Abstract

Background

Although resecting colorectal cancer (CRC) pulmonary metastasis is associated with long-term survival, identification of prognostic groups is needed for future randomized trials, and construction of a lung metastasectomy prognostic model (LMPM) is warranted.

Patients and methods

We searched the PubMed database for retrospective studies evaluating prognostic factors following resecting CRC lung metastasis. Individual patient data were analyzed. Independent prognostic factors were used to construct an LMPM.

Results

Between 1983 and 2008, 1112 metastasectomies were carried out on 927 patients included in eight studies. Five-year survival rate was 54.3% following the first lung resection. Multivariate analysis identified three independently poor prognostic factors: pre-thoracotomy carcinoembryonic antigen ≥5 ng/ml, disease-free interval <36 months, and more than one metastatic lesion. Patients with good-, intermediate-, and high-risk groups according to the LMPM had a 5-year survival of 68.2%, 46.4%, and 26.1%, respectively (P < 0.001). Perioperative chemotherapy and previously resected liver metastasis had no influence on survival.

Conclusions

The low- and intermediate-risk groups have a good chance of long-term survival following metastasectomy. However, more studies are needed to investigate whether surgery offers any advantage over systemic therapy for the poor-risk group.

introduction

Around 10%–15% of patients undergoing surgical resection for colorectal cancer (CRC) will develop lung metastasis [1–3].

For the minority of patients with CRC where lungs are the only site of metastatic disease, although there is absence of randomized studies, many retrospective studies suggested improved survival with metastasectomy and curative treatment of the primary CRC. In several series, encouraging 5-year survival rates of 40%–61% had been reported [4–7].

Previously, this practice of pulmonary metastasectomy was applied to few patients with solitary lung metastasis and long disease-free interval (DFI); however, many retrospective studies suggested that patients with multiple lung metastases might benefit as well [4, 8].

Most retrospective studies evaluating lung metastasectomy in the setting of metastatic CRC had evaluated prognostic factors for survival [9–13]; however, conflicting data regarding some prognostic factors had been reported.

Some studies had shown better outcome for patients with longer DFI [10–14], solitary rather than multiple metastatic lesions [11, 15], smaller size of the largest metastatic pulmonary lesion [15, 16], and normal rather than elevated pre-thoracotomy carcinoembryonic antigen (CEA) level [10, 12, 16, 17]. On the other hand, some studies failed to show any influence on survival based on DFI [8, 18], presence of single versus multiple lesions [18], or number of metastatic lesions [8].

The international registry of lung metastases had accrued 5206 cases of lung metastasectomy, which is the largest reported series [19]. A simple prognostic system based on resectability, DFI, and number of metastases was proposed; however, it was not specific as it included variety of cancers including sarcomas, germ cell tumors, melanomas, and carcinomas.

Constructing a prognostic model specific for CRC pulmonary oligometastases would be more informative in assessing the prognosis of this patient population. Furthermore, such a model may be needed for stratification of patients enrolled in future randomized trials, as well as for comparing different cohorts of patients from different institutions.

methods

search strategy

we searched the PubMed database to identify retrospective studies that included patients with pulmonary oligometastases from CRC who had undergone metastasectomy, using the words lung metastasis, colon cancer, surgery, or metastasectomy.

Articles that were published before the year 2000 and/or included <20 patients were excluded.

The authors of the articles were contacted to provide individual patient data for each patient included in their studies.

definitions and patient's data

We gathered individual patient's data about 13 possible prognostic factors for each of the patients as long as they are available: age, gender, status of regional lymph node (LN) involvement for the primary CRC, histological grade of the metastatic cancer, DFI, number of lung metastatic nodules at first detection of pulmonary metastasis, maximum diameter of the largest lung metastatic lesion, type of surgical resection, laterality of metastasis at first detection of lung metastasis, pre-thoracotomy CEA level, intrathoracic LN involvement, whether the patient had received perioperative chemotherapy or not, and whether the patient had a history of previous liver metastases.

DFI was defined as the interval from definitive treatment of the primary CRC until detection of pulmonary metastases. For patients with history of resection of oligometastases involving another organ, the DFI was defined as the time from resection of the previous metastasis until the first detection of the pulmonary oligometastases. These patients had to be free of other organ metastases at the time of pulmonary metastasectomy.

