BACKGROUND: STUDY DESIGN: RESULTS: CONCLUSIONS:

Transcription

1 Improved Survival after Resection of Liver and Lung Colorectal Metastases Compared with Liver-Only Metastases: A Study of 112 Patients with Limited Lung Metastatic Disease Antoine Brouquet, MD, Jean Nicolas Vauthey, MD, FACS, Carlo M Contreras, MD, Garrett L Walsh, MD, FACS, Ara A Vaporciyan, MD, FACS, Stephen G Swisher, MD, FACS, Steven A Curley, MD, FACS, Reza J Mehran, MD, FACS, Eddie K Abdalla, MD, FACS BACKGROUND: STUDY DESIGN: RESULTS: CONCLUSIONS: Lung metastases are considered a poor prognostic factor in patients with resectable colorectal liver metastases. We reviewed records of 1,260 consecutive patients with liver-only or liver-plus-lung (L L) metastases from colorectal cancer who underwent resection with curative intent (1995 to 2009). Survival and prognostic factors were analyzed. There were 112 patients who underwent resection of L L (mean 2 liver, 2 lung metastases). Mean tumor sizes were 3 cm and 1 cm, respectively. Thirty-four (31%) had bilateral lung metastases. Ten (9%) had synchronous L L metastases, 60 (54%) had diagnosis of lung metastases within 1 year of liver resection. Most (108 of 112, 96%) had resection of liver before or at the same time as lung. Preoperative chemotherapy was used in 77 (69%) before liver resection and 56 (50%) before lung resection. Among L L patients, no postoperative deaths occurred; postoperative morbidity rates were 26% after liver resection and 4% after lung resection. After a median of 49 months follow-up, L L patients (n 112) had better survival than liver only (n 1,148) (5-year overall survival, L L, 50% vs liver only, 40%; p 0.01). CEA level 5 ng/dl (hazard ratio [HR] 2.1, 95% CI 1.1 to 4.4, p 0.04) and rectal primary (HR 2.9, 95% CI 1.4 to 6, p 0.004) were associated with worse survival in L L patients. The survival rate for patients who undergo resection of L L metastases from colorectal cancer is greater than the survival rate of the general population of patients who undergo resection of liver metastases only. The presence of resectable lung metastases is neither a poor prognostic factor nor a contraindication to resection of liver metastases. (J Am Coll Surg 2011;213: by the American College of Surgeons) The presence of extrahepatic disease in patients with colorectal liver metastases (CLM) is traditionally associated with poor long-term prognosis and is still considered a relative contraindication to resection. 1-3 Resection of CLM Disclosure Information: Nothing to disclose. Supported in part by the National Institutes of Health through MD Anderson Cancer Center Support Grant CA Presented at the Western Surgical Association 118th Scientific Session, Chicago, IL, November Received January 7, 2011; Revised April 14, 2011; Accepted May 4, From the Departments of Surgical Oncology (Brouquet, Vauthey, Contreras, Curley, Abdalla) and Thoracic Surgery (Walsh, Vaporciyan, Swisher, Mehran), The University of Texas MD Anderson Cancer Center, Houston, TX. Correspondence address: Eddie K Abdalla, MD, Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, 1400 Hermann Pressler Dr, Unit 1484, Houston, TX eabdalla@ mdanderson.org or lung metastases from colorectal cancer is associated with a 5-year survival rate of up to 58% and 68%, respectively. 4,5 Improved surgical technique and better patient selection 6,7 probably account for these good results; the increasing use of combined modality approaches, including chemotherapy, may contribute to improving outcomes. 8,9 As outcomes have improved, interest has grown in a more aggressive oncosurgical approach to patients with liver and lung colorectal metastases, including surgery, despite the concern that the presence of metastatic sites in multiple organs might be a sign of diffuse cancer dissemination. To date, few studies have evaluated the outcomes of patients who underwent resection of both liver and lung colorectal metastases, and only 1 study includes more than 100 patients (treated over a 20-year time period). 13 No comparative data are available that assess 2011 by the American College of Surgeons ISSN /11/$36.00 Published by Elsevier Inc. 62 doi: /j.jamcollsurg

