Ann Thorac Surg 1995;59:1385-1390
© 1995 The Society of Thoracic Surgeons
Metastasectomy for Sarcomatous Pediatric Histologies: Results and Prognostic Factors
Barbara K. Temeck, MD,
Leonard H. Wexler, MD,
Seth M. Steinberg, PhD,
Linda L. McClure, RN,
Marc Horowitz, MD,
Harvey I. Pass, MD
Thoracic Oncology Section, Pediatric Oncology Section, and Biostatistics and Data Management Section, National Cancer Institute, National Institutes of Health, Bethesda, Maryland
 |
Abstract
|
|---|
We reviewed our experience of pediatric metastasectomy to define (1) morbidity/mortality in this population and (2) any preoperative or intraoperative prognostic predictors of survival. One hundred fifty-two patients with median age 19 years (range, 5 to 33 years) had 258 thoracic explorations (Ewing's sarcoma, 28; rhabdomyosarcoma, 6; nonrhabdomyosarcoma soft tissue sarcoma, 42; and osteosarcoma, 76). Resections were accomplished by 218 wedge resections, 19 anatomic resections, 14 wedge and anatomic resections, 4 wedge and chest wall resections, and 3 wedge resections/other procedures. An initial complete resection was accomplished in 121/152 patients (80%). With a median potential follow-up of 10.6 years, median survival from initial thoracotomy is 2.2 years. By the Cox proportional hazards model, three or more positive nodules (p = 0.021), histology other than osteosarcoma (p = 0.0054), and incomplete resection (p < 0.0001) were unfavorable prognostic factors for survival. Two or more positive nodules (p = 0.0049), left location (p = 0.0031), age 14 years or greater at diagnosis (p = 0.0052), or rhabdomyosarcoma (p = 0.0066) predicted shorter pulmonary progression–free survivals after resection. Nonrhabdomyosarcoma pediatric metastasectomy can yield selected long-term survival. Morbidity/mortality is low, and a complete resection, if possible, is paramount. Prognostic factors can be defined that can be used to define the limits of this therapy to the patient and family.
|
Introduction
|
|---|
In young patients with osteogenic sarcoma, rhabdomyosarcoma, and soft tissue sarcomas, the lungs are frequently the initial and only site of metastatic disease. The aggressive resection of these pulmonary metastases has been reported in numerous investigations [1–16] yet analysis of prognostic factors for successful metastasectomy in patients with these ``pediatric histologies'' has not been defined. This study was undertaken to determine the morbidity and mortality, long-term results, and the preoperative and postoperative prognostic predictors of survival in this subset of patients.
|
Patients and Methods
|
|---|
Patient Population
From January 1, 1975, to June 1, 1993, 152 patients were referred from the NCI Pediatric or Surgery Branches to the Thoracic Oncology Section for consideration for pulmonary metastasectomy. In general, these patients were being treated under the auspices of various protocols using chemotherapy with or without radiation for primary head and neck, truncal, or extremity sarcomas. Specifically, the histologic subtype of sarcomas targeted were those commonly encountered in the pediatric population including osteogenic sarcoma, Ewing's sarcoma, rhabdomyosarcoma, and nonrhabdomyosarcoma soft tissue sarcoma (synovial cell sarcoma, malignant fibrous histiocytoma, alveolar soft parts sarcoma, spindle cell aracoma, chondrosarcoma, fibrosarcoma, angiosarcoma, undifferentiated sarcoma, leiomyosarcoma).
These 152 patients had 258 thoracic explorations. There were 91 male and 61 female patients with a median age of 19 years and age range of 5 to 33 years. Seventy-six patients had osteosarcoma, 42 nonrhabdomyosarcoma soft tissue sarcoma, 28 Ewing's sarcoma, and 6 rhabdomyosarcoma. The primary tumor was extremity in 123, truncal in 27, and head/neck in 2 patients. Chemotherapy was part of the treatment regimen during some time in the clinical course of all but 4 patients. Eighteen patients had preoperative symptoms including cough (8), chest pain (5), hemoptysis (3), and dyspnea (2). Preoperative evaluation excluded extrapulmonary metastases in all patients. Pulmonary metastases were evaluated preoperatively by thoracic computed tomography, linear tomography, or both.
Operative Data
Sixty-eight patients had two or more resections, 25 had three or more, 11 had four or more, and 2 had five explorations. Of the 258 thoracic operations, median sternotomy was the operative approach in 133, thoracotomy in 124, and combined sternotomy/thoracotomy in 1. Resection was accomplished by 218 wedge resections, 19 anatomic resections, 14 wedge and anatomic resections, 4 wedge and chest wall resections, and 3 wedge resections with other procedures (2 photodynamic therapy, 1 esophageal resection). The initial operative approach was via median sternotomy in 91 and thoracotomy in 61. Resections in this group were 136 wedge, 7 anatomic, 7 wedge with anatomic, and 2 wedge with chest wall.
