The metastatic behavior of renal cell carcinoma (RCC) Renal Cell Carcinoma Lung Metastases Surgery: Pathologic Findings and Prognostic Factors

Transcription

1 Renal Cell Carcinoma Lung Metastases Surgery: Pathologic Findings and Prognostic Factors Jalal Assouad, MD, Boriana Petkova, MD, Pascal Berna, MD, Antoine Dujon, MD, Christophe Foucault, MD, and Marc Riquet, MD, PhD Department of Thoracic Surgery, Georges Pompidou European Hospital, Paris, France Background. Renal cell carcinoma metastases are more frequently located in the lungs, with surgical results better than in other anatomic locations. Prognosis is darkened by incomplete resection, short disease-free interval, and number of lung metastases (LM). Our purpose was to further review these prognostic factors and related renal cell carcinoma disease characteristics. Methods. From 1984 to 2005, 65 consecutive patients underwent surgery for LM in view of cure. Studied factors were age, sex, smoking habits, forced expiratory volume in 1 second, disease-free interval, adjuvant therapy, size and number of metastases, lymph node involvement, and renal cell carcinoma pathologic staging. These factors were compared with those of 23 patients with previously resected renal cell carcinoma and undergoing surgery for lung cancer during the same period. Results. There were 44 unilateral and 21 bilateral LM; 83 operations were performed, with no postoperative deaths. Lung metastases were classified in four subgroups: single metastasis (n 23), multiple unilateral metastases (n 8), LM and other organ metastasis (n 13), and bilateral LM (n 21). Five-year overall survival (37.2% when resection was complete) was not statistically different among subgroups nor dependent on age, sex, smoking, forced expiratory volume in 1 second, diseasefree interval, and adjuvant therapy, but was significantly influenced by the size of LM and lymph node involvement (univariate and multivariate analyses). Lymph node involvement was less frequent than in patients operated on for lung cancer: respectively, 13 of 65 (20%) and 13 of 23 (56.5%; p ). Intrathoracic metastatic spread was not related to a particular renal cell carcinoma pathologic tumor staging (pt) subgroup. Conclusions. Size of LM and lymph node involvement are important prognostic factors. They suggest a metastatic mode of spread involving the renal lymphatic drainages and specific biologic characteristics acquired by selected tumor cells. (Ann Thorac Surg 2007;84: ) 2007 by The Society of Thoracic Surgeons The metastatic behavior of renal cell carcinoma (RCC) is often bizarre and unpredictable [1], but is of much concern for the thoracic surgeon. In effect, RCC metastases to the lungs (LM) are the most frequent, with prevalence rates as high as 72% and 76% in autopsy studies [1, 2]. Follow-up of patients operated on for localized or locally advanced RCC also demonstrate the lungs to be the most common site of metastases [3]. Worthwhile survival rates are achieved when surgery can remove RCC metastases [4], and better survival is found for LM when compared with other locations [5]. The first case treated by surgery was reported in 1939 [6], and until 1992 published series were based on small cohorts, commonly reporting fewer than 50 patients [7], which was sufficient to assess the prognostic value of surgical treatments but not sufficient to define the best prognostic factors. Since 1994 [8], more important series including between 50 and 191 patients [8 14] aimed to define some relevant prognostic factors for long-term survival after surgery for LM. The most important predictors related with a poor survival were incomplete resection [11 14], Accepted for publication April 27, Address correspondence to Dr Riquet, Department of Thoracic Surgery, Georges Pompidou European Hospital, rue Leblanc, Paris, Cedex 15, 75015, France; marc.riquet@hop.egp.ap.ap-hop-paris.fr. shorter disease-free interval between RCC and lung surgery [8, 10, 11, 13, 14], a higher number of pulmonary lesions [8, 10, 11, 13, 14], and pathologic characteristics such as size of the largest metastasis [12, 14] and