Eur J Cardiothorac Surg 2001;19:908-913
© 2001 Elsevier Science NL
Survival, disease-free interval, and associated tumor features in patients with colon/rectal carcinomas and their resected intra-pulmonary metastases
Stefan Zinka,
Gian Kayserb,
Hans-Joachim Gabiusc,
Klaus Kaysera
a Department of Pathology, Thoraxklinik, Heidelberg, Germany
b Department of Cardio-Vascular Surgery, University of Freiburg, Freiburg, Germany
c Institute of Physiological Chemistry, University Munich, Munich, Germany
Received 18 October 2000;
received in revised form 22 February 2001;
accepted 3 April 2001.
Corresponding author. Tel.: +49-6221-396496; fax: +49-6221-396238
e-mail: klkayser{at}lung.de
 |
Abstract
|
|---|
Objective: Colon/rectum cancer often presents with intrapulmonary metastases. Surgical resection can be performed in a selected group of patients. In this study, the search for possible prognostic factors of patients with primary colon/rectum cancer and lung metastases was performed. Methods: Medical records of 110 patients operated on pulmonary metastases of primary colon/rectum cancer were reviewed. The clinical parameters include age, sex, pTNM/UICC stage, grading, localization, surgical and adjuvant therapy of the primary cancer. The number, maximum diameter and total intra-thoracic resected tumor-mass (load), the pre-thoracotomy serum carcinoembryonic antigen (CEA) levels, localization of the metastases (uni- vs. bilateral), the presence of hilar/mediastinal tumor-infiltrated lymph nodes, the surgical procedure and performed therapy schemes of lung metastases were recorded. Results: The cumulated 5- and 10-year total survival after diagnosis of the primary carcinomas was estimated to 71 and 33.7%, respectively. After resection of the pulmonary metastases, the 3- and 5-year post-thoracotomy survival measured 57 and 32.6%, respectively. The median time interval between diagnosis of the primary cancer and thoracotomy (disease free interval (DFI)) was found to be 35 months. A non-negligible percentage of patients (15.4%) displayed limited tumor stages of the primary cancer (pT1/2, pN0). The median diameter of the largest metastasis measured 28 mm, and the median resected intrathoracic tumor-load was calculated to 11.4 cm3. In only 8 patients hilar or mediastinal tumor-involved lymph nodes were found. A potentially curative resection of lung metastases was recorded in 96 patients. The overall survival was significantly correlated with the DFI and the number of intrapulmonary metastases. The DFI correlated significantly with the tumor load and the number of metastases; the post-thoracotomy survival with the number of metastases, tumor-load and pre-thoracotomy serum CEA level. Treatment, stage and grade of the primary cancer, occurrence of liver metastases and local recurrences, mode of treatment of metastases and postoperative residual stage had no significant correlation with either total nor post-thoracotomy survival. Conclusions: Pulmonary metastases occur even in patients with limited tumor-stages of primary colon/rectum cancer. DFI is the major parameter to estimate the total survival of patients with lung metastases. The survival after thoracotomy depends on the number of metastases, the intrapulmonary tumor load and the presence of elevated serum CEA level prior to thoracotomy.
Key Words: Colon/rectum cancer Lung metastasis Tumorload Serum carcinoembryonic antigen level Survival
 |
1. Introduction
|
|---|
Second to lung cancer in men and breast cancer in women, colon/rectum cancer is an important cause of death with increasing incidence over the past 20 years [1,2]. After potential curative resection of the primary tumor, dedicated surveillance in order to detect distant metastases or local recurrence in early stage is recommended [3]. The measurement of the serum carcinoembryonic antigen (CEA) level can be used in patients potentially being candidates for resection of liver and lung metastases [4]. The lungs are the most common extra-abdominal site of metastases in patients affected by cancer of the large bowel [5]. The treatment of metastases is controversially discussed; however, non-surgical treatment still is of only limited benefit to the patients [68], and surgical intervention currently is the therapy of choice if a metastatic intrapulmonary process is confined [9,10]. About 10% of patients with primary colon/rectum cancer will develop lung metastases [11]. The majority of patients additionally have recurrent extra-pulmonary disease. Only 24% of patients with recurrent colon/rectum cancer display metastases limited to the lungs, and surgical treatment of lung metastases may be indicated [10]. The 5-year post-thoracotomy survival of this selected group of patients is reported to range between 9 and 40.5%. The overall 5-year survival rate of potentially curative resected cancer of the large bowel has been estimated to 71% [12]. Several reports report a median time interval between the resection of the primary tumor and the pulmonary metastases (disease free interval, DFI) between 29 and 41 months [1115].