Statistical Methods
Survival duration was calculated from the date of the first operation until the date of death. Pulmonary progression–free intervals were calculated as follows: If the patient had a pulmonary relapse after the first operation date, the pulmonary progression–free interval was taken to be the difference between the date of relapse and the date of operation. All patients were followed up for pulmonary progression, including those who could not be rendered disease-free at the initial thoracotomy. If the patient did not have a failure after operation, or if the failure was not pulmonary, then the pulmonary progression–free interval extends from the date of operation until the date of death or last follow-up, and is censored at that point.
The probability of survival or pulmonary progression–free survival was calculated using the Kaplan-Meier method [17], and the significance of the difference between pairs of Kaplan-Meier curves was calculated using the Mantel-Haenszel procedure [18]. The Cox proportional hazards model was used to identify which factors are jointly significant in their association with survival or pulmonary progression–free survival [19]. The resulting model parameters (b1) were converted to relative risks by computing exp(bi) where exp(a) = 2.7183a. The 95% confidence interval for the relative risk was computed as [exp (biL), exp(biH)], where biL = bi - 1.96 [estimated standard error (bi)] and biH = bi + 1.96 [estimated standard error (bi)]. The relative risk indicated the risk associated with dying or experiencing a pulmonary relapse while being in a greater risk category compared with that of being in a lower risk category. All p values are two-sided and denoted by p2.
Prognostic factors to be evaluated by the study included the total number of nodules found, the side(s) of the nodules, the number of positive nodules found, the side(s) of the positive nodules, and the maximum dimension of any positive nodules found, as well as the patients' sex, age, diagnosis, disease-free interval (initial), previous pulmonary progression–free interval duration, and whether the operation was complete or incomplete. The initial disease-free interval was calculated as zero if the patient presented with lung metastases at diagnosis (synchronous) and as the difference between the date of initial pulmonary failure and the date of diagnosis for patients without lung metastases at diagnosis (metachronous).
|
Results
|
|---|
Operative Results
An initial complete resection was accomplished in 121/152 patients (80%). Mortality due to respiratory insufficiency occurred 27 and 48 days postoperatively in 2 patients (1.3%) having incomplete resections. These were the second and fourth metastasectomies in these patients. There were ten complications after initial metastasectomy (6.5%), which included prolonged air leak (3), pneumonia (2), superficial wound infections (2), respiratory insufficiency (1), pneumonitis (1) and atrial tear (1). In the 106 subsequent repeat explorations, there were six complications (5.6%): air leak (3), pneumonia (2), and superficial wound infection (1).
Survival and Recurrence Results
The median potential follow-up was 10.6 years. For all 152 patients the median survival from the initial operation was 2.2 years (Fig 1
) and initial pulmonary progression–free survival was 0.6 years (Fig 2). Median survival was 3.1 years for osteosarcoma (Fig 3), 2.3 years for nonrhabdomyosarcoma soft tissue sarcoma, 1.7 years for Ewing's sarcoma, and 0.4 years for rhabdomyosarcoma. The inability to accomplish a complete resection was associated with significantly shortened survival (Fig 4), and younger patients had a longer progression-free survival (Fig 5).
View larger version (14K):
[in this window]
[in a new window]
|
Fig 1. . Overall survival for 152 patients with pediatric histologies from the initial pulmonary metastasectomy.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig 2. . Pulmonary progression–free survival available in 149 patients with pediatric histologies (see text for details).
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig 3. . Survival of patients having pulmonary metastasectomy according to their histology. Patients with osteogenic sarcoma had significantly longer survivals compared with Ewing's sarcoma (p2 = 0.044) and soft tissue sarcoma (p2 = 0.044).
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig 4. . Survival of patients having a complete or incomplete removal of all metastases. Patients having complete resection survived significantly longer than with incomplete resection (p2 = 0.0001).
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig 5. . Progression-free interval as a function of age. Younger patients had longer progression-free intervals compared with older patients (p2 = 0.025).
|
|
Sex, age, and initial disease-free interval were not predictive for survival. By the Cox proportional hazards model, the presence of three or more metastases at operation (p = 0.021), histology other than osteosarcoma (p = 0.0054), and incomplete resection (p < 0.001) were unfavorable prognostic factors for survival. Two or more positive nodules (p = 0.0049), left location of the nodules (p = 0.0031), age 14 years or greater at diagnosis (p = 0.0052) and rhabdomyosarcoma (p = 0.0066), predicted a shorter pulmonary progression–free survival after resection in the 149 patients in whom these data were available. These results are outlined in Tables 1 and 2.