lymph node (LN) involvement [9, 11, 12, 14]. However, LN involvement was not detailed and RCC pathologic characteristics were not commonly studied, although mentioned in three papers [8, 12, 14]. The purpose of this study was to review these prognostic factors and to analyze the RCC disease-related characteristics. Patients and Methods From January 1984 to December 2005, 76 consecutive patients underwent lung surgery for RCC metastases at Georges Pompidou European Hospital and Boisguillaume Medico-surgical Center. Surgery was performed for diagnostic purpose in 8 patients, for tumor harvest for immunology in 3 patients, and in view of cure in 65 patients. These 65 patients form the basis of this retrospective analysis, for which our institutional review board waived obtaining patient consent. Data were obtained from hospital case records. The RCC was treated by radical nephrectomy in all patients. There were by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg ASSOUAD ET AL 2007;84: RENAL CELL CARCINOMA METASTASES SURGERY men and 13 women aged between 42 and 82 years (mean age, ). Metastases were distributed according to four main subgroups (single metastasis in one lung, more than one metastasis in one lung, LM associated with other organresected metastasis, and bilateral metastases). Renal cell carcinoma nephrectomy study was obtained for all patients but 3 (operated on at the beginning of the study). The LM and RCC pathologic (and TNM [15]) characteristics were studied in each subgroup, and the pathologic study results were analyzed and compared among them. During the same period, 23 patients with a past history of nephrectomy for RCC underwent surgery for a solitary pulmonary nodule demonstrated to be a non small-cell lung cancer (NSCLC): squamous cell carcinoma (n 7), adenocarcinoma (n 10), large cell carcinoma (n 3), and neuroendocrine large cell carcinoma (n 3). These patients were compared with those patients who underwent surgery for unilateral LM to compare their pathologic and prognostic characteristics. Follow-up information was obtained either from hospital case records or from a questionnaire completed by the local chest physician or general practitioner, or from death certificates. The overall survival was the main outcome, and was defined as the time interval between the date of surgery for LM and the date of death or the last follow-up for censored patients. No patient was lost at follow-up. Actuarial survival curves were estimated by Kaplan Meier method. Statistical comparisons among subgroups were made using the log-rank test. Univariate analyses used the following explanatory variables: age, sex, smoking habits, forced expiratory volume in 1 second, disease-free interval, adjuvant therapy, diameter of the largest metastasis, number of metastases, and mediastinal LN involvement. Hazard ratios with their 95% confidence intervals were estimated. The cut-off p value Table 1. Main Characteristics of the Patients Variable Result Age (y) Sex Male (n) 52 (80%) Female (n) 13 (20%) Smoker (n) 41 (63.1%) FEV 1 (%) Past medical history (n) 23 (35.4%) Other cancer history (n) 3 (4.6%) DFI 12 (mo) 16 (24.6%) (mo) 37 (56.9%) 60 (mo) 12 (18.5%) Postoperative complications (n) 10 (15.4%) Adjuvant therapy (n) 13 (20%) 5-year survival 34.4% Median (mo) 35 DFI disease-free interval; FEV 1 forced expiratory volume in 1 second. used for entering specific explanatory variables into the multivariate analysis was The p values were adjusted according to the multiplicity of tests using the Bonferroni method. All analyses were conducted using a two-sided test. A p value less than 0.05 was considered statistically significant. The statistical software used for analysis was SEM (CAC J. Perrin, Clermont Ferrand, France) [16]. Results 1115 The main characteristics of the patients are shown in Table 1. The mean interval between the nephrectomy and the first surgical procedure for lung metastases was months. Metastases were discovered at the initial diagnosis in 7 RCC patients (synchronous LM for stage IV patients). The LM were unilateral in 44 patients and bilateral