Metachronous presentation was defined as detection of pulmonary oligometastases later than 6 months following the definitive treatment of the primary CRC or a curative resection of liver metastases.

Synchronous presentation was defined as detection of the pulmonary oligometastases within 6 months of definitive treatment of the primary CRC or the curative liver resection, at the same time of initial evaluation of the primary cancer or before detection of the primary cancer.

patient's inclusion and exclusion criteria

Included patients are those who were aged ≥18 years with oligometastases confined to one or both lungs and who had undergone surgical resection of the metastasis.

Patients with any of the following were excluded: extrapulmonary metastasis at the time of first detection of lung metastasis, gross residual disease following resection of the metastases (R2 resection) or positive resection margins (R1 resection), the primary CRC was not treated definitively, absence of follow-up data, and if other metastatic sites could not be ruled out.

statistical analysis

Survival times were calculated from the time of the first pulmonary metastasectomy until death or last follow-up.

Survival curves were plotted using the Kaplan–Meier method.

The 13 possible prognostic factors were evaluated in univariate analysis using the log-rank test. A P value of <0.05 was considered statistically significant.

Factors found to be significant in the univariate analysis were evaluated in the multivariate analysis using the backward stepwise regression model.

The impact of each independent prognostic factor on survival was interpreted for the entire cohort of patients, as well as per study, using hazard ratios with corresponding 95% confidence intervals (CI).

An assessment of heterogeneity among patients included from the different studies was carried out using a fixed effect model, which tests variation using a χ2 test.

All analyses were carried out with SPSS version 13.0 J software (SPSS, Tokyo, Japan).

results

search results and patients

We identified 83 retrospective studies evaluating outcomes following resection of CRC pulmonary oligometastases. Of the 83 studies, 43 were excluded for the following reasons: 18 studies for being published before the year 2000, 14 studies because each included <20 patients, and 10 studies were excluded for both reasons.

Of the remaining 41 articles, we were able to collect individual patient data from eight retrospective studies (Table 1).

Table 1.

Retrospective studies that were included in the pooled analysis

First author Year of publication Number of included patients Study area Study period 
Watanabe [172009 101 Japan 1992–2004 
Zabaleta [202011 84 Spain 1998–2008 
Nojiri [212011 29 Japan 1992–2006 
Riquet [222010 102 France 1983–2007 
Borasio [112011 131 Italy 1989–2008 
Park [232010 187 Korea 1995–2007 
Hwang [242010 125 Korea 2001–2007 
Welter [252007 168 Germany 1993–2003 
First author Year of publication Number of included patients Study area Study period 
Watanabe [172009 101 Japan 1992–2004 
Zabaleta [202011 84 Spain 1998–2008 
Nojiri [212011 29 Japan 1992–2006 
Riquet [222010 102 France 1983–2007 
Borasio [112011 131 Italy 1989–2008 
Park [232010 187 Korea 1995–2007 
Hwang [242010 125 Korea 2001–2007 
Welter [252007 168 Germany 1993–2003 

These eight articles included 988 patients. Of those, 61 were excluded for the following reasons: 24 for synchronous lung and liver metastases, 36 for R1 or R2 resection, and 1 for uncertain survival data. After excluding the 61 patients, 927 patients were eligible for analysis.

patients’ characteristics

The characteristic of patients and their metastatic CRC are summarized (Table 2).

Table 2.

Patient's characteristics at the time of first metastasectomy of colorectal cancer (CRC) lung metastases