The aim of this study is to evaluate parameters of the primary cancer and the pulmonary metastases which might be of prognostic value for patients with potentially curative resected colon/rectum carcinomas and solitary pulmonary metastases treated surgically.
 |
2. Material and methods
|
|---|
The medical records of 110 patients with potentially curative resected (R0) colon/rectum carcinomas and surgically treated lung metastases operated on at the Thoraxklinik Heidelberg between 1.1.1988 and 1.3.1999 were reviewed. All patients who underwent a potentially curative resection of colon/rectum metastases were included in this study. The surgical approach was based upon routinely tumor follow-up after resection of the primary colon/rectum carcinoma, and specifically upon findings in CT scans. Bronchoscopy prior to operation was a routine procedure, whereas a bilateral surgical approach was not systematically performed. In case of bilateral abnormal findings a sequential operation of left and right lung (or vice versa) was the normal procedure. The patients underwent a chemotherapy of the colon/rectum carcinoma in case of extended tumor stage. The analyzed tumor features include age, sex of patients, maximum diameter of primary tumor, localization, pTNM/UICC stage, development of local recurrent disease and distant metastasis in the liver, number and maximum diameter of metastases, tumor load of metastases (calculated from the number and maximum diameter of occurring lung metastases), presence of mediastinal or hilar lymph node involvement, serum CEA levels prior to resection of the lung metastases, post-operative treatment, and surgical procedures of primary and secondary tumors. The statistical tests include parametric and non-parametric correlation tests, log-rank and COX regression tests including KaplanMeier analysis [16], and multivariate discriminant analysis (number cruncher statistical system, NCSS60 program package, Utah, USA).
 |
3. Results
|
|---|
The median age of patients at time of the diagnosis of the colon/rectum cancer was 58 years (range: 3478 years). Patients with primary tumors located in the rectum (55 years) were significantly younger compared to those with colon carcinomas (59 years) (P<0.01). Sixty-eight patients were men and 42 were women. In 104 cases (94.5%) the primary tumor stage was accessible, which was not documented in six cases (5.5%). The distribution of pTNM stages of 104 patients with lung metastases is demonstrated in Table 1. Although the majority of patients was operated on in advanced tumor stages, and 15% of patients displayed limited tumor stages (pT1/2, pN0). The majority (53.6%) of tumors arose from the colon (n=59) and in only 51 patients from the rectum. The rectum carcinomas were located in the lower region in 10 patients, in the middle part in 26 patients and in the upper part in 12 patients. Colon cancer was found in the sigmoid in 32 patients and in the descend colon in seven patients. Right-sided cancer was detected in the transverse colon in one patient only, in the ascending part in 11 patients, and in the caecum in five patients. All patients underwent a radical resection of the primary tumor. In the majority of cases an anterior resection was performed (n=47) followed by a hemicolectomy (n=36), abdominosacral resection (ASR, n=15), abdominoperineal resection (APR, n=11), and transanal disc-excision (one patient). The median follow-up time after resection of the primary tumor was calculated to 70 months (range: 14252 months). No significant association of tumor localization with the DFI or total survival could be obtained. 39 patients (35.5%) underwent a course of postoperative adjuvant therapy. Seventeen patients received a postoperative chemotherapy, seven patients radiotherapy, and 15 patients a combined radio-chemotherapy. Local recurrence of the primary cancer occurred in 15 patients (13.6%). Fourteen patients could be potentially curative resected, residual tumor mass after resection was found in only one patient. The presence of local recurrences had no statistically significant correlation neither with (DFI) nor the total survival. Liver metastases were found in 20 patients (18.2%), in ten cases metastatic hepatic spread was detected at time of primary diagnosis. All liver metastases, detected