in 21. Thoracic surgery was performed by staged posterolateral thoracotomy, 6 of which were video-assisted (bilateral, n 2; unilateral, n 2). There were 53 unilateral and 12 bilateral procedures. Four patients underwent reoperation on the first operated side, and 1 patient underwent three operations on the same lung. Total lung resections (n 83) consisted of one or more wedges in the same lung (n 47), lobectomies (n 17), segmentectomies (n 11), lobectomies plus wedges (n 5), pneumonectomies (n 2), and completion pneumonectomy (n 1). Mediastinal lymphadenectomy was performed in 67.7% of patients (44 of 65). There were no postoperative deaths. All resections of metastases were complete. However, complete resections themselves accounted for only 83% of patients (54 of 65): incomplete resection was determined by the presence of bilateral metastases for which the planned second thoracotomy was not carried out (because of progression of the disease, n 6; contralateral small metastases kept under observation, n 2; and regression of metastases after adjuvant therapy, n 1) and by the fact of unilateral metastases with other organ involvement (vertebroplasty, n 1; brain metastasis stereotaxic radiosurgery, n 1). The four subgroups of lung metastases were as follows: single metastasis in one lung (n 23), more than one metastasis in one lung (n 8), LM associated with another organ resected RCC metastasis (n 13), and bilateral LM (n 21). Twelve patients with bilateral LM had an operation on both sides, whereas 9 patients underwent only one lung surgery: among them, 1 patient had a removed other organ metastasis (adrenal gland). The main clinical and pathologic characteristics of the subgroups of patients are shown in Table 2. The subgroups were similar, except when lung surgery was performed after resection of another organ metastasis. In such patients (n 14), the disease-free interval after nephrectomy was longer. Eight anatomic structures were involved: pancreas (n 4), adrenal gland (n 4), contralateral kidney (n 1), bone (n 2), brain (n 2), diaphragm (n 2), thyroid gland (n 1), and extraabdominal LN (n 1). GENERAL THORACIC

3 1116 ASSOUAD ET AL Ann Thorac Surg RENAL CELL CARCINOMA METASTASES SURGERY 2007;84: Table 2. Main Clinical and Pathologic Characteristics of the Lung Metastases Subgroups of Patients a Variable 1 Metastasis (n 13) Unilateral 2 Metastasis (n 8) Other is Metastasis (n 13) 2 Operated Sides (n 12) Bilateral 1 Operated Side (n 8) p Value DFI 12 mo 5 (21.7%) 3 (37.5%) 0 1 (8.3%) 6 (75%) mo 14 (60.9%) 4 (50%) 9 (69.2%) 9 (75%) 1 (12.5%) 60 mo 4 (17.4%) 1 (12.5%) 4 (30.8%) 2 (16.7%) 1 (12.5%) Largest size metastasis (mm) NS Lung metastasis side NS R 11 (47.8%) 5 (62.5%) 8 (61.5%) L 12 (52.2%) 3 (37.5%) 5 (38.5%) RCC side NS R 11 (47.8%) 3 (37.5%) 4 (30.8%) 7 (58.3%) 4 (50%) L 12 (52.2%) 5 (62.5%) 9 (69.2%) 5 (41.7%) 4 (50%) Lung metastasis and RCC 11 (47.8%) 6 (75%) 7 (53.8%) 5 (41.7%) 4 (50%) NS same side (RCC) T b T1 5 (22.7%) 0 2 (15.4%) 1 (9.1%) 2 (20%) NS T2 6 (27.3%) 1 (16.7%) 6 (46.2%) 1 (9.1%) 1 (10%) T3a 5 (22.7%) 4 (66.6%) 0 5 (45.5%) 1 (10%) T3b 6 (27.3%) 1 (16.7%) 5 (38.4%) 4 (36.3%) 4 (40%) Lung metastasis N0 19 (82.6%) 6 (80%) 10 (76.9%) 11 (91.7%) 5 (62.5%) NS LN involvement N 4 (17.4%) 2 (20%) 3 (23.1%) 1 (8.3%) 3 (37.5%) Survival 5 years (%) Median a Lymph node (LN) renal cell cancer (RCC) involvement was present in only 2 patients at the time of nephrectomy. The patient with bilateral lung and other organ metastasis is not incorporated in this table. b Data not available for 4 patients. DFI disease-free interval; NS not significant. Five-year survival was 34.4% for the whole group (n 65), and 37.2% when the surgical resection was complete (n 54). Survival was not different among the subgroups (Table 2). However, there was a clear tendency to poor prognosis for bilateral LM when the planned second intervention was not performed, without long-term survivors when the surgical contraindication was related to disease progression. Survival was not significantly related to age, sex, smoking habits, forced expiratory volume in 1 second, or