing that the incidence of the development of the second malignant tumour depends on the exactitude of evaluation, length of follow-up, curability of the first primary tumour, and patterns of tobacco and alcohol use, we should approach the problem of the second malignancies in lung cancer from several angles in the future. When a second primary cancer appears, a careful search for the metastatic disease should be made and a consideration for the curative intervention should be given. An aggressive curative approach conserving as much organ function as possible still offers the greatest chance for the long-term survival.
References
1.   Billroth T. Die allgemeine chirurgische pathologie und therapie in 51 vorlesungen. In: Reimer G, editor. Handbuch für studierende und ärzte. Berlin; 1882. p. 908.
2.   Warren S, Gates O. Multiple primary malignant tumors. Am J Cancer 1932; 16: 1358-414.
3.   Cahan WG. Lung cancer associated with cancer primary to other sites. Am J Surg 1955; 89: 494-514.
Kurishima K et al /Multiple malignancies in lung cancer
53
4.   Antakli T, Schaefer RF, Rutherford JE, Reed RC. Second primary lung cancer. Ann Thorac Surg 1995; 59: 863-6.
5.   Reynolds RD, Pajak TF, Greenberg BR, Shirley JH, Lucus RN, Hill RP, et al. Lung cancer as a second primary. Cancer 1978; 42: 2887-93.
6.   Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest 1997; 111: 1710-7.
7.   Carter D, Eggleston JC. Tumours of the lower res-piratory tract. Washington, DC: Armed Forces Institute of Pathology; 1974. p. 70-94; p 113-27.
8.   Cox DR. Regression models and life tables. J Royal Stat Soc B 1972; 34: 187-220.
9.   Johnson BE, Linnoila RI, Williams JP, Venzon DJ, Okunieff P, Anderson GB, et al. Risk of second aerodigestive cancers increases in patients who survive free of small-cell lung cancer for more than 2 years. J Clin Oncol 1995; 13: 101-11.
10. Johnson BE. Second lung cancers in patients after treatment for an initial lung cancer. J Natl Cancer Inst 1998; 90: 1335-45.
11. Waller CF, Fetscher S, Lange W. Secondary chron-ic myelogenous leukemia after chemotherapy fol-lowed by adjuvant radiotherapy for small cell lung cancer. Leuk Res 1999; 23: 961-4.
12. Watanabe S, Kodama T, Shimosato Y, Arimoto H, Sugimura T, Suemasu K, et al. Multiple primary cancers in 5456 autopsy cases in the National Cancer Center of Japan. J Natl Cancer Inst 1984; 72: 1021-7.
13. Barbin A. Role of etheno DNA adducts in carcino-genesis induced by vinyl chloride in rats. IARC Sci Publ 1999; 150: 303-13.
14. Kogelnik HD, Fletcher GH, Jesse RH. Clinical course of patients with squamous cell carcinoma of the upper respiratory and digestive tracts with no evidence of disease 5 years after initial treat-ment. Radiology 1975; 115: 432-27.
Radiol Oncol 2005; 39(1): 49-53.