at primary diagnosis, could be resected with clean tumor boundaries. The presence of liver metastases was not associated with the DFI or the total survival at a statistically significant level (P>0.05). The median DFI was calculated to 35 months (range: 2216 months) and the follow-up after resection of pulmonary metastases to 31 months (range: 4142 months). The intra-pulmonary tumor load was correlated with the pre-thoracomy CEA serum levels at a significant level (Table 2). In the majority of cases only one metastasis was detected (n=56). Twenty-one patients had two, 10 patients three, seven patients four, one had five, three had six, and four patients had seven metastases. Two patients had eight and nine metastases, respectively. The maximum number of lung metastases accounted to 18 tumor nodules (one patient). The mean number was computed to 2.6 intrapulmonary metastases per patient. The median of the maximum diameter defined in the largest metastasis measured 28 mm (range: 6110 mm) and the median total intrapulmonary tumor load was calculated to 11.4 cm3 (range: 0.11342 cm3). The thoracic intervention was performed most frequently (83 patients) through lateral thoracotomy (76%). Median sternotomy was documented in 21 patients (19%), transverse thoracotomy in four patients (3.6%), anterior thoracotomy in one patient, and thoracoscopic guided intervention in one case. The mode of the pulmonary resection is presented in Fig. 1. The pulmonary resection was potentially curative in 96 patients. Residual tumor were documented in 14 patients (R1: 10 patients, R2: 4 patients). The presence of residual tumor and/or the presence of mediastinal or hilar lymph node involvement (eight patients) was not statistically significantly associated with the post-surgical survival. After thoracotomy 15 patients underwent adjuvant therapy. Fourteen patients had chemotherapy and a single patient had combined radio-chemotherapy. Twenty-nine patients developed recurrent lung metastases, and 23 patients underwent a secondary operation. The surgically treated patients had a statistically significant longer post-thoracotomy survival (P=0.03). A third lung operation was undertaken in eight patients, and one patient underwent four intrapulmonary resections of metastases. The pre-thoracotomy serum CEA level was documented in 101 cases (92%) and found to be elevated (
5 ng/ml) in 42 patients. The DFI is statistically significantly associated with the intra-pulmonary tumor load and the maximum diameter of metastases (Table 3). The post-thoracotomy survival was statistically significantly correlated with the number of metastases (Fig. 2), the tumor load (Fig. 3), and the pre-thoracotomy CEA serum level (Fig. 4), whereas the total survival only displayed a correlation with the number of intra-pulmonary metastases and the length of the DFI (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 2. Pre-thoracotomy CEA serum levels and intrapulmonary tumor load, number of metastases, and maximum diameter of pulmonary metastases (largest metastasis in case of multiple tumors)
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Relationship of the disease-free interval (DFI) and tumor features of primary and secondary tumors (log rank test)
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Multivariant analysis of parameters associated with total survival and post-thoracotomy survival (Cox-regression)
|
|
 |
4. Discussion
|
|---|
Patients with colon/rectum carcinomas are subject to develop lung metastases even at early tumor stages of the primary cancer. The cumulated 2-, 3-, 5-, and 10-year overall survival rates were calculated to 98, 92, 71 and 34%, respectively. In contrast to the prognosis of non-selected patients developing colon/rectum carcinomas [3,17], the total survival of this selected cohort has not been found to be significantly associated with the stage (pTNM, UICC [18]) (P=0.2) or grade of the primary tumor. However, the DFI, which is significantly correlated with the total survival (P<0.001), measured 48.5 months in stage I (n=14 patients), 35 months in Stage II (n=28 patients), and 36 months in Stage III (n=40 patients) disease. The 22 patients at stage IV (18.5 months) had a significantly (P<0.002) shorter DFI. The patients age and sex, or mode of primary