Surgery for Pulmonary Metastases: A Comprehensive Database
ABSTRACT & COMMENTARY
Many tumors produce pulmonary metastases as the main and even the sole site of dissemination. The technical means for surgical removal of these lesions are widely available, with low morbidity and mortality rates. However, the potential benefits of metastasectomy have not been clearly defined. The procedure is performed routinely in some centers, especially for patients with sarcoma, germ-cell tumors, and pediatric malignancies, but only a few patients with epithelial cancers are considered. Currently, data suggest that metastasectomy produces five-year survivals of 20-40%, which is higher than anticipated for treatment of metastatic disease with chemotherapy or radiation. Unfortunately, the reported literature consists of a small series of highly selected case reports with limited follow-up. This situation clouds the assessment of the use of this procedure.
A cooperative multicenter clinical study was launched in 1990 to overcome the limits of present knowledge. The International Registry of Lung Metastases was set up to: a) establish a common database in major centers of thoracic surgery in the United States and Europe; b) perform homogenous evaluation of results; c) define prognostic factors by multivariate analysis; d) propose a novel system of stage grouping; and e) generate testable hypotheses for future randomized clinical trials.
Pastorino et al report the first analysis of the data collected by the International Registry. The Registry was started with the design of a database at the Instituto Nazionale Tumori of Milan to collect records of patients who had undergone metastasectomy with curative intent. Major centers of thoracic surgery were invited to participate. The data collection process was piloted in Milan and at the Royal Brompton Hospital in London before the other centers were brought on board. Major contributors were Memorial Sloan-Kettering Cancer Center in New York (1075 cases), Hôpital Porte de Choissy, Paris ( 561 cases), the Instituto Nazionale Tumori, Milan (548 cases), and Thoraxchirugie, Gerlingen, Germany (528 cases). Data were collected from 1991 to 1995, but the patient records actually covered four decades of surgical experience.
A total of 5206 patients were available for analysis. There was slight male predominance, and the mean age was 44 years with a range of 2-93 years. Epithelial tumors comprised 43% of the cases, sarcoma 42%, germ cell tumors 7%, melanoma 6%, and 2% various other tumor types. The surgical approach was unilateral thoracotomy in 58%, bilateral thoracotomy in 11%, median sternotomy in 27%, and thoracoscopy in 2%. The resections were generally sublobar, with 67% wedge resections, 9% segmentectomies, 21% lobectomies, and 3% pneumonectomies. Forty-six percent of patients had a single metastasis, and 52% had multiple lesions; 26% had four or more, 9% ten or more, and 3% had 20 or more metastases. One-fifth of the patients required multiple operations including one patient who had seven procedures. The total 30-day perioperative mortality rate was 1.3%. Eighty-eight percent of the patients were able to undergo a complete resection. Ten percent had macroscopic, 2% had microscopic residual disease. Chemotherapy was given to 38% of the patients, before metastasectomy in 22% and after in 16%. Chemotherapy was used more often in patients with multiple metastases or in whom resections were incomplete. The median disease-free interval for all patients was 19 months. Median follow-up was 46 months.
The patients who had complete resection had a 36% survival rate at five years while a 13% five-year survival was observed for patients undergoing incomplete resections. For patients with a disease-free interval (DFI) of 0-11 months, the five-year survival was 33%. For a DFI of 12-35 months, the five-year survival was 31%, and a DFI of 36 months or more meant 45% survival at five years.
Results varied by tumor type. Patients with germ-cell tumors had the best results and enjoyed a 68% five-year survival. Melanoma patients had the worst results with a 21% five-year survival; epithelial tumors and sarcomas had intermediate results with a five-year survival of 37% and 31%, respectively. Recurrences tended to remain pulmonary for sarcomas (66%), whereas melanoma patients had a 73% rate of extrathoracic recurrence. The nature of pulmonary relapses in sarcoma patients allowed these patients to undergo a second (and sometimes third or more) metastasectomy. Results were good in these patients with a 44% five-year survival.
A multivariate analysis of this database showed that DFI, number of metastases, and tumor type were highly significant prognostic variables; DFI longer than 36 months, single metastases, and germ-cell histology were the favorable factors. Germ-cell and Wilms’ tumor had the best results. Breast cancer had the worst prognosis of the epithelial tumors. The multivariate analysis allowed the creation of prognostic groups that were simple, discriminating, and valid in various tumor types. Germ-cell and Wilms’ tumor are excluded because of their uniquely good results. Four groups could be identified:
Group I Resectable, no risk factors (DFI ³ 36 months and single metastasis)
Group II Resectable, one risk factor (DFI < 36 months or multiple metastases)
Group III Resectable, two risk factors (DFI < 36 months and multiple metastases)
Group IV Unresectable
Median survival was 61 months for group I, 34 months for group II, 24 months for group III, and 14 months for group IV.
It was possible to assess the accuracy of clinical staging in 2988 patients. Radiological assessment was accurate in 61%, it underestimated disease in 25%, and overestimated it in 14%. The joint experience indicates that radiologic staging is inaccurate in a high proportion of cases, and thorough intraoperative exploration by an experienced surgeon is required to optimize results. The results indicate that many patients can be treated successfully by surgical removal of pulmonary metastases. Multiple operations may be required to cure patients, and repeat salvage procedures can be safe and effective. Treatment of melanoma and breast cancer, median sternotomy, or bilateral thoracotomies, and the possible contributions of chemotherapy are areas for future clinical trials. (Pastorino U, et al. J Thorac Cardiovasc Surg 1997;113:37-49.)
COMMENTARY
Pulmonary metastases are a common site of recurrence of many malignant neoplasms, and as many as 20% of patients with them have no other sites of disease at autopsy. The number of patients who are potential candidates for metastasectomy appears to be quite large. The prognostic groups defined by the International Registry of Lung Metastases are congruent with the generally accepted recommendations from other studies. However, previous studies were much smaller clinical series that had little more than anecdotal validity. A series of 5206 cases must command our attention. Taken at face value, this series would warrant a change in our thinking about patients with pulmonary recurrences. It is not likely that chemotherapy can produce a 30-40% five-year survival with a single $25,000 treatment in this patient population.
Should we greatly increase the number of patients who undergo pulmonary metastasectomy? A bit of caution seems appropriate. The current series had an excellent perioperative mortality of 1.3%. The procedures were performed in internationally recognized centers of excellence, and it may not be possible to achieve this standard in the community. On the other hand, surgery for bronchogenic cancer and other thoracic procedures are being performed in communities throughout North America. Technical skills to do more of these procedures should be available.
A more serious concern involves patient selection. The multivariate analysis did not find significant discriminating value for patient age. Many patients outside of referral centers will be older than the median of 44 years in this study. It seem reasonable to be concerned about the ability of older patients to tolerate major thoracic surgery. This series included patients up to age 93 years, but the procedure is rarely considered in people over age 60. More patients may permit a better assessment of age as an adverse prognostic factor.
The International Registry included as many sarcoma cases as epithelial cancers. This will surely not be the case in the community. The multivariate analysis suggested that breast cancer is associated with a less satisfactory result with pulmonary metastasectomy. Epithelial malignancies are obviously a heterogenous group of diseases and more experience may be needed to identify candidate diseases for the procedure.
The current report does contain important information. Clearly, patients with germ-cell tumors should be considered for surgery. Sarcoma patients appear to benefit when they fall into one of the favorable groups. Epithelial tumors, with the exception of breast cancer, can be considered. Melanoma remains an unfavorable disease.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content