2 Vol. 213, No. 1, July 2011 Brouquet et al Resection of Colorectal Liver and Lung Metastases 63 Abbreviations and Acronyms CLM colorectal liver metastases L L liver lung metastases LV levamisole outcomes after resection of liver and lung metastases vs liver-only metastases. We hypothesized that recent improvements in multidisciplinary management and patient selection of patients with stage IV colorectal cancer would enable similar outcomes after resection in patients with both liver and lung colorectal metastases to outcomes after resection of CLM alone. In this study, the feasibility and outcomes after resection of both liver and lung colorectal metastases (synchronous or metachronous) were analyzed in a large series of patients and compared with outcomes for those who underwent resection of CLM alone. Predictors of outcomes in patients undergoing resection of both liver and lung colorectal metastases were evaluated. METHODS Patients Between December 1995 and May 2009, data on 1,260 consecutive patients who underwent surgical resection for CLM at one institution were collected prospectively. Of these patients, 117 underwent liver and lung resection for metastatic colorectal cancer. Patients in whom the diagnosis of metastatic colorectal cancer was not confirmed on pathology were excluded. Radiofrequency ablation was used in addition to resection at the initial operation for CLM at the surgeon s discretion for unresectable disease in the remnant liver. The study was approved by the Institutional Review Board of MD Anderson Cancer Center. Resection of liver plus lung metastases Patients commonly received preoperative chemotherapy before resection of liver metastases. Typically, resection of CLM was performed after a short course of chemotherapy (4 to 6 cycles) in patients with resectable liver metastases and as soon as the lesions became resectable in patients with more advanced disease. 14 In contrast, indications for chemotherapy before resection of lung metastases were more restricted, especially in the case of small lesions, because chemotherapy-induced shrinkage of small lung metastases makes intraoperative identification and resection more difficult and because of the lack of data supporting a preoperative approach to chemotherapy in patients with lung metastases. Whether preoperative chemotherapy was or was not delivered, resection of progressing metastatic disease was uncommon, and liver and lung resections were performed only with curative intent, ie, only if it was believed that a complete tumor resection (R0) could be achieved. Progression was defined as increasing diameter of lesions measured on cross-sectional imaging. Resectability was determined not on number or size of liver or lung metastases but on potential to completely resect all metastatic deposits ever identified on imaging, leaving an adequate liver remnant 15 or lung function. At the time of resection of liver metastases, in all patients, systematic hepatic intraoperative ultrasonography was carried out and a complete exploration of the peritoneal cavity was performed to rule out extrahepatic spread of the disease. Major hepatectomy was defined as hepatic resection including 3 or more contiguous liver segments. Resection of pulmonary metastases was performed at the same time as resection of CLM (combined) or at another operation using either a transdiaphragmatic, a thoracotomy, or a video-assisted thoracoscopic approach as indicated. Some patients with bilateral pulmonary metastases underwent resection of all the lesions at the same time via median sternotomy; others underwent staged resection via bilateral thoracotomies or a video-assisted thoracoscopic approach. Lymphadenectomy in the chest was performed as indicated. Data on postoperative morbidity and 90-day mortality were collected prospectively and postoperative complications were classified according to their severity. 16 Long-term outcomes The decision whether to deliver additional chemotherapy after resection was made at the time of the follow-up postoperative visit on an individualized basis, based on the presence of possible residual disease (R0, R1), response to preoperative chemotherapy on imaging, 7,17 tolerance and number of cycles of preoperative chemotherapy, and pathologic response, at the discretion of the treating oncologist. 8 Patients were reassessed by physical examination, CEA serum level, multiphase (liver protocol) CT scan, or MRI and chest radiography or CT every 3 to 4 months after resection, and decisions about further treatment were made according to the findings on follow-up work-up. Statistical analysis Quantitative and qualitative variables were expressed as mean standard deviation, median (range), and frequency. Overall survival was calculated using the Kaplan Meier method from the date of the resection of the last metastatic site (pulmonary or hepatic) in patients who underwent resection of liver plus lung colorectal metastases and from the date of liver resection in patients who underwent resection of CLM alone. Disease-free survival was calculated from the date of resection of the last metastatic