resection, postoperative adjuvant therapy, development of liver metastases or local recurrence of the primary tumor, no significant correlation to the overall survival could be observed. Calculated by multivariant analysis, the DFI, the number of lung metastases, the pulmonary metastatic tumor-mass and the occurrence of lymphonodal tumor-infiltration of the primary tumor (pN) was found to significantly correlate with the total survival of patients. After thoracotomy the 3-year survival was calculated to 57% and the 5-year survival to 32.6%, postoperatively. The median length of the DFI measures about 3 years, i.e. an intensive control of operated tumor patients should be performed at least within this period. The median survival after post-thoracotomy amounts to 41 months, i.e. is comparable or even in favor to that of patients suffering from primary bronchial carcinomas with comparable tumor mass (tumor load) [19]. This statement is important as the cohort with lung metastases including patients with multiple tumors is comparable to patients with primary lung tumors [19]. In this study only 51% (n=56) of patients had a singular metastasis and 49% multiple intrapulmonary tumors. In studies with comparable number of patients Okamura et al. [20] report 67.3% and McAfee et al. [21] report 70.5% of patients with singular metastasis at the time of operation. In other studies [2225] singular metastasis was present in 62.978.9% of patients. In the study of Okamura et al. [20] the 5-year survival rate was calculated to 40.5% and McAfee et al. [21] reported 30.5% of patients survived 5 years. Other studies [11,13,14,2225] calculated 5 year survival rates between 9% [5] and 38.7% [15]. In contrast to the data of Okamura et al. [20], we could not find, that cancer patients had different survival rates between patients with/without liver metastases or with/without developed local recurrences at the site of the primary cancer. Furthermore, none of the clinical parameters (age at time of diagnosis of lung metastases, sex, type and extent of surgical intervention on lung metastases) had a potential prognostic value in predicting the post-thoracotomy survival. Additionally, in agreement with data of other studies [5,11,14,15,21,23], the tumor stage of the colon/rectal cancer had no influence on the survival after resection of metastases at a statistically significant level. Reviewing the literature, only Okamura et al. [20] and McCormack et al. [25] reported a significantly improved post-thoracotomy survival of patients whose large bowel cancer was staged to Duke's A. Although the DFI was statistically significantly correlated with the total survival of patients, a statistically significant correlation between the DFI and the post-thoracotomy survival could not be found. These findings are in agreement to most of the other studies [5,13,15,2023,25]; however, in contrast to the results of Briester et al. [14]. Interestingly, the patients with an elevated serum CEA level prior to thoracotomy had a significantly reduced survival after pulmonary metastasectomy. These findings have also been reported by McAfee et al. [21] in their study comprising 52 patients with recorded serum CEA level. Patients with mediastinal or hilar lymph node involvement (n=8) displayed no major differences in survival after thoracotomy in comparison to their normal counterparts. Goya et al. [23] reported similar results whereas Okamura et al. [20] found a statistically significant negative association of hilar or mediastinal lymph node metastases with the post-thoracotomy survival.
In aggregate, the total 5-year survival rate of patients with potentially curatively operated colon/rectum carcinomas and subsequent excised lung metastases measures 71% which is in the range of the total survival of large bowel cancer patients treated surgically in a potentially curative manner [12].
 |
Acknowledgments
|
|---|
The financial support from the Verein zur Förderung des biologisch-technischen Fortschritts in der Medizin and the Dr M.-S. Scheel-Stiftung für Krebsforschung is gratefully acknowledged.
 |
References
|
|---|
-
Gatta G., Faivre J., Capocaccia R., Ponz de Leon M., The EUROCARE Working Group. Survival of colorectal cancer patients in Europe during the period 19781989. Eur J Cancer 1998;34:2176-2183.