Clinical variable No. of patients (%) 
Gender  
 Male 568 (29) 
 Female 359 (71) 
LN in the primary CRCa  
 Involved 478 (52) 
 Not involved 309 (33) 
 Missing data 140 (15) 
Pre-thoracotomy CEA (ng/ml)  
 ≥5 240 (26) 
 <5 474 (51) 
 Missing data 213 (23) 
History of liver metastasis  
 Yes 236 (25) 
 No 691 (75) 
Status of lymph nodes in thorax  
 Involved 75 (8) 
 Not involved 424 (46) 
 Sampling or dissection not carried out 297 (32) 
 Missing data 131 (14) 
Type of surgical resection  
 Sublobectomy 555 (60) 
 Lobectomy 186 (20) 
 Pneumonectomy 12 (1) 
 Missing data 174 (19) 
Laterality of metastasisb  
 Unilateral 635 (68) 
 Bilateral 91 (10) 
 Missing data 201 (22) 
Maximum diameter of largest nodule (cm)  
 <2 386 (42) 
 ≥2 523 (56) 
 Missing data 18 (2) 
No. of metastatic nodules:  
 1 558 (60) 
 ≥2 364 (39) 
 Missing data 5 (1) 
DFI (months)  
 ≥36 310 (33) 
 <36 603 (65) 
 Missing data 14 (2) 
Histological grade  
 Well/moderately differentiated 526 (57) 
 Poorly differentiated 26 (3) 
 Missing data 585 (40) 
Perioperative chemotherapy  
 Given 171 (18) 
 Not given 273 (29) 
 Missing data 483 (52) 
Clinical variable No. of patients (%) 
Gender  
 Male 568 (29) 
 Female 359 (71) 
LN in the primary CRCa  
 Involved 478 (52) 
 Not involved 309 (33) 
 Missing data 140 (15) 
Pre-thoracotomy CEA (ng/ml)  
 ≥5 240 (26) 
 <5 474 (51) 
 Missing data 213 (23) 
History of liver metastasis  
 Yes 236 (25) 
 No 691 (75) 
Status of lymph nodes in thorax  
 Involved 75 (8) 
 Not involved 424 (46) 
 Sampling or dissection not carried out 297 (32) 
 Missing data 131 (14) 
Type of surgical resection  
 Sublobectomy 555 (60) 
 Lobectomy 186 (20) 
 Pneumonectomy 12 (1) 
 Missing data 174 (19) 
Laterality of metastasisb  
 Unilateral 635 (68) 
 Bilateral 91 (10) 
 Missing data 201 (22) 
Maximum diameter of largest nodule (cm)  
 <2 386 (42) 
 ≥2 523 (56) 
 Missing data 18 (2) 
No. of metastatic nodules:  
 1 558 (60) 
 ≥2 364 (39) 
 Missing data 5 (1) 
DFI (months)  
 ≥36 310 (33) 
 <36 603 (65) 
 Missing data 14 (2) 
Histological grade  
 Well/moderately differentiated 526 (57) 
 Poorly differentiated 26 (3) 
 Missing data 585 (40) 
Perioperative chemotherapy  
 Given 171 (18) 
 Not given 273 (29) 
 Missing data 483 (52) 

aStatus of regional LNs involvement at the time of CRC surgery.

bLaterality of the metastases at the time of first resection of pulmonary metastases: unilateral or bilateral.

CEA, carcinoembryonic antigen; CRC, colorectal cancer; DFI, disease-free interval; LN, lymph nodes.

Median age at first metastasectomy was 62.6 years.

The median DFI was 25.6 months (range, 0–181.6 months). Eight hundred and ten patients (87%) had metachronous metastasis, 104 patients (11%) had synchronous metastasis, while data were not available for 13 patients (2%). Previous liver resection was carried out on 236 patients (25%).

Mean follow-up after the first pulmonary metastasectomy was 43.3 months, standard deviation = 32.5 months.

Data on the administration of perioperative chemotherapy was available for only 444 patients (48%); 171 patients received chemotherapy and 273 did not.

The assessment of heterogeneity among patients included from the different studies using a fixed effect model did not demonstrate significant differences regarding age, number of patients with more than one metastatic nodule, number of patients with synchronous versus metachronous metastasis, type of surgical resection, pre-thoracotomy CEA level, DFI, and number of patients with maximum size of the largest lung metastatic lesion of ≥2 cm, χ2 < 0.0001.

surgical outcomes

Between 1983 and 2008, 1112 thoracotomies were carried out on the 927 patients. One hundred and forty-eight patients (16%) had repeated metastasectomy. Of those, 32 (22%) had a third thoracotomy and 5 (3.4%) had a fourth thoracotomy.

The mode of resection at first metastasectomy was a sublobar resection (wedge or segmentectomy or both) in 555 patients (59.9%), a lobectomy with or without wedge or segmentectomy in 186 patients (20.1%) and a pneumonectomy in 12 patients (1.3%), while data were not available for the rest of patients.