disease-free interval, even if a tendency to poor prognosis was observed after short intervals (disease-free interval 12 months, 5-year survival, 27.1%; median, 17 months; 13 disease-free interval 60 months, 5-year survival, 35.9%; median, 43 months; and disease-free interval 61 months, 5-year survival, 42.3%; median, 37 months; p 0.21) and after adjuvant treatment. Five-year survival was significantly influenced by the size of the largest LM (Fig 1) and by the hilar and mediastinal LN involvement; no 5-year survivors were observed in cases of metastatic LN (Fig 2). Multivariate analysis confirmed size of LM (p ) and LN involvement (p ) to be independent indicators of poor prognosis. Mean size of the largest LM was mm in cases with LN involvement and mm in cases without (p 0.48). Lymph node involvement was significantly less frequent in patients with LM than in patients operated on for NSCLC after a previously resected RCC: 13 of 65 (20%), and 13 of 23 (56.5%), respectively (p ). The type of LN involvement was similar in each population (N1 6 and N2 7). However, there were more skip mediastinal LN metastases in cases of RCC LM (4 of 7 versus 1 of 7; p 0.26), and more intralobar (lobar, segmental, and subsegmental) LN involvement in cases of NSCLC patients (8 of 13 versus 1 of 13; p 0.013). Survival rates were not different between both populations (NSCLC, 29.5%; median, 27 months versus RCC LM, 34.4%; median, 35 months; p 0.69), and remained not significant when only unilateral RCC LM (n 31) were considered (5-year survival, 32.2%; median, 37 months; p 0.71). Five-year survival rates were not different in cases of N0 (NSCLC, 59.8% versus RCC LM, 43.1%; p 0.62), nor in cases of LN involvement (NSCLC, 12.4% versus RCC LM, 0%; p 0.53). When RCC LM were analyzed with regard to RCC disease, it appeared that (1) the RCC regional LN involvement was rare: 2 patients of 65 (one with bilateral and the other with other organ metastases) whose survivals were short (16 and 25 months); (2) the largest RCC tumors were observed in the subgroup of patients with multiple unilateral LM. These tumors more frequently presented with an extension in the pericapsular fat (66%):

4 Ann Thorac Surg ASSOUAD ET AL 2007;84: RENAL CELL CARCINOMA METASTASES SURGERY Figure 1. Survival and lung metastases size. Curve 1 is lung metastases less than 20 mm, median, 54 months (n 37); curve 2 is lung metastases greater than mm, median, 22 months (n 28). their 5-year survival rates after lung surgery was poor (25%; Table 2); (3) the lowest rates of LN involvement were observed in cases of completely resected bilateral LM (1 of 12, 8.3%), whereas higher rates were registered when the contralateral lung had not been operated on (3 of 8: 37.5%); and (4) LM with LN involvement and their corresponding RCC pathologic T staging are shown in Figure 3. No relationship could be established between the RCC ptnm and the intrathoracic lung and LN metastatic mode of spread, notably concerning the role played by the renal and vena caval venous involvement (pt3b) resection is complete. The number of resected metastases was also reported to be an important prognostic factor of mortality [8, 10 14]. Murthy and coworkers [14] demonstrated that the number of pulmonary metastases is inversely related to the ability to achieve complete resection, and presented high-resolution spiral computed tomography as an essential component of the workup: when three or fewer metastases are identified on preoperative computed tomography, the chance of complete resection exceeds 80%; conversely, if more than six lesions are seen, the chance of incomplete resection is 80% or higher. However, as Fourquier and coworkers [9] noted, we found that the number of metastases had no influence on prognosis, and as van der Poel and coworkers reported [5], we observed that consecutive resection was related to longer survival. We suggest that the poor prognosis reported by some publications in cases of multiple LM may be explained by the uncertainty of the completeness of resection. Similar conclusions can be applied concerning results of surgery