-
Levi F., Lucchini F., Negri E., La Vecchia C. Worldwide patterns of cancer mortality, 19901994. Eur J Cancer Prev 1999;8:381-400.[Medline]
-
Markowitz A.J., Winawer S.J. Screening and surveillance for colorectal cancer. Semin Oncol 1999;26:485-498.[Medline]
-
American Society of Clinical Oncology. Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer. J Clin Oncol 1996;14:2843-2877.[Abstract/Free Full Text]
-
Wilking N., Petrelli N.J., Herrera L., Regal A.M., Mittelmann A. Surgical resection of pulmonary metastases from colorectal adenocarcinoma. Dis Colon Rectum 1985;28:562-564.[Medline]
-
Fong Y., Cohen A.M., Fortner J.G., Enker W.E., Turnbull A.D., Coit D.G., Marrero A.M., Prasad M., Blumgart C.H., Brennan M.F. Liver resection for colorectal metastases. J Clin Oncol 1997;15:938-946.[Abstract/Free Full Text]
-
Karapetis C.S., Yip D., Harper P.G. The treatment of metastatic colorectal cancer. Int J Clin Pract 1999;53:287-294.[Medline]
-
Koehne C.H., Midgley R., Seymour M., Kerr D.J. Advanced colorectal cancer: which regimes should we recommend?. Ann Oncol 1999;10:877-882.[Free Full Text]
-
McCormack P.M., Ginsberg R.J. Current management of colorectal metastases to the lung. Chest Surg Clin N Am 1998;8:119-126.[Medline]
-
Turk P.S., Wanebo H.J. Results of surgical treatment of non-hepatic recurrence of colorectal carcinoma. Cancer 1993;71:4267-4277.[Medline]
-
Shirouzu K., Isomoto H., Hayashi A., Nagamatsu Y., Kakegawa T. Surgical treatment for patients with pulmonary metastases after resection of primary colorectal carcinoma. Cancer 1995;76:393-398.[Medline]
-
Lehnert T., Schaible A., Herfarth C. Oncologic principles associated with colon carcinoma. Diagnosis, therapy and follow-up. Chirurg 1999;70:499-510.[Medline]
-
Mansel J.K., Zinsmeister A.R., Pairolero P.C., Jett J.R. Pulmonary resection of metastatic colorectal adenocarcinoma. A ten year experience. Chest 1986;89:109-112.[Abstract/Free Full Text]
-
Briester S.J., De Varennes B., Gordon P.H., Sheiner N.M., Pym J. Contemporary operative management of pulmonary metastases of colorectal origin. Dis Colon Rectum 1988;31:786-792.[Medline]
-
Baron O., Amini M., Duveau D., Despins P., Sagan C.A., Michaud J.L. Surgical resection of pulmonary metastases from colorectal carcinoma. Five-year survival and main prognostic factors. Eur J Cardiothorac Surg 1996;10:347-351.[Abstract]
-
Kaplan E.L., Meier P. Non-parametric estimations from incomplete observations. Am Stat Assoc J 1953;53:457-481.
-
Compton C.C. Pathology report in colon cancer: what is prognostically important. Dig Dis 1999;17:67-79.[Medline]
-
American Joint Committee on Cancer. AJCC Cancer staging manual, 5th edition Philadelphia: Lippincott-Raven, 1997:83-90.
