Surgical resection of pulmonary metastases is an established

Long-Term Results After Pulmonary Resection of Renal Cell Carcinoma Metastases Stefan Piltz, MD, Georgios Meimarakis, MD, Matthias W. Wichmann, MD, Rudolf Hatz, MD, Friedrich Wilhelm Schildberg, MD, and Heinrich Fuerst, MD Department of Surgery and Thoracic Surgery, Klinikum Grosshadern, Ludwig-Maximilia-University Munich, Munich, Germany Background. Until now no conclusive data exist regarding the factors influencing long-term survival after pulmonary resection of renal cell carcinoma metastases. The aim of the present study, therefore, was to discover definitive prognostic factors for survival using a large and homogeneous single center patient cohort. Methods. Between 1980 and 2000, 105 patients, after curative resection of lung metastases from renal cell carcinoma, were followed in this long-term study. These patients underwent a total of 150 surgical procedures. analysis was done using the Kaplan-Meier method and the log-rank test. Multivariate analysis of prognostic factors was performed using the Cox multivariate proportional hazard model. Results. Median survival after curative resection reached 43 months (range, 1 to 218 months). at 3, 5, and 10 years was 54%, 40%, and 33%, respectively. Univariate analysis revealed that a complete resection, a less than 4-cm diameter of the metastases and tumor-free lymph nodes at the time of primary operation, were highly significant dependent prognostic factors (p < 0.001). These factors were also shown to be independent prognostic factors as suggested by multivariate analysis (p < 0.05). Conclusio. The size of the metastatic nodule, the completeness of pulmonary resection, and the lymph node status at the time of nephrectomy are the most important prognostic factors that influence survival after resection of pulmonary metastases. Recurrence of resectable pulmonary metastases does not impair survival, thus favoring repeated resection in patients with recurrent disease. (Ann Thorac Surg 2002;73:1082 7) 2002 by The Society of Thoracic Surgeo Surgical resection of pulmonary metastases is an established treatment modality in patients with metastatic disease. Metastases of primary tumors that do not respond well to chemotherapy, radiotherapy, or a combination of both therapies are especially best suited for surgical resection. Renal cell carcinoma is known to belong to this category of malignant disease. To date, reports in the literature do not adequately discern those factors that may have the highest impact on long-term survival after resection of lung metastases in renal cell carcinoma. A major problem of most studies is that the number of patients included does not allow for a proper statistical evaluation. A second problem is that patients with different primary tumors are recruited for these follow-up studies. The analysis of factors of possible prognostic relevance for survival is therefore questionable, as the different biological aspects of certain tumors were not coidered in these studies [1, 2]. The present study, therefore, aimed to define prognostic factors that are of significant relevance for long-term survival after resection of pulmonary renal cell carcinoma metastases in a homogeneous and large single center patient series. Accepted for publication Dec 3, 2001. Address reprint requests to Dr Piltz, Department of Surgery and Thoracic Surgery, Klinikum Grosshadern, Ludwig-Maximilia-University Munich, Marchioninistrasse 15, D-81377 Munich, Germany; e-mail: spiltz@ gch.med.uni-muenchen.de. Patients and Methods Between January 1980 and January 2000, a total of 105 patients (73 men, 32 women) underwent thoracotomies for curative resection of pulmonary renal cell carcinoma metastases at our Department of Surgery. A total of 150 procedures were performed on these patients. The mean time period between the primary procedure on the kidney and pulmonary metastasectomy was 59 months. The median patient age at the time of metastasectomy was 59 years, with a range from 35 to 78 years. Histopathologic evaluation of the resected specime revealed true renal cell carcinoma metastases in all patients. Complete resection for cure was defined as follows: the lung is the unique site of metastasis, no additional extrapulmonary sites of metastatic disease or locoregional recurrence in the preoperative examinatio exist (roentgenogram, thoracic computed tomographic scan, ultrasound, or abdominal computed tomographic scan). Furthermore, no macroscopic tumor tissue is left behind and the tumor is not opened. Finally, tumor-free resection margi are required. The following factors of potential prognostic significance were selected for statistical analysis: age, gender, diameter of the largest resected lung metastasis, TNM classification, tumor grading, number of lung metastases in each patient, and disease-free interval. Disease-free interval was defined as time period between curative 2002 by The Society of Thoracic Surgeo 0003-4975/02/$22.00 Published by Elsevier Science Inc PII S0003-4975(01)03602-5