3 64 Brouquet et al Resection of Colorectal Liver and Lung Metastases J Am Coll Surg Table 1. Preoperative Characteristics of 112 Patients Who Underwent Resection of Liver and Lung Metastases from Colorectal Cancer Liver plus lung (n 112) Liver only (n 1,148) p Value Characteristic Mean age, y (SD) 55 (11) 57 (12) NS Sex (M/F) 78/34 678/ Rectal primary, n (%) 32 (29) 254 (22) NS Node positive, n (%) 73 (65) 697 (61) NS Liver metastases Synchronous, n (%) 50 (45) 588 (51) NS Mean number (SD) 2.5 (2) 2.6 (2.7) NS Mean size, cm (SD) 3.5 (2.5) 3.6 (2.7) NS Preoperative chemotherapy, n (%) 77 (69) 655 (57) 0.01 Median preoperative CEA serum level, ng/ml (range) 3.2 (1 400) 3.7 (1 1501) NS Lung metastases Synchronous, n (%) 10 (9) Bilateral, n (%) 34 (31) Mean number (SD) 2.3 (2.1) Mean size, cm (SD) 1.5 (1.2) Preoperative chemotherapy before resection of lung metastases, n (%) 56 (50) Median preoperative CEA serum level, ng/ml (range) 1 (1 67) site (liver or lung). For detection of factors associated with survival in patients who underwent resection of liver plus lung colorectal metastases, univariate analysis was used to examine the relationship between the survival and the following variables: age older than 60 years, sex, rectal (vs other) location of the primary tumor, presence of regional lymph node metastases (vs absence), synchronous (vs metachronous) CLM, multiple (vs single) CLM, size of liver metastases measured at the time of diagnosis (greater or less than 5 cm), preoperative serum CEA level 5 ng/dl, preoperative chemotherapy before resection of CLM, positive (vs negative) surgical margins at resection of CLM, postoperative chemotherapy after resection of CLM, synchronous (vs metachronous) lung metastases, progressionfree interval (greater or less than 12 months) between the diagnosis of liver and lung metastases, multiple (vs single) lung metastases, size of lung metastases measured at the time of diagnosis (greater or less than 2 cm), preoperative chemotherapy for lung metastases, and positive (vs negative) surgical margins at resection of lung metastases. All variables associated with survival with p 0.2 in a univariate proportional hazard model were subsequently entered into a Cox multivariate regression model with backward elimination. Values of p 0.05 were considered statistically significant. Comparisons between groups were analyzed using the chi-square or Fischer exact test for proportions, the Mann-Whitney U test for medians, and the Student t- test for means, as appropriate. Statistical analysis was performed using the statistical software package SPSS version 17.2 (SPSS). RESULTS Patient population Among the 117 patients who underwent both liver and lung resection for a suspicion of metastatic colorectal cancer, pathologic evaluation revealed that 5 patients had resection of a benign lesion of the lung (n 5) or concomitant mediastinal malignancy (n 1) and were excluded. One hundred twelve patients had resection of liver plus lung metastases of colorectal origin, based on pathologic analysis of surgical specimens. Clinicopathologic features of these 112 patients are summarized in Table 1. The site of first metastasis was the liver only in 92 patients (82%), liver and lung in 15 patients (13%), and lung only in 5 patients (5%). The majority of patients had multiple CLM (62 of 112; 55%) and 21 patients (19%) had a tumor size of 5 cm or more. The majority of patients (77 of 112, 69%) received preoperative chemotherapy before resection of CLM, and first line chemotherapy regimens included 5-fluorouracil and levamisole (5-FU and LV) in 7 patients, 5-FU and LV oxaliplatin in 37 patients, and 5-FU and LV irinotecan in 33 patients. The mean number of cycles of preoperative chemotherapy was Thirty-six patients (32%) received bevacizumab and 7 patients (6%) received 2 or more lines of preoperative chemotherapy. The disease-free interval between resection of the first site of metastasis and recurrence in another site (liver or lung) was less than 1 year in 60 patients (54%). Half of the patients (55 of 112, 49%) had multiple lung metastases, and lesions were bilateral in 43 patients (31%). Half of the patients (56