Data about LN sampling or dissection during the first thoracotomy were available for 796 (86%) patients. Of those, LN sampling or dissection was not carried out in 279 patients (35%), showed positive lymph nodes in 75 patients (9%), and no involvement in 424 patients (53%).

The type of surgical resection during the 185 repeated thoracotomies were as follows: a sublobar resection in 105 (57%), a lobectomy in 52 (28%), and a pneumonectomy in 2 (1%), while data were not available for 26 cases (14%).

postoperative morbidity and mortality

Data on postoperative morbidity was available for 808 thoracotomies carried out on 672 patients. Of those, 77 patients (11%) had any major postoperative morbidity.

The major postoperative complications were air leak for >7 days in 24 patients (3.6%), atrial fibrillation in 17 patients (2.5%), pneumonia in 14 patients (2.1%), empyema in 7 patients (1%), major hemorrhage in 7 patients (1%), intestinal obstruction in 3 patients (0.4%), and respiratory failure not attributable to any of the above reasons for additional 5 patients (0.7%).

Of all 927 patients, 5 (0.5%) died within 2 months of first thoracotomy, and their death was considered surgery related.

survival outcome

Five-year overall survival rate calculated from the time of first thoracotomy for the entire cohort of patients was 54.3%, (Figure 1).

Figure 1.

Overall Kaplan–Meier survival estimation after the first metastasectomy.

Figure 1.

Overall Kaplan–Meier survival estimation after the first metastasectomy.

Univariate analysis demonstrated five statistically significant prognostic factors: DFI, ≥36 versus <36 months (5-year survival rate of 61.3% and 50.1%, respectively, P < 0.0398); pre-thoracotomy, CEA <5 versus ≥5 (5-year survival rate of 63.0% and 45.1%, respectively, P < 0.0001); number of metastatic lesions, 1 versus ≥2 (5-year survival rate of 60.5% and 44.6%, respectively, P < 0.0001); maximum diameter of largest metastatic lesion, <2 versus ≥ 2 cm (5-year survival rate of 57.7% and 51.8%, respectively, P = 0.0015); and status of lymph nodes in the thorax, not involved versus involved (5-year survival rate of 55.4% and 27.0%, respectively, P < 0.0001).

The multivariate analysis identified three independent clinical poor prognostic factors: pre-thoracotomy CEA (<5 versus ≥5), number of metastatic lesions (1 versus ≥2), and the DFI (≥36 versus <36 months) (Table 3).

Table 3.

Independent clinical prognostic factors

Variable No. of patients (%) Hazard ratio 95% confidence interval P value 
Pre-thoracotomy CEA (ng/ml)     
 ≥5 240 (26) 1.84 1.425–2.381 <0.001 
 <5 474 (51)   
DFI (months)     
 <36 603 (65) 1.39 1.030–1.862 0.031 
 ≥36 310 (33)   
Number of metastatic nodules     
 ≥2 364 (39) 2.05 1.583–2.648 <0.001 
 1 558 (60)   
Variable No. of patients (%) Hazard ratio 95% confidence interval P value 
Pre-thoracotomy CEA (ng/ml)     
 ≥5 240 (26) 1.84 1.425–2.381 <0.001 
 <5 474 (51)   
DFI (months)     
 <36 603 (65) 1.39 1.030–1.862 0.031 
 ≥36 310 (33)   
Number of metastatic nodules     
 ≥2 364 (39) 2.05 1.583–2.648 <0.001 
 1 558 (60)   

The three independent poor prognostic factors were used to construct a lung metastasectomy prognostic model. Risk categories according to number of poor prognostic factors: low risk = 0 or 1 factor, intermediate risk = 2 factors, high risk= 3 factors.

CEA, carcinoembryonic antigen; DFI, disease-free interval.

The hazard ratios with the corresponding 95% CIs of each independent prognostic factor for all patients and per study are illustrated in forest plot (figure 2).

Figure 2.

Forest plot illustration of the hazard ratios with corresponding 95% confidence intervals for each of the independent poor prognostic factors per study: (A) pre-thoracotomy CEA ≥5 ng/ml, (B) DFI < 36 months, (C) number of metastatic lung nodules ≥2. The diamond at the bottom of each forest plot corresponds to the hazard ratio for all pooled patients.

Figure 2.