Ann Thorac Surg PILTZ ET AL 2002;73:1082 7 PULMONARY RESECTION OF RENAL CELL CARCINOMA METASTASES 1083 primary operation on the kidney and first detection of a local recurrence or metastatic disease. To evaluate the impact of complete surgical resection we compared our 105 patients with 17 patients undergoing operation without curative intention during the same period. The reason for incomplete resection in these patients was that microscopic (n 3) or macroscopic (n 8) amounts of malignant tissue were being left behind. Three operatio with palliative intention (eg, for treatment of metastasis-related pneumonia) and three exploratory thoracotomies were done. Except for the calculation of the impact of completeness, statistical analyses are based on data obtained in patients treated for cure. Follow-up was based on data provided by the Munich tumor register, by the regional general practitioners, as well as the hospitals participating in the Munich tumor center. All patients or their families were contacted by phone every second year. Statistical analysis was done with the Kaplan-Meier method and the log-rank test [3, 4]. Multivariate analysis of prognostic factors was performed using the Cox multivariate proportional hazard model [5]. A p value less than 0.05 was coidered statistically significant. Results At the last follow-up evaluation (January 2000) 48 of the 105 patients were still alive (46%). Median survival was 99 months (range, 6 to 248 months) after resection of the primary renal cancer and 61 months after pulmonary metastastectomy. Fifty-seven patients had died after a median time period of 50 months after removal of their renal tumor and 19 months after resection of the pulmonary metastasis. All patients but one had died because of progressive renal cancer. Operative Procedures Sixty-eight patients underwent unilateral thoracotomy for metastases that only affected one side and 28 patients with metastases on both sides had two sequential thoracotomies, within a time interval ranging from 2 to 6 weeks. Metastases on both sides were removed by midline sternotomy in 8 patients. In 1 patient exploration of the contralateral side was performed through a sternotomy, despite not having any preoperative proof of metastatic disease on both sides. This exploration did not reveal a malignant lesion. One hundred fifty procedures were performed on 105 patients. Of these procedures, 118 were performed as wedge resectio (79%), in 9 patients as segmental resectio, and in 23 major resectio (19 lobectomies and 4 bilobectomies) were necessary. All patients undergoing resection for cure were operated on through an open thoracotomy. Number of Resected Metastases Forty-nine patients (47%) had a single lung metastasis, 14 patients had two metastases, 12 had three metastases, and 30 patients had more than three metastases. The median number of metastases resected was two. One patient in whom 32 histologically confirmed metastases were resected (20 on the right side, 12 on the left side) survived pulmonary metastasectomy for 132 months, and at his last clinical follow-up evaluation no metastases were detected. Complicatio In 16 procedures (of 150, 10.7%), a complication was observed. One patient had a rethoracotomy for postoperative bleeding from the chest wall, and 1 patient hemorrhaged and required a blood trafusion but no additional procedure. Three patients were reintubated because of respiratory iufficiency and one had prolonged respiratory support for more than 12 hours. Atelectasis occurred in 2 patients and required bronchoscopic clearance. Five patients had superficial wound infection that was resolved under local wound therapy. After a thoracotomy on the left side, 1 patient suffered from paralysis of the recurrent nerve. One patient (1 of 105; 0.95%) died at 34 days from severe sepsis after a lower bilobectomy and angioplastic resection of the pulmonary vein and a subsequent pleural empyema. Adjuvant Treatment Regime Additional nourgical adjuvant treatment was administered to 44 patients (42%). Eighteen patients had radiotherapy alone. In 2 patients radiotherapy was combined with chemotherapy and immunotherapy and 1 patient underwent chemoradiation. Seven patients were treated with chemotherapy only. A combined chemotherapy and immunotherapy was administered to 5 patients and another 11 patients received immunotherapy only. Twelve percent (13 of 105) of all of the study patients underwent rethoracotomy. In 1 patient a thoracotomy was performed six times. The median time period until recurrence of pulmonary metastases was 10 months. The median survival time of patients who underwent a rethoracotomy was 46 months, which is longer than survival for patients after a single metastastectomy (median survival, 40 months). The