4 Vol. 213, No. 1, July 2011 Brouquet et al Resection of Colorectal Liver and Lung Metastases 65 Table 2. Intraoperative and Postoperative Characteristics of 112 Patients Who Underwent Resection of Liver and Lung Metastases from Colorectal Cancer Characteristic Data Liver resection Preoperative portal vein embolization, n (%) 6 (5) 2-stage liver resection, n (%) 8 (5) Major liver resection ( 3 contiguous liver segments), n (%) 67 (59) Radiofrequency ablation, n (%) 12 (11) Mean estimated blood loss, ml (SD) 450 (650) Transfusion requirement, n (%) 11 (9) Postoperative mortality, n (%) 0 (0) Postoperative morbidity, n (%) 29 (26) Major postoperative complication, n (%) 8 (7) Positive margin, n (%) 20 (18) Mean length of postoperative hospital stay, d (SD) 6.6 (2) Postoperative chemotherapy, n (%) 90 (80) Lung resection Combined resection of liver and lung metastases, n (%) 9 (8) Staged lung resection for bilateral lung metastases, n (%) 18 (16) Type of procedure, n (%) Thoracotomy 81 (72) Median sternotomy (for bilateral metastases) 10 (9) Thoracoscopy 16 (14) Transdiaphragmatic 5 (4) Mean estimated blood loss, ml (SD) 120 (120) Transfusion requirement, n (%) 0 (0) Postoperative mortality, n (%) 0 (0) Postoperative morbidity, n (%) 4 (4) Major postoperative complication, n (%) 1 (1) Mean length of postoperative hospital stay, d (SD) 4 (2) Positive margin, n (%) 15 (13) Postoperative chemotherapy, n (%) 51 (46) of 112, 50%) received preoperative chemotherapy before resection of lung metastases; chemotherapy regimens were 5-FU and LV in 9 patients, 5-FU and LV irinotecan in 30 patients, 5-FU and LV oxaliplatin in 17 patients. Twenty-one patients (19%) received bevacizumab and 3 patients received 2 or more lines of chemotherapy before resection of lung colorectal metastases. The mean number of chemotherapy cycles before resection of lung metastases was Perioperative findings Intraoperative and postoperative characteristics of resection of lung and liver metastases are summarized in Table 2. The majority of patients (108 of 112, 96%) had resection of lung metastases after or at the same time as liver metastases. The median time between resection of lung and liver metastases was 19 months (range 0 to 79 months). Eight patients (7%) developed severe postoperative complications after liver resection including intra-abdominal collection or bilioma requiring drainage (n 6), severe hepatic insufficiency (n 1), and acute stroke (n 1). Among the 34 patients who had bilateral lung metastases, 18 (53% of patients with bilateral metastases, 16% of total population of patients with liver plus lung metastases) had sequential resection in both lungs at different times. One patient (0.9%) developed a severe complication after resection of lung metastases (necrotizing pancreatitis). There were neither unexpected complications related to lung surgery nor toxicities in the lung that could be attributed to chemotherapy. The complication rate after resection of lung metastases was the same with or without preoperative chemotherapy (2 of 55 vs 2 of 57, respectively, not significant). With regard to the liver intervention, in the group that underwent resection of liver plus lung metastases, 12 (11%) patients underwent resection plus radiofrequency ablation of CLM vs 221 (19%) patients in the resection of CLM alone group (p 0.026). Survival after resection of liver plus lung metastases vs resection of CLM alone After a median follow-up of 49 months (range 0.1 to 159 months), patients who underwent resection of both liver and lung colorectal metastases (n 112) experienced longer survival than patients who underwent resection of CLM alone (n 1,148) (median 3-year and 5-year overall survival rates at 58 months were 71% and 50%, respectively, after resection of both liver and lung metastases vs at 44 months were 58% and 40%, respectively, after resection of liver metastases alone, p 0.01, Fig. 1). Seventy-eight patients (69%) developed recurrence after resection of both liver and lung metastases. Three-year and 5-year disease-free survival rates after resection of both liver and lung metastases were 28% and 25%, respectively. Among patients who developed recurrences after resection of liver plus lung metastases, recurrence occurred most often and included the lungs (56 of 78, 72%) followed by recurrence including the liver (41 of 78, 53%). Specifically, recurrence patterns included lung only (n 26), liver plus lung (n 21), liver only (n 17), lung plus peritoneum (n 5), lung plus brain or bone (n 4), liver plus peritoneum (n 3), nodes only (n 1), or primary only (n 2). Among the 78 patients who developed recurrence, 33 (42%) underwent potentially curative retreatment including reresection of lung recurrence (12 patients), reresection of liver recurrence (10 patients), radiofrequency ablation of liver recurrence (7 patients), and repeat treatment of liver and lung recurrences (4 patients).