-
Kayser K.W., Bülzebruck H., Merkle N., Vogt-Moykopf I. Survival of operated bronchus carcinoma patients: a prospective study. J Surg Oncol 1987;36:84-92.[Medline]
-
Okomura S., Kondo H., Tsuboi M., Nakayama H., Asamura H., Tsuchiya R., Navake T. Pulmonary resection for metastatic colorectal cancer: experiences with 159 patients. J Thorac Cardiovasc Surg 1996;112:867-874.[Abstract/Free Full Text]
-
McAfee M.K., Allen M.S., Trastek V.F., Ilstrup D.M., Deschamps C., Pairolero P.C. Colorectal Lung Metastases: Results of surgical excision. Ann Thorac Surg 1992;53:780-786.[Abstract]
-
Cahan W.G., Castro E.B., Hajdu S.I. Therapeutic pulmonary resection of colonic carcinoma metastatic to lung. Dis Colon Rectum 1974;17:302-309.[Medline]
-
Goya T., Miyazawa N., Kondo H., Tsuchiya R., Naruke T., Suemasu K. Surgical resection of pulmonary metastases from colorectal cancer 10 year follow-up. Cancer 1989;64:1418-1421.[Medline]
-
Kamiyoshihara M., Hirai T., Kawashima O., Morishita Y. Resection of pulmonary metastases in six patients with disease-free interval greater than 10 years. Ann Thorac Surg 1998;66:231-233.[Abstract/Free Full Text]
-
McCormack P.M., Attiyeh F.F. Resected Pulmonary Metastases from Colorectal Cancer. Dis Colon Rectum 1979;22:553-556.[Medline]
This article has been cited by other articles:

|
 |

|
 |
 
R. Maeda, N. Isowa, H. Onuma, H. Miura, T. Harada, H. Touge, H. Tokuyasu, and Y. Kawasaki
Pulmonary resection for metastases from colorectal carcinoma
Interactive CardioVascular and Thoracic Surgery,
October 1, 2009;
9(4):
640 - 644.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
M. W. Onaitis, R. P. Petersen, J. C. Haney, L. Saltz, B. Park, R. Flores, N. Rizk, M. S. Bains, J. Dycoco, T. A. D'Amico, et al.
Prognostic Factors for Recurrence After Pulmonary Resection of Colorectal Cancer Metastases.
Ann. Thorac. Surg.,
June 1, 2009;
87(6):
1684 - 1688.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
N. Rama, A. Monteiro, J. E. Bernardo, L. Eugenio, and M. J. Antunes
Lung metastases from colorectal cancer: surgical resection and prognostic factors
Eur. J. Cardiothorac. Surg.,
March 1, 2009;
35(3):
444 - 449.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
J. Nakajima, T. Murakawa, T. Fukami, A. Sano, M. Sugiura, and S. Takamoto
Is Finger Palpation at Operation Indispensable for Pulmonary Metastasectomy in Colorectal Cancer?
Ann. Thorac. Surg.,
November 1, 2007;
84(5):
1680 - 1684.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
G. Melloni, C. Doglioni, A. Bandiera, A. Carretta, P. Ciriaco, G. Arrigoni, and P. Zannini
Prognostic Factors and Analysis of Microsatellite Instability in Resected Pulmonary Metastases From Colorectal Carcinoma
Ann. Thorac. Surg.,
June 1, 2006;
81(6):
2008 - 2013.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
I. Chau, M. J. Allen, D. Cunningham, A. R. Norman, G. Brown, H. E.R. Ford, N. Tebbutt, D. Tait, M. Hill, P. J. Ross, et al.
The Value of Routine Serum Carcino-Embryonic Antigen Measurement and Computed Tomography in the Surveillance of Patients After Adjuvant Chemotherapy for Colorectal Cancer
J. Clin. Oncol.,
April 15, 2004;
22(8):
1420 - 1429.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
R. H.V. Reddy, B. Kumar, R. Shah, S. Mirsadraee, K. Papagiannopoulos, P. Lodge, and J. A.C. Thorpe
Staged pulmonary and hepatic metastasectomy in colorectal cancer--is it worth it?
Eur. J. Cardiothorac. Surg.,
February 1, 2004;
25(2):
151 - 154.
[Abstract]
[Full Text]
[PDF]
|
 |
|