Forest plot illustration of the hazard ratios with corresponding 95% confidence intervals for each of the independent poor prognostic factors per study: (A) pre-thoracotomy CEA ≥5 ng/ml, (B) DFI < 36 months, (C) number of metastatic lung nodules ≥2. The diamond at the bottom of each forest plot corresponds to the hazard ratio for all pooled patients.

Perioperative chemotherapy had no significant influence on survival, 5-year survival of 54.9% with chemotherapy versus 56.8% without chemotherapy (P = 0.7).

A clinical prognostic model was constructed using the three independent clinical prognostic factors.

Of the 712 patients assessable for all three factors, the 5-year survival for the 68 patients (10%) with three poor prognostic factors (high-risk group), the 250 patients (35%) with two poor prognostic factors (intermediate-risk group), and the 394 patients (55%) with zero or one risk factor (low-risk group), were 26.1%, 46.4%, and 68.2%, respectively, P < 0.001 (overall and for comparison between any two groups). The corresponding 3-year survival rates were 35.6%, 69.8% and 82.5%, respectively (P < 0.001) (Figure 3).

Figure 3.

Survival estimation according to risk group (712 assessable patients).

Figure 3.

Survival estimation according to risk group (712 assessable patients).

discussion

Around 50%–60% of patients diagnosed with CRC will develop metastatic disease [26, 27]. Metastases are more likely to develop metachronously and most likely involve the liver, which is the most common metastatic site [28].

Lung metastases complicate the course of around 10%–15% of those who had undergone curative CRC surgery [1–3, 29, 30]. Nevertheless, it is uncommon to have patients where the lungs are the only metastatic site. Tan et al. [31] reported a 7.4% rate of isolated lung metastasis in a cohort of 754 patients with CRC.

In this setting, similar to the results after surgical resection of liver oligometastases, retrospective studies suggested that resecting pulmonary oligometastases offers a better chance of survival when compared with the expected 5-year survival rates of <10% for stage IV CRC managed with palliative chemotherapy [32, 33].

How to select those patients who are most likely to benefit from surgical resection of pulmonary oligometastases is the important question that needs to be answered. Selection criteria had been the main focus of numerous studies evaluating factors that influence survival after metastasectomy.

To provide a better understanding about the prognosis of these patients, we elected to construct a prognostic model which takes all the probable prognostic factors into consideration and can be applied in daily practice for every patient considered for resection.

Our analysis demonstrates significant and clinically meaningful differences in survival between the three risk groups. Five-year survival rate was 26.1% for the highest risk group versus 68.2% for the most favorable group.

Although patients who belong to the low- and intermediate-risk groups could have up to two poor prognostic factors, we found the prognosis of these two groups very favorable when compared with the known poor survival of stage IV CRC treated with systemic chemotherapy. However, the survival benefit from surgery in the high-risk group is unclear and needs further evaluation.

We found that a history of resected liver metastasis has no detrimental influence on survival. As a consequence, pulmonary metastasectomy should not be denied for patients with history of previous liver metastasectomy with otherwise favorable prognostic factors.

An interesting finding was the low postoperative mortality of 0.5%. This could be explained by our exclusion criteria for articles, which included <20 patients and articles published before the year 2000. Studies including smaller number of patients are more likely to be reported from centers with low volume of such surgeries and less experience. Additionally, excluding older publications would exclude patients treated in the era of less advanced supportive and intensive care medicine.

The fact that we could not find a benefit of chemotherapy prior or after metastasectomy has to be interpreted with caution. We understand that we have several limitations that affect our conclusions. First, data about delivery of chemotherapy was missing for half of patients. Second, we could not get data about the type of chemotherapy administered for most patients since many of them had received it in other centers before or after their referral for surgery. Additionally, differences in chemotherapy regimens, doses, and schedules as well as selection bias could influence our conclusions.

The role of perioperative chemotherapy requires further evaluation. This approach was evaluated by the European Organization for Research and Treatment of Cancer Intergroup trial 40983 for patients with isolated liver metastasis, where the use of perioperative FOLFOX4 improved progression-free survival in patients undergoing liver resection [34]; however, this approach of perioperative chemotherapy was never evaluated for CRC pulmonary oligometastases.