5 66 Brouquet et al Resection of Colorectal Liver and Lung Metastases J Am Coll Surg Figure 1. Overall survival of 112 patients who underwent resection of liver and lung metastases from colorectal cancer. Factors associated with overall survival following resection of liver plus lung metastases Univariate analysis showed that the location of the primary tumor was associated with survival after resection of lung and liver colorectal metastases (Table 3), specifically, 3-year and 5-year survival rates were 77% and 58%, respectively, in patients who had a colon primary and 63% and 28%, respectively, in patients who had a rectal primary (p 0.01). In multivariate analysis, rectal primary tumor (hazard ratio 2.9, 95% CI 1.4 to 6, p 0.004) and carcinoembryonic antigen plasma level 5 ng/ml (hazard ratio 2.1, 95% CI 1.1 to 4.4, p 0.04) were independently associated with shorter survival after resection of lung and liver colorectal metastases (Table 3 and Fig. 2). Simultaneous presentation of liver and lung metastases occurred in a small minority of patients, such that subset analysis of survival in this group would be inappropriate. However, presentation of lung metastases within 6 months of liver resection was not prognostic. The 5-year overall survival rate was 21% in 12 patients undergoing resection of liver and lung metastases before 2000 and was 55% in 100 patients undergoing resection after 2000 (p 0.04). The proportion of patients who received preoperative chemotherapy before resection of liver metastases was significantly higher after 2000 than before 2000 (73 of 100 vs 4 of 12, p 0.01). There was no significant difference in number or size of liver metastases, synchronous presentation, use of postoperative chemotherapy after resection of CLM, number or size of lung metastases, and preoperative or postoperative chemotherapy with regard to resection of lung metastases in patients treated before vs after DISCUSSION This study is the first to compare oncologic outcomes after resection of liver plus lung colorectal metastases with those after resection of CLM alone in a large series of patients. Analysis shows that survival after resection of liver plus lung colorectal metastases can be better than survival after resection of CLM alone. The presence of resectable lung metastases should not be considered a poor prognostic factor or a contraindication to resection of CLM. These findings do not suggest that it is favorable for a patient to develop lung metastases vs not develop additional metastases, but demonstrates that patients who develop lung metastases and undergo resection do better than the general population of patients who undergo resection of CLM. In this study, we report a 50% 5-year overall survival rate after resection of liver plus lung colorectal metastases from the date of resection of last metastatic site. This rate compares favorably with previous reports of survival, which range from 11% to 40% 5 years from the date of last resection Independent evaluation of the lung disease by thoracic surgeons and the liver disease by hepatobiliary surgeons may have contributed to the good outcomes reported here. As is the case in many hepatobiliary units, restrictions on resectability, such as the number and size of liver metastases, 4,8,18 per se are not considered exclusion criteria for surgery, rather, CLM are considered resectable based on the potential to remove all tumor deposits from the liver with negative margins, leaving an adequate liver remnant. 15 When both liver and lung metastases are identified, the best sequence (usually liver first or liver plus lung) and operative approach (abdominal with transdiaphragmatic approach or abdominal plus thoracic approaches) are tailored to the disease and to the patient. Although the presence of extrahepatic metastases is reported to be associated with poor prognosis in patients undergoing resection of CLM, 1-3 we found that the survival rate in patients undergoing resection of liver plus lung colorectal metastases was better than that in the general population of patients undergoing resection of CLM alone. The majority of our patients received preoperative chemotherapy before resection of CLM and patients were selected based on response to chemotherapy (decrease in tumor size on cross-sectional imaging), a change in morphologic radiologic appearance, 7 and resectability with negative margins. Patients with multiple liver (and lung) metastases are not considered for surgery in our unit when tumors are increasing in size on chemotherapy. Good outcomes of liver plus lung resection may be explained by improved surgical technique and better patient selection, especially if the use of preoperative chemotherapy improved selection of patients for lung resection, as