for LM and extrathoracic metastases. Surgery must be performed when all other visceral metastases can be completely resected. In our series, the results for this subgroup of patients are as good as those obtained for the whole series, which Murthy and coworkers [14] and Cerfolio and coworkers [8] also reported. There is no explanation for this phenomenon, probably attributable to a particular biology of these multiorgan metastases. In our series, this subgroup of patients is the one presenting the longest disease-free interval (Table 2). Shorter disease-free interval is an important risk factor for mortality [8, 10, 11, 13]. However, it is not any longer an important risk factor for mortality when the resection is complete, which confirms that the residual disease is the more critical factor. When resection is complete, it is GENERAL THORACIC Comment Our study demonstrates that RCC LM surgery is safe and potentially curative in more than one third of patients, results commonly reported in the literature. Postoperative mortality is low, between 0% [8, 13, 14] and 2.1% [11]. Postoperative morbidity is generally around 10% [8, 12, 14], and has been reported as low as 1.5% [9]. Five-year survival rates range from 31% to 40% [8, 10 14]. Age, sex, past medical history, and other clinical data do not appear to be prognostic factors, but the published series are still too small to disclose any significant difference. Murthy and coworkers [14] report decreased preoperative forced expiratory volume in 1 second to be an important risk factor for mortality (p 0.02), which was not observed in our series. All these prognostic factors are not related to the RCC pathologic characteristics and cannot give any information on the RCC metastases surgery prognostic value itself. The main prognostic factor of long-term survival is the completeness of the resection, and 5-year survival rates are commonly between 40% and 45% [9, 11 14] when the Figure 2. Survival and lung metastases lymph node involvement. Curve 1 is no involvement, median, 37 months (n 52); curve 2 is lymph node involvement, median, 14 months (n 13).

5 1118 ASSOUAD ET AL Ann Thorac Surg RENAL CELL CARCINOMA METASTASES SURGERY 2007;84: Figure 3. Lung metastases lymph node involvement and corresponding renal cell carcinoma pt. (a) Bilateral lung metastases. (b) Lung metastases and other organ metastasis. (c) Unilateral lung metastases. (DFI diseasefree interval; 5 subaortic lymph node; 4R and 4L right and left paratracheal lymph node; N1, N2 hilar and mediastinal lymph node involvement; 9 pulmonary ligament lymph node; 7 subcarinal lymph node; T1a, T1b, T2, T3a, T3b renal cell carcinoma TNM classification [15]; 3a phrenic lymph node.) conceivable that longer disease-free interval reflects less aggressive cancer [14]. Patients with synchronous metastases have a higher risk [13]. Although affecting prognosis, synchronous presentation of RCC and LM does not preclude operation if complete resection is expected and can be achieved [14]. Two prognostic factors are particularly relevant: the size of the largest LM and the LN involvement. Jett and coworkers [17] demonstrated in 1983 that the diameter of the largest resected metastasis affected survival, but a study published in 1994 that included more patients was not conclusive [8]. In our cohort, patients with metastases larger than 2 cm had significantly shorter survival rates than patients with smaller metastases, which Piltz and coworkers [12] also observed. Murthy and coworkers [14] report that the prognosis is poor when the LM diameter is larger than 3 cm. Prognosis is even poorer for diameters larger than 4 cm [12]. This prognostic value does not appear to be related to the number of resected LM at the same occasion. The reason why this pathologic feature is