difference, however, did not reach statistical significance. Statistical Analysis The results of the univariate analysis are shown in Table 1. Highly significant differences (p 0.001) were found for the following measures: complete versus incomplete resection of the metastases (Fig 1); size of the metastasis ( 4 cm versus 4 cm); and lymph node status at the time of the primary renal operation (Fig 2). Significant differences (p 0.05) were observed for the number ( 2 versus 2; Fig 3) and the size ( 2 cm versus 2 cm; Fig 4) of the metastases. As shown in Table 1, other factors were found not to be significant. Multivariate analysis revealed that complete and incomplete resection of pulmonary metastasectomy (p 0.0016; relative risk 3.56; odds ratio 1.61 to 7.83) as well as the lymph node status at the time of primary tumor operation (pn) (p 0.031; relative risk 1.56; odds ratio 1.04 to 2.34) are independent prognosis factors. If only patients resected for cure are coidered (n 105) lymph node status at the time of primary operation (pn) (p

1084 PILTZ ET AL Ann Thorac Surg PULMONARY RESECTION OF RENAL CELL CARCINOMA METASTASES 2002;73:1082 7 Table 1. Univariate Analysis of Prognostic Variables for Factor 0.049, relative risk 1.61; odds ratio 1.0 to 2.59) and the size of the metastasis (p 0.0016; relative risk 2.42; odds ratio 1.40 to 4.20) are independent prognostic factors. Comment p Value Complete resection a 0.001 Sex Number of lesio 2 0.029 Diameter 2 cm 0.008 Diameter 3 cm 0.006 Diameter 4 cm 0.001 Disease-free interval pt pn 0.001 pm Grade a Analysis based on 105 complete and 17 incomplete resected patients. not significant. The selection of patients with metastatic disease to the lung who will profit from pulmonary metastasectomy and subsequently show a significant improvement in the survival rate is still a major issue of thoracic surgery. Some patients are refused proper surgical therapy, or the operation performed is not radical enough, whereas others are overtreated and the indication for operation is exaggerated. It appears that more than 50% of the patients with lung metastases originating from sarcomas, germ-line tumors, and pediatric tumors are coidered for metastasectomy; such a broad indication for surgical treatment does not seem to exist for pulmonary metastases from epithelial tumors [6]. Preoperative diagnostic assessment must analyze functional operability of the patient as well as curative resectability of the metastases. Our data clearly indicate that, after pulmonary metastasectomy, residual tumor tissue prohibits long-term survival and there is no obvious benefit for the patient. The same holds true for extended resectio (eg, partial resection of the chest wall or the diaphragm) where the tumor has to be completely resected and must not be opened. In our series none of the 17 patients with incomplete resection lived longer than 29 months as opposed to 62% of the patients after complete resection, who were still alive at that time. These data confirm the results of other investigators, who also observed significant survival benefits only for patients with complete resection of lung metastases [7, 8]. On the basis of our findings we cannot recommend video-assisted thorascopic surgery for metastases if curative resection is intended. Sufficient palpation of the lung in atelectasis cannot be guaranteed using thoracoscopic techniques that can result in a higher risk of missing the smaller lesio. In a prospective study Loehe and colleagues [9] demotrated that even if a helical computed tomographic scan is used before operation one-third of all patients still suffer from additional undetected malignant lesio. These findings are also supported by other studies [6, 7, 10 12]. In the future, intraoperative ultrasound may prove to be sufficient for the detection of additional metastases that were not seen on preoperative computed tomographic sca. It is not generally accepted among thoracic surgeo to principally explore both lungs if metastases are discovered preoperatively only on one side [6, 7]. For the time being, we do not routinely perform bilateral exploration in patients with metastases from renal carcinoma as opposed to patients suffering from germ-line tumors or sarcomas. Ten of the 13 patients who underwent repeated thoracotomies suffered from ipsilateral recurrence, two recurrences were contralateral and one was bilateral. Our findings indicate that in most patients (11 of 13) recurrent metastases are found on the side that had already been completely explored during the previous operation. The survival data reported here (Table 2) support the notion that exploration may be limited to the Fig 1. Probabiltity of survival of the patients undergoing curative (n 105) and incomplete pulmonary resection (n 17) of renal cell carcinoma metastases. Zero on the x-axis represents the date of the first pulmonary resection (p 0.001). (pat. patients.)