Another limitation is that our proposed model was assessable for only 712 patients (77%); this limitation was attributable to missing data about pre-thoracotomy CEA level for 213 patients (23%). To limit the influence of these missing data on our results, we assessed the model only for the 712 patients for whom we have data about all three independent prognostic factors.

Our analysis was focused on survival and prognostic factors directly associated with the malignancy. This is an important limitation as other factors, not related to malignancy might be prognostic as well. These factors are pulmonary function and comorbidities. Furthermore, associated factors like quality of life, post-thoracotomy pain syndrome, as well as long-standing postoperative morbidity could not be addressed in this paper. These limitations were attributed to limited data. We believe that these are important factors to be considered for future studies.

Although numerous retrospective studies in pulmonary metastasectomy for metastatic CRC had shown encouraging survival results, we cannot derive a firm conclusion that the observed long-term survival in these patients is attributable to surgery rather than the fact that they are a highly selected population with good performance status, minimal comorbidities, better pulmonary function, and younger age and the fact that the natural biology of cancer may had played a role in their favorable survival, given the absence of randomized studies in this setting. In fact, both the CEA level and the number of metastatic sites were reported to have independent prognostic significance for poor survival in patients with unresectable and disseminated metastatic CRC [35, 36], which adds to the complexity of interpretation of data derived from single-arm studies evaluating surgical resection, and suggests that at least the biology of cancer in these patients may had contributed to their apparent favorable survival. Although the literature is rich in these retrospective studies and the survival results with surgery are encouraging, many authors still argue that the available level of evidence is not sufficient to draw firm conclusions about the effectiveness of surgery [37, 38].

We think, although randomized studies are lacking, that the 5-year survival approaching 70% for the low-risk group following surgery is unlikely to be achieved with other modalities. We recommend surgical resection of lung metastasis for the low- and intermediate-risk groups until strong evidence confirming a non-inferiority of systemic therapy becomes available. On the other hand, for the high-risk group, conducting a randomized study that compares metastasectomy with palliative chemotherapy and targeted therapies is highly justified, given the poor prognosis of this subset of patients following metastasectomy and the absence of a strong evidence proving the benefit of surgery.

conclusions

According to our proposed prognostic model, patients with CRC pulmonary oligometastases in the low- and intermediate-risk groups have an excellent chance of long-term survival following resection of the pulmonary oligometastases. We recommend meticulous resection of these metastases after adequate staging workup had ruled out other metastatic sites. For patients who belong to the poor-risk group, more studies are needed to evaluate whether surgery offers any advantage over systemic therapy.

disclosure

The authors have declared no conflicts of interest.

acknowledgements

We thank all of the following: Dalia Al-Rimawi, Statistical Programmer at King Hussein cancer center, for statistical help; Mervat Al-Risheq for some technical assistance in organizing the data; Salah Abbasi and Alaa' Addasi for reviewing the manuscript.

references

1
McCormack
PM
Burt
ME
Bains
MS
, et al.  . 
Lung resection for colorectal metastases. 10-year results
Arch Surg
 , 
1992
, vol. 
127
 (pg. 
1403
-
1406
)
2
August
DA
Ottow
RT
Sugerbaker
PH
Clinical perspective of human colorectal cancer metastasis
Cancer Metastasis Rev
 , 
1984
, vol. 
3
 (pg. 
303
-
324
)
3
Goya
T
Miyazawa
N
Kondo
H
, et al.  . 
Surgical resection of pulmonary metastases from colorectal cancer. 10-year follow-up
Cancer
 , 
1989
, vol. 
64
 (pg. 
1418
-
1421
)
4
Suemitsu
R
Takeo
S
Kusumoto
E
, et al.  . 
Results of a pulmonary metastasectomy in patients with colorectal cancer
Surg Today
 , 
2011
, vol. 
41
 
1
(pg. 
54
-
59
)
5
Landes
U
Robert
J
Perneger
T
, et al.  . 
Predicting survival after pulmonary metastasectomy for colorectal cancer: previous liver metastases matter
BMC Surg
 , 
2010
, vol. 
10
 pg. 
17
 
6
Inoue
M
Ohta
M
Iuchi
K
, et al.  . 
Thoracic Surgery Study Group of Osaka University
Benefits of surgery for patients with pulmonary metastases from colorectal carcinoma
Ann Thorac Surg
 , 
2004
, vol. 
78
 