6 Ann Thorac Surg ASSOUAD ET AL 2007;84: RENAL CELL CARCINOMA METASTASES SURGERY an indicator of poor prognosis remains speculative. However, our study demonstrated that the largest metastases seem to coexist with hilar and mediastinal LN involvement, despite the fact that both features are independent factors of prognosis. This should motivate further research. Mediastinal LN involvement frequency ranges from 12.5% [8] to 56.3% [14]. Whatever the percentage, intrapulmonary LN (N1) involvement appears as frequently as mediastinal LN (N2) involvement: 6 and 5 for Fourquier and coworkers [9] and Cerfolio and coworkers [8], 8 and 10 for Murthy and coworkers [14], and 6 and 7 in our series, respectively. Lymph node involvement is an important prognostic factor for mortality: this was first reported in the case of mediastinal LN (N2) involvement [9], proved commonly significant for global LN (N1 N2) involvement, and appears also confirmed for separate LN (N1) involvement [11]. Lymph node involvement by hematogenous metastasis to the lung being unusual, Cerfolio and coworkers [8] suggested that LM be considered as NSCLC in cases of LN involvement and to manage it as such if the LM cannot be differentiated from NSCLC at the time of operation. Because LN involvement is found when performing LM surgery, Pfannschmidt and coworkers [11] and Murthy and coworkers [14] hypothesize that LM metastases induce LN metastases, and also suggest treating solitary LM similarly to NSCLC. However, regardless of the hypothesis, the surgical guidelines are the same. In our series, LN involvement was possible whatever the number and the location of LM, but its frequency was not as high as in the cases of NSCLC. This does not suggest an involvement based on the lymphatic drainage of the LM, but a phenomenon more likely based on a particular RCC metastatic mode of spread. Reinke and coworkers [18] and more recently Assouad and coworkers [19], reported that hilar and mediastinal LN involvement can be present without associated LM. Lymph node involvement may be explained in those cases by the renal lymph drainage: the renal lymph rapidly and preponderantly drains into the thoracic duct and may backflow into the lymphatics and mediastinal or hilar LN [18, 19]. The tendency we observed to more skip metastases and less intralobar LN involvement in patients with RCC lung metastases, and the tendency to more hilar or interlobar LN involvement observed by others in such patients, may also be explained by this kind of malfunction of the renal lymphatic drainage. The pathologic and TNM characteristics of the primary RCC are rarely available in the literature, and when available they are generally not correlated with the pathologic characteristics of the LM when discussing the results of surgery [8, 14, 20]. Piltz and coworkers [12] reported that there are no survivors after LM resection when regional LN are involved at the time of the RCC surgery. In our series, RCC LN metastases were present in only 2 patients (none survived), but the number is too small to be supportive. Pantuck and coworkers [21] studied the role of LN dissection in RCC surgery and demonstrated that survival of patients with regional LN involvement only was identical to that of patients with distant metastatic disease only, and that patients with regional nodes and distant metastases had significantly inferior survival compared with those with only one metastatic condition. To explain what is observed in our and other series, we suggest that most patients amenable to LM surgery are mainly those without retroperitoneal LN involvement and thus already pertaining to a selected subgroup. We observed that RCC pt subgroup distribution (Table 2) was not significantly correlated with the pathologic pattern (size, number, N involvement) of the resected LM. Especially T3b, which means renal and inferior caval venous involvement, was not associated with a greater systemic metastatic dissemination. These findings do not support an exclusive hematogenous metastatic mode of spread. They may tally with a mode of spread involving the renal lymphatic drainage and its anatomic, physiologic, and biologic characteristics, acquired or already present in some selected renal tumor cells. Surgery for RCC LM provides worthwhile survival rates. However, the pattern of metastases occurrence is not yet definitely understood and appears as a pathologic and