Ann Thorac Surg PILTZ ET AL 2002;73:1082 7 PULMONARY RESECTION OF RENAL CELL CARCINOMA METASTASES 1085 Fig 2. Probability of survival after complete resection according to involvement of regional lymph nodes (pn) at the time of primary renal operation. Patients with tumorfree lymph nodes have a highly significant better survival rate (p 0.001) than patients with regional (local) lymph node metastases. (pat. patients.) affected side in patients with metastases from renal cancer. Until now, there is no report in the literature that a bilateral exploration performed in every patient may lead to an improvement in rate of survival. Furthermore, unnecessary thoracotomy leads to adhesio and radiologic artifacts that could limit the seitivity and specificity of diagnostic tools such as computed tomographic sca during follow-up. Our observatio indicate that patients with a maximum metastasis diameter larger than 2 cm had a significantly shorter survival rate than did patients with smaller metastases. In a report by Cerfolio and associates [7], it was found that the size of the metastasis was of no prognostic significance. In an earlier study from the same group, Jett and colleagues [13] reported a significantly increased survival rate in patients with metastases larger than 3 cm. These reports are not in agreement with our observation that larger size metastases have a significant influence on long-term survival and this becomes even more obvious with metastases larger than 4 cm (p 0.001). A number of studies [6, 7, 14, 15] have shown cut-off values for disease-free interval ranging between 20 and 36 months with significantly improved survival in patients with a longer disease-free interval. In our series we were unable to confirm this effect of disease-free interval on survival. Two recent reports [8, 16] appear to support our results. Nonetheless, there is only one study that reports findings in a cohort with the same sample size and homogeneity as reported here [7]. Interestingly, we observed a significant relation between the existence of lymph node metastasis at the time of removal of the primary tumor and long-term survival. Multivariate analysis revealed that lymph node metastases at the time of primary cancer resection is an independent prognostic factor. A similar trend was observed by Fourquier and co-workers [8] in an earlier study. Due to the small patient number in this study the difference in Fig 3. Probability of survival after complete resection in relation to the number of resected pulmonary metastases (p 0.029). (pat. patients.)

1086 PILTZ ET AL Ann Thorac Surg PULMONARY RESECTION OF RENAL CELL CARCINOMA METASTASES 2002;73:1082 7 Fig 4. Probability of survival after complete resection in selection to the largest diameter of the resected pulmonary lesion. Patients with a metastasis smaller than 2 cm have a significantly better survival rate (p 0.008). (pat. patients.) survival did not reach statistical significance. These data confirm an autopsy study [17] of 554 perso in whom renal carcinoma had remained undiagnosed during their lifetimes. In 88 patients renal cancer was the cause of death. Lymphatogenous dissemination was detected in 91% of all these patients and 85% had additional, mostly multifocal metastatic spread. Therefore, it appears that tumors that metastasize into the lung through the bloodstream alone display better tumobiologic behavior than tumors that also metastasize by way of the lymphatics. Because the nodal status is a routine measure of primary cancer resection it should be coidered in the decisionmaking process regarding operation for recurrent disease. Univariate analysis of our data shows that the absolute number of resected metastases is a relevant prognostic factor. Patients with up to two metastases survived significantly longer than all other patients. Until now, only a few researchers analyzed the prognostic impact of the number of metastases being resected during operation. Most reports [7, 8, 13] limit their analysis to subgroups of patients with solitary versus multiple lesio. Our findings support the notion that differentiation into only two groups is not sufficient. Furthermore, our data confirm the observation of other investigators [6 8, 14] who report that repeated curative lung resection in patients with recurrent disease is of benefit to the patient and that survival is not adversely affected by removal of recurrent metastases. Adjuvant treatment regime in this study did not seem to have a significant impact on survival. However, in our study the number of patients allocated into each single adjuvant treatment group was too small to allow for comparison between treatment groups. In summary, the long-term follow-up data after resection for pulmonary renal carcinoma metastases indicate satisfying survival rates and, therefore, support an aggressive surgical approach in this particular metastatic disease. In contradiction to earlier studies that reported no cure after metastasectomy in renal cell carcinoma the present analysis supports the contention that a cure is possible [18]. In view of the low perioperative morbidity and mortality rates and due to the lack of appropriate treatment alternatives we therefore recommend that metastasectomy should be performed whenever possible. In patients with unilateral distribution renal cell carcinoma metastases it appears safe to refrain from bilateral exploration. If the metastatic lesio cannot be completely resected, operation should only be performed in selected patients. Table 2. Following Resection of Lung Metastases of Renal Cell Carcinoma Reference Year No. of Patients Median (mo) 3-Year 5-Year 10-Year Discrimination complete versus incomplete Morrow [2] 1980 30...... 24%... No Jett [13] 1983 44 33 (45%) 27%... No Pogrebniak [19] 1992 23... (60%) (60%)... Yes Cerfolio [7] 1994 96 36 (57%) 36% 14% Yes Fourquier [8] 1997 45 (48) (60%) 44% 4% Yes Kavolius [20] 1998 50...... 54%... Yes Friedel [21] 1999 77 37 (50%) 39% 21% Yes Present series 2002 105 43 54% 40% 33% Yes () as estimated by Kaplan-Meier analysis.