6 Vol. 213, No. 1, July 2011 Brouquet et al Resection of Colorectal Liver and Lung Metastases 67 Table 3. Univariate and Multivariate Analysis of Factors Associated with Survival after Resection of Lung and Liver Colorectal Metastases Univariate analysis Multivariate analysis Factors p Value p Value Hazard ratio (95% CI) Age 60 y 0.22 Sex 0.86 Rectal primary (1.4 6) Locoregional nodes 0.64 Synchronous liver metastases 0.4 Preoperative chemotherapy before resection of liver metastases 0.46 CEA serum level before resection of liver metastases 5 ng/ml ( ) Multiple liver metastases 0.28 Liver metastasis size 5cm Positive margins at resection of liver metastases Postoperative chemotherapy after resection of liver metastases 0.32 Disease-free interval 12 months between resection of first and last metastatic site 0.32 Synchronous lung metastases 0.25 Preoperative chemotherapy before resection of lung metastases 0.63 Bilateral lung metastases 0.59 Lung metastasis size 2cm 0.34 Positive margins at resection of lung metastases 0.94 Postoperative chemotherapy after resection of lung metastases 0.27 suggested by better outcomes after 2000 despite similar patient characteristics in the early and late periods. Comparatively better outcomes for the subset with resected lung metastases may also be explained by the fact that more than half of all patients with resected CLM develop recurrence, and only a minority can undergo complete treatment of that recurrence (eg, those with resectable lung metastases). From a clinical perspective, complete response of lung metastases can be a concern related to use of preoperative chemotherapy before pulmonary metastasectomy; the majority of patients had a lung metastasis size of 1 cm or less, making intraoperative localization of lesions more difficult in the event of major response to preoperative therapy. For this reason, preoperative chemotherapy before resection of lung metastases was not routinely used (57 of 112, 51%); rather, chemotherapy before lung surgery was often related to treatment in patients with synchronous diagnosis of liver and lung metastases or concern that the disease in the lung might represent more widespread disease. In such cases, the multidisciplinary team was more comfortable proceeding to lung resection after a period without treatment during which no other disease became evident. It is increasingly clear that assessment of response to chemotherapy helps to select which patients with advanced disease will benefit most from aggressive surgery. 9,19 Further, the finding of a high survival rate in patients with relatively advanced colorectal metastases is expected, and has been reported, for example, in patients undergoing repeat hepatectomy for recurrent CLM. 20,21 In the past decade, resectability criteria of patients with CLM have expanded to include patients with advanced disease with multiple metastatic sites. 3,22 Our results reflect the positive effect of an aggressive surgical approach to patients with liver and lung colorectal metastases. In this study, all patients had resection with curative intent of metastatic disease regardless of the initial involved site, and a significant proportion of patients were candidates for and subsequently underwent repeat surgical treatment of further recurrence after resection of liver and lung colorectal metastases. Approximately 10% to 20% of patients with colorectal cancer develop lung metastases at some point during the course of their disease. 23,24 Lung metastases are typically small volume lesions and are usually not the cause of death in patients with advanced colorectal cancer. Benign pulmonary nodules can be detected with high quality imaging in patients with colorectal cancer as well. A recent report suggested that the presence of subcentimeter pulmonary nodules did not affect disease-specific survival in patients with CLM; importantly, only a minority of the patients in that series had confirmed metastases (35%) and less than 8% of patients underwent resection. 25 For these reasons, patients with pathologically proven (resected) lung metastases should not be compared with patients with observed pul-