therapeutic major topic that deserves further research. References Weiss L, Harlos JP, Torhost J, et al. Metastatic patterns of renal carcinoma: an analysis of 687 necropsies. J Cancer Res Clin Oncol 1988;114: Saitoh H. Distant metastasis of renal adenocarcinoma. Cancer 1981;48: Sivaramakrishna B, Gupta NP, Wadhwa P, et al. Pattern of metastases in renal cell carcinoma: a single institution study. Indian J Cancer 2005;42: Kierney PC, van Heerden JA, Segura JW, Weaver AL. Surgeon s role in the management of solitary renal cell carcinoma metastases occurring subsequent to initial curative nephrectomy: an institutional review. Ann Surg Oncol 1994;1: Van der Poel HG, Roukema JA, Horenblas S, van Geel AN, Debruyne FM. Metastasectomy in renal cell carcinoma: a multicenter retrospective analysis. Eur Urol 1999;35: Barney JD, Churchill EJ. Adenocarcinoma of the kidney with metastasis to the lung. J Urol 1939;42: Pogrebniak HW, Haas G, Linehan WM, Rosenberg SA, Pass HI. Renal cell carcinoma: resection of solitary and multiple metastases. Ann Thorac Surg 1992;54: Cerfolio RJ, Allen MS, Deschamps C, et al. Pulmonary resection of metastatic renal cell carcinoma. Ann Thorac Surg 1994;57: Fourquier P, Regnard JF, Rea S, Levi JF, Levasseur P. Lung metastases of renal cell carcinoma: results of surgical resection. Eur J Cardiothorac Surg 1997;11: Friedel G, Hurtgen M, Penzestadler M, Kyriss T, Toomes H. Resection of pulmonary metastases from renal cell carcinoma. Anticancer Res 1999;19: Pfannschmidt J, Hoffmann H, Muley T, Krysa S, Trainer C, Dienemann H. Prognostic factors for survival after pulmonary resection of metastatic renal cell carcinoma. Ann Thorac Surg 2002;74: Piltz S, Meimarakis G, Wichmann MW, Hatz R, Schildberg FW, Fuerst H. Long-term results after pulmonary resection of renal cell carcinoma metastases. Ann Thorac Surg 2002; 73: GENERAL THORACIC

7 1120 ASSOUAD ET AL Ann Thorac Surg RENAL CELL CARCINOMA METASTASES SURGERY 2007;84: Hofmann HS, Neef H, Krohe K, Andreev P, Silber RE. Prognostic factors and survival after pulmonary resection of metastatic renal cell carcinoma. Eur J Urol 2005;48: Murthy SC, Kim K, Rice TW, et al. Can we predict long term survival after pulmonary metastasectomy for renal cell carcinoma? Ann Thorac Surg 2005;79: Wittekind C, Green FL, Hutter RVP, Klimpfinger M, Sobin LH. TNM atlas, 5th ed. UICC. Berlin: Springer-Verlag, Kwiatkowski F. SEM (Statistiques, Epidémiologie, Médecine): un outil de gestion informatique et statistique adapté à la recherche en cancérologie. Bull Cancer 2000;87: Jett JR, Hollinger CG, Zinsmiester AR, Pairolero PC. Pulmonary resection of metastatic renal cell carcinoma. Chest 1983;84: Reinke RT, Higgins CB, Nikayama G, Harris RH, Friedman PJ. Bilateral pulmonary hilar lymphadenopathy. An unusual manifestation of metastatic renal cell carcinoma. Radiology 1976;121: Assouad J, Riquet M, Berna P, Danel C. Intrapulmonary lymph node metastasis and renal cell carcinoma. Eur J Cardiothorac Surg 2007;31: Katzenstein AL, Purvis R, Gmelich J, Askin F. Pulmonary resection for metastatic renal adenocarcinoma. Pathologic findings and therapeutic value. Cancer 1978;41: Pantuck AJ, Zisman A, Dorey F, et al. Renal cell carcinoma with retroperitoneal lymph nodes: role of lymph node dissection. J Urol 2003;169: Notice From the American Board of Thoracic Surgery Regarding Trainees and Candidates for Certification Who Are Called to Military Service Related to the War on Terrorism The Board appreciates the concern of those who have received emergency calls to military service. They may be assured that the Board will exercise the same sympathetic consideration as was given to candidates in recognition of their special contributions to their country during the Vietnam conflict and the Persian Gulf conflict with regard to applications, examinations, and interruption of training. If you have any questions about how this might affect you, please call the Board office at (312) Carolyn E. Reed, MD Chair The American Board of Thoracic Surgery 2007 by The Society of Thoracic Surgeons Ann Thorac Surg 2007;84: /07/$32.00 Published by Elsevier Inc