Ann Thorac Surg PILTZ ET AL 2002;73:1082 7 PULMONARY RESECTION OF RENAL CELL CARCINOMA METASTASES 1087 We thank Rolf Weitkunat for his expert support in statistical analysis of our data. Furthermore, we are grateful for the superb editorial assistance of Margy Lynch (MSU, East Laing, MI). References 1. Schildberg FW, Meyer G, Piltz S, Koebe HG. Surgical treatment of tumor metastases: general coideratio and results. Surg Today 1995;25:1 10. 2. Morrow CE, Vassilopoulos PP, Grage TB. Surgical resection for metastatic neoplasms of the lung: experience at the University of Minnesota Hospitals. Cancer 1980;45:2981 5. 3. Kaplan EL, Meier P. Non-parametric estimation from incomplete observatio. J Am Stat Assoc 1958;53:457 81. 4. Mantel N. Evaluation of survival data and two new rank order statistics arising in its coideration. Cancer Chemother Rep 1966;50:163 70. 5. Cox DR. Regression models and life tables. J R Stat Soc B 1972;34:187 220. 6. Pastorino U, Buyse M, Friedel G, et al. Long-term results of lung metastasectomy: prognostic analyses based on 5206 cases. The International Registry of Lung Metastases. J Thorac Cardiovasc Surg 1997;113:37 49. 7. Cerfolio RJ, Allen MS, Deschamps C, et al. Pulmonary resection of metastatic renal cell carcinoma. Ann Thorac Surg 1994;57:339 44. 8. 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;11:17 21. 9. Loehe F, Kobinger S, Hatz R, Helmberger T, Loehrs U, Fuerst H. Value of systematic mediastinal lymph node dissection during pulmonary metastasectomy. Ann Thorac Surg 2001;72:225 9. 10. Ambrogi V, Paci M, Pompeo E, Mineo TC. Traxiphoid video-assisted pulmonary metastasectomy: relevance of helical computed tomography occult lesio. Ann Thorac Surg 2000;70:1847 52. 11. McCormack PM, Bai MS, Begg CB, et al. Role of videoassisted thoracic surgery in the treatment of pulmonary metastases: results of a prospective trial. Ann Thorac Surg 1996;62:213 7. 12. McAfee MK, Allen MS, Trastek VF, Ilstrup DM, Deschamps C, Pairolero PC. Colorectal lung metastases: results of surgical excision [see comments]. Ann Thorac Surg 1992;53: 780 5. 13. Jett JR, Hollinger CG, Zimeister AR, Pairolero PC. Pulmonary resection of metastatic renal cell carcinoma. Chest 1983; 84:442 5. 14. Kandioler D, Kromer E, Tuchler H, et al. Long-term results after repeated surgical removal of pulmonary metastases. Ann Thorac Surg 1998;65:909 12. 15. Maldazys JD, dekernion JB. Prognostic factors in metastatic renal carcinoma. J Urol 1986;136:376 9. 16. Robert JH, Ambrogi V, Mermillod B, Dahabreh D, Goldstraw P. Factors influencing long-term survival after lung metastasectomy. Ann Thorac Surg 1997;63:777 84. 17. Johen JA, Hellsten S. Lymphatogenous spread of renal cell carcinoma: an autopsy study. J Urol 1997;157:450 3. 18. Dineen MK, Pastore RD, Emrich LJ, Huben RP. Results of surgical treatment of renal cell carcinoma with solitary metastasis. J Urol 1988;140:277 9. 19. Pogrebniak HW, Haas G, Linehan WM, Rosenberg SA, Pass HI. Renal cell carcinoma: resection of solitary and multiple metastases. Ann Thorac Surg 1992;54:33 8. 20. Kavolius JP, Mastorakos DP, Pavlovich C, Russo P, Burt ME, Brady MS. Resection of metastatic renal cell carcinoma. J Clin Oncol 1998;16:2261 6. 21. Friedel G, Hurtgen M, Penzetadler M, Kyriss T, Toomes H. Resection of pulmonary metastases from renal cell carcinoma. Anticancer Res 1999;19:1593 6.