7 68 Brouquet et al Resection of Colorectal Liver and Lung Metastases J Am Coll Surg colorectal cancer 1,5,8,27,28 and probably reflects the total tumor burden. Of note, rectal primary tumor was the most reliable predictor of outcome in our patients after resection of lung and liver colorectal metastases. Rectal primary has been associated with worse outcomes in patients undergoing resection of CLM 27 and an increase in the risk of lung relapse after resection of nonmetastatic colorectal cancer. 29 The influence of the primary tumor site has been purported to relate to differences in the aggressiveness of the tumor and also to the risk of pelvic recurrence associated with rectal cancer. The occurrence of systemic recurrences might also be related by the access of circulating tumor cells to the systemic circulation via nonmesenteric veins. In contrast, we did not observe any relationship between outcomes and the number and size of lung or liver metastases or the disease-free interval between the diagnosis of lung and liver metastases. 13 No finding of a prognostic factor in this study can be considered a specific contraindication to resection in an otherwise suitable patient. This retrospective review of a prospective cohort was not designed to be an analysis of the incidence of extrahepatic metastasis in patients with CLM, nor was analysis of the influence of extrahepatic metastasis on survival the aim of the study. Figure 2. Factors independently associated with survival after resection of liver and lung colorectal metastases in multivariate analysis. (A) Overall survival stratified by primary tumor location (colon vs rectum); and (B) overall survival stratified by preoperative carcinoembroyonic antigen (CEA) level, ( 5 ng/ml vs 5 ng/ml). monary nodules; such a comparison would be misleading because a significant proportion of patients with pulmonary nodules do not have pulmonary metastases. Further, recent improvements in treatment, including more effective modern chemotherapy, usually prevent rapid progression of lung colorectal metastases. Long-term cancer control can be expected in patients with stage IV cancer, especially those with small volume lung disease. This study suggests that complete resection of lung disease may contribute to significantly prolonged survival in these patients. The finding that CEA serum level is associated with survival in patients undergoing resection of liver and lung colorectal metastases is confirmed in this study. 26 The CEA serum level has been consistently reported as a strong prognostic factor in patients undergoing resection of metastatic CONCLUSIONS This study shows that survival outcomes of patients undergoing resection of lung and liver colorectal metastases are excellent and may be better than survival in the general population of patients undergoing resection of CLM alone. These results reiterate the value of an aggressive oncosurgical approach to improve outcomes of patients with advanced colorectal cancer. Our study also confirms that in the era of modern treatment strategies for metastatic colorectal cancer, the presence of resectable lung metastases should not be considered a poor prognostic factor or contraindication for resection of CLM. Author Contributions Study conception and design: Brouquet; Abdalla, Vauthey, Contreras Acquisition of data: Brouquet, Vauthey, Contreras, Walsh, Vaporciyan, Swisher, Curley, Mehran, Abdalla Analysis and interpretation of data: Brouquet, Vauthey, Contreras, Walsh, Vaporciyan, Swisher, Curley, Mehran, Abdalla Drafting of manuscript: Brouquet, Abdalla, Vauthey Critical revision: Brouquet, Vauthey, Contreras, Walsh, Vaporciyan, Swisher, Curley, Mehran, Abdalla Acknowledgment: The authors particularly thank Ruth J Haynes and Stephanie Deming for editing.

8 Vol. 213, No. 1, July 2011 Brouquet et al Discussion 69 REFERENCES 1. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999;230: Rees M, Tekkis PP, Welsh FK, et al. Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: a multifactorial model of 929 patients. Ann Surg 2008;247: Carpizo DR, Are C, Jarnagin W, et al. Liver resection for metastatic colorectal cancer in patients with concurrent extrahepatic disease: results in 127 patients treated at a single center. Ann Surg Oncol 2009;16: Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004;239: Watanabe K, Nagai K, Kobayashi A, et al. Factors influencing survival after complete resection of pulmonary metastases from colorectal cancer. Br J Surg 2009;96: Fernandez FG, Drebin JA, Linehan DC, et al. Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 2004;240: Chun YS, Vauthey JN, Boonsirikamchai P, et al. Association of computed tomography morphologic criteria with pathologic response and survival in patients treated with bevacizumab for colorectal liver metastases. JAMA 2009;302: Blazer DG 3rd, Kishi Y, Maru DM, et al. Pathologic response to preoperative chemotherapy: a new outcome end point after resection of hepatic colorectal metastases. J Clin Oncol 2008;26: Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008;371: Regnard JF, Grunenwald D, Spaggiari L, et al. Surgical treatment of hepatic and pulmonary metastases from colorectal cancers. Ann Thorac Surg 1998;66: Kobayashi K, Kawamura M, Ishihara T. Surgical treatment for both pulmonary and hepatic metastases from colorectal cancer. J Thorac Cardiovasc Surg 1999;118: Shah SA, Haddad R, Al-Sukhni W, et al. Surgical resection of hepatic and pulmonary metastases from colorectal carcinoma. J Am Coll Surg 2006;202: Miller G, Biernacki P, Kemeny NE, et al. Outcomes after resection of synchronous or metachronous hepatic and pulmonary colorectal metastases. J Am Coll Surg 2007;205: Kopetz S, Vauthey JN. Perioperative chemotherapy for resectable hepatic metastases. Lancet 2008;371: Abdalla EK, Adam R, Bilchik AJ, et al. Improving resectability of hepatic colorectal metastases: expert consensus statement. Ann Surg Oncol 2006;13: Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240: Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92: Elias D, Baton O, Sideris L, et al. Local recurrences after intraoperative radiofrequency ablation of liver metastases: a comparative study with anatomic and wedge resections. Ann Surg Oncol 2004;11: Chun YS, Vauthey JN, Ribero D, et al. Systemic chemotherapy and two-stage hepatectomy for extensive bilateral colorectal liver metastases: perioperative safety and survival. J Gastrointest Surg 2007;11: Nordlinger B, Vaillant JC, Guiguet M, et al. Survival benefit of repeat liver resections for recurrent colorectal metastases: 143 cases. Association Francaise de Chirurgie. J Clin Oncol 1994; 12: Pinson CW, Wright JK, Chapman WC, et al. Repeat hepatic surgery for colorectal cancer metastasis to the liver. Ann Surg 1996;223: Elias D, Baton O, Sideris L, et al. Hepatectomy plus intraoperative radiofrequency ablation and chemotherapy to treat technically unresectable multiple colorectal liver metastases. J Surg Oncol 2005;90: McCormack PM, Burt ME, Bains MS, et al. Lung resection for colorectal metastases. 10-year results. Arch Surg 1992;127: August DA, Ottow RT, Sugarbaker PH. Clinical perspective of human colorectal cancer metastasis. Cancer Metastasis Rev 1984;3: Maithel SK, Ginsberg MS, D Amico F, et al. Natural history of patients with subcentimeter pulmonary nodules undergoing hepatic resection for metastatic colorectal cancer. J Am Coll Surg 2010;210: Headrick JR, Miller DL, Nagorney DM, et al. Surgical treatment of hepatic and pulmonary metastases from colon cancer. Ann Thorac Surg 2001;71: ; discussion Nordlinger B, Guiguet M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection, based on 1568 patients. Association Francaise de Chirurgie. Cancer 1996;77: Brouquet A, Mortenson MM, Vauthey JN, et al. Surgical strategies for synchronous colorectal liver metastases in 156 consecutive patients: classic, combined or reverse strategy? J Am Coll Surg 2010;210: Bohm B, Schwenk W, Hucke HP, et al. Does methodic longterm follow-up affect survival after curative resection of colorectal carcinoma? Dis Colon Rectum 1993;36: Discussion INVITED DISCUSSANT: DR JOHN BREMS (Maywood, IL): This is a retrospective review of 1,260 consecutive patients, over a 14-year period, who underwent liver only or liver plus lung resections for metastases from colorectal cancer. The mean tumor size resected was 3 cm for the liver metastases and 1 cm for the lung metastases. The great majority of these patients developed their lung metastases after the liver metastases. The authors conclude that survival may be better for patients with liver and lung metastases as opposed to liver metastases only.