View this table:
[in this window]
[in a new window]
|
Table 2. . Factors Predictive of Shortened Pulmonary Progression–Free Interval: Two Cox Proportional Hazards Models
|
|
|
Comment
|
|---|
Although there is a wealth of information regarding the surgical management of pulmonary metastases in the adult population with regard to mortality/morbidity as well as the attempted definition of prognostic factors for survival, there is only a small database that relates this information to the patients afflicted with ``pediatric'' histologies. There are a number of issues that make these data difficult to find, specifically the rarity of these neoplasms and the inability of many institutions to accumulate a large series of patients with more than just one of these subtypes. Moreover, the criteria for resection of these tumors, except for osteogenic sarcoma, remain undefined in the adolescent population. Further, the definition of which population of patients to analyze can become problematic. Does one define a given age as the ``pediatric population'' or does one group these patients by the histologic cell type, acknowledging that the adult type soft tissue sarcomas behave differently than soft tissue sarcomas that are seen mainly in children but that can be diagnosed in older patients, ie, those in their 30s? Our institution recognizes the latter situation, and therefore these analyses do include a select number of patients who could not be categorized as children, but who were diagnosed by immunohistochemical and electron microscopic means to have tumors that usually would be seen in the pediatric age group, hence termed ``pediatric histologies.'' These include osteogenic sarcoma, Ewing's sarcoma, rhabdomyosarcoma, and selected nonrhabdomyosarcomas that were treated under the auspices of intensive chemoradiation protocols by our Pediatric Branch.
Our approach to these patients has been consistent with our previous strategies in the management of adult soft tissue sarcoma [7, 20]. For initial and the first few reexplorations, median sternotomy has been used to permit bilateral exploration and resection. This approach consistently allows for low operative mortality and morbidity, minimum hospital stay, and relatively ``easy to deal with'' reversible complications. The operative mortality of 1.3% is very compatible with that in the literature, and the only two perioperative deaths occurred in patients after reexplorations. Therefore, in patients having multiple explorations in whom a number of wedge resections may have left them with decreased pulmonary reserve, preoperative pulmonary function tests should be performed. Because these patients also have been treated with chemotherapies known to have pulmonary toxicity, anesthetic management must be compulsive with respect to fluid resuscitation and oxygen concentrations.
This analysis of prognostic factors for pediatric metastasectomy was twofold. We attempted to define those factors that would predict shorter overall survival after metastasectomy as well as those that foretold a greater chance of subsequent pulmonary recurrence (the pulmonary progression–free interval). We made use of prognostic factors that had been investigated in the pulmonary metastatic series of adults in the literature and factors from our own adult database. These included age, sex, number of nodules seen on preoperative studies, laterality, size of resected nodules, and disease-free interval, as well as the number of nodules revealing malignancy resected, the completeness of resection, and the histology of the disease studied. Because virtually 100% of these patients received adjuvant chemotherapy and radiation therapy, we did not analyze the impact of these parameters on survival. Moreover, to do so would have led to multiple, small group subset analyses because many primary and salvage regimens were used.
For the group of 152 patients discussed, age, sex, and size of positive nodules did not contribute significantly to survival. Previous data from our institution chiefly in adult patients with osteogenic sarcoma revealed no effect of age and sex [6]; however, women with soft tissue sarcoma had a lower risk of dying [7]. Although improved survival has been associated with various numbers of nodules (usually fewer than five) on computed or linear tomography in osteogenic or soft tissue sarcoma [2, 6, 7, 8, 21] no relationship between preoperative nodule number and survival was detected in this study. Survival was affected adversely by left location of the nodules, but laterality has not influenced prognosis in other reports [2, 6]. Although selected papers have reported lesions greater than 2 cm [9] or 3 cm [22] as having survival impact (chiefly in adult patients), no size association was seen with survival in this study. We found no association between survival and the interval from primary resection or treatment to the development of pulmonary metastases (disease-free interval). Although two studies [23, 24] found an association between survival and disease-free interval in osteogenic sarcoma, this was not evident in others [3, 4, 21, 25–28]. Conflicting data also have been reported in soft tissue sarcoma, where disease-free interval was related to survival by several authors [6, 7, 24] but not by others [4, 11, 26]. In a recent report, disease-free interval did not