1
(pg. 
238
-
244
)
7
Saito
Y
Omiya
H
Kohno
K
, et al.  . 
Pulmonary metastasectomy for 165 patients with colorectal carcinoma: a prognostic assessment
J Thorac Cardiovasc Surg
 , 
2002
, vol. 
124
 
5
(pg. 
1007
-
1013
)
8
Mongil Poce
R
Pagés Navarrete
C
Ruiz Navarrete
JA
, et al.  . 
[Survival analysis of resection of lung metastases from colorectal cancer]
Arch Bronconeumol
 , 
2009
, vol. 
45
 
5
(pg. 
235
-
239
)
9
Chen
F
Shoji
T
Sakai
H
, et al.  . 
Lung metastasectomy for colorectal carcinoma in patients with a history of hepatic metastasis
Ann Thorac Cardiovasc Surg
 , 
2011
, vol. 
17
 (pg. 
13
-
18
)
10
Rama
N
Monteiro
A
Bernardo
JE
, et al.  . 
Lung metastases from colorectal cancer: surgical resection and prognostic factors
Eur J Cardiothorac Surg
 , 
2009
, vol. 
35
 (pg. 
444
-
449
)
11
Borasio
P
Gisabella
M
Billé
A
, et al.  . 
Role of surgical resection in colorectal lung metastases: analysis of 137 patients
Int J Colorectal Dis
 , 
2011
, vol. 
26
 
2
(pg. 
183
-
190
)
12
Takakura
Y
Miyata
Y
Okajima
M
, et al.  . 
Short disease-free interval is a significant risk factor for intrapulmonary recurrence after resection of pulmonary metastases in colorectal cancer
Colorectal Dis
 , 
2010
, vol. 
12
 
7 Online
(pg. 
e68
-
e75
)
13
Onaitis
MW
Petersen
RP
Haney
JC
, et al.  . 
Prognostic factors for recurrence after pulmonary resection of colorectal cancer metastases
Ann Thorac Surg
 , 
2009
, vol. 
87
 
6
(pg. 
1684
-
1688
)
14
Lin
BR
Chang
TC
Lee
YC
, et al.  . 
Pulmonary resection for colorectal cancer metastases: duration between cancer onset and lung metastasis as an important prognostic factor
Ann Surg Oncol
 , 
2009
, vol. 
16
 
4
pg. 
1026
 
15
Nakajima
J
Murakawa
T
Fukami
T
Takamoto
S
Is thoracoscopic surgery justified to treat pulmonary metastasis from colorectal cancer?
Interact Cardiovasc Thorac Surg
 , 
2008
, vol. 
7
 
2
(pg. 
212
-
216
discussion 216–217
16
Muñoz Llarena
A
Carrera Revilla
S
Gil-Negrete Laborda
A
, et al.  . 
[Prognostic factors associated with resectable pulmonary metastases from colorectal cancer]
Arch Bronconeumol
 , 
2007
, vol. 
43
 
6
(pg. 
309
-
316
)
17
Watanabe
K
Nagai
K
Kobayashi
A
, et al.  . 
Factors influencing survival after complete resection of pulmonary metastases from colorectal cancer
Br J Surg
 , 
2009
, vol. 
96
 
9
(pg. 
1058
-
1065
)
18
Zapf
I
Molnár
FT
Benko
I
, et al.  . 
[Surgical treatment of pulmonary metastases of colorectal cancers]
Magy Seb
 , 
2007
, vol. 
60
 
3
(pg. 
130
-
135
)
19
The International Registry of Lung Metastases
Long-term results of lung metastasectomy: prognostic analyses based on 5206 cases
J Thorac Cardiovasc Surg
 , 
1997
, vol. 
113
 
1
(pg. 
37
-
49
)
20
Zabaleta
J
Aguinagalde
B
Fuentes
MG
, et al.  . 
Survival after lung metastasectomy for colorectal cancer: importance of previous liver metastasis as a prognostic factor
Eur J Surg Oncol
 , 
2011
, vol. 
37
 
9
(pg. 
786
-
790
)
21
Nojiri
K
Tanaka
K
Nagano
Y
, et al.  . 
Efficacy of surgery for lung metastases from colorectal cancer synchronous to or following that for liver metastases
Anticancer Res
 , 
2011
, vol. 
31
 
3
(pg. 
1049
-
1054
)
22
Riquet
M
Foucault
C
Cazes
A
, et al.  . 
Pulmonary resection for metastases of colorectal adenocarcinoma
Ann Thorac Surg
 , 
2010
, vol. 
89
 
2
(pg. 
375
-
380
)
23
Park
JS
Kim
HK
Choi
YS
, et al.  . 
Outcomes after repeated resection for recurrent pulmonary metastases from colorectal cancer
Ann Oncol
 , 
2010
, vol. 
21
 
6
(pg. 
1285
-
1289
)
24
Hwang
MR
Park
JW
Kim
DY
, et al.  . 
Early intrapulmonary recurrence after pulmonary metastasectomy related to colorectal cancer
Ann Thorac Surg
 , 
2010
, vol. 
90
 
2
(pg. 
398
-
404
)
25
Welter
S
Jacobs
J
Krbek
T
, et al.  . 
Prognostic impact of lymph node involvement in pulmonary metastases from colorectal cancer
Eur J Cardiothorac Surg
 , 
2007
, vol. 
31
 (pg. 
167
-
172
)
26
Van Cutsem
E
Nordlinger
B
Adam
R
, et al.  . 
Towards a pan-European consensus on the treatment of patients with colorectal liver metastases
Eur J Cancer
 , 
2006
, vol. 
42
 (pg. 
2212
-
2221
)
27
Yoo
PS
Lopez-Soler
RI
Longo
WE
Cha
CH
Liver resection for metastatic colorectal cancer in the age of neoadjuvant chemotherapy and bevacizumab
Clin Colorectal Cancer
 , 
2006
, vol. 
6
 (pg. 
202
-
207
)
28
Dimitroulis
D
Nikiteas
N
Troupis
T
, et al.  . 
Role of surgery in colorectal liver metastases: too early or too late?
World J Gastroenterol
 , 
2010
, vol. 
16
 
28
(pg. 
3484
-
3490
)
29
Pihl
E
Hughes
ES
McDermott
FT
, et al.  . 
Lung recurrence after curative surgery for colorectal cancer
Dis Colon Rectum
 , 
1987
, vol. 
30
 (pg. 
417
-
419
)
30
Foster
JH
Treatment of metastatic disease of the liver: a skeptic's view
Semin Liver Dis
 , 
1984
, vol. 
4
 (pg. 
170
-
179
)
31
Tan
KK
Lopes Gde
L
Jr
Sim
R
How uncommon are isolated lung metastases in colorectal cancer? A review from database of 754 patients over 4 years
J Gastrointest Surg
 , 
2009
, vol. 
13
 (pg. 
642
-
648
)
32
O'Connell
JB
Maggard
MA
Ko
CY
Colon cancer survival rates with the new American Joint Committee on Cancer sixth edition staging
J Natl Cancer Inst
 , 
2004
, vol. 
96
 
19
(pg. 
1420
-
1425
)
33
Sanoff
HK
Sargent
DJ
Campbell
ME
, et al.  . 
Five-year data and prognostic factor analysis of oxaliplatin and irinotecan combinations for advanced colorectal cancer: N9741
J Clin Oncol
 , 
2008
, vol. 
26
 
35
(pg. 
5721
-
5727
)
34
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
, vol. 
371
 
9617
(pg. 
1007
-
1016
)
35
Hsu
CW
King
TM
Wang
HT
Wang
JH
Factors that influence survival in unresectable metastatic or locally advanced colorectal cancer
Int J Colorectal Dis
 , 
2011
, vol. 
26
 
12
(pg. 
1559
-
1566
)
36
Tournigand
C
André
T
Achille
E
, et al.  . 
FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study
J Clin Oncol
 , 
2004
, vol. 
22
 
2
(pg. 
229
-
237
)
37
Treasure
T
Pulmonary metastasectomy: a common practice based on weak evidence
Ann R Coll Surg Engl
 , 
2007
, vol. 
89
 (pg. 
744
-
748
)
38
Fiorentino
F
Hunt
I
Teoh
K
, et al.  . 
Pulmonary metastasectomy in colorectal cancer: a systematic review and quantitative synthesis
J R Soc Med
 , 
2010
, vol. 
103
 
2
(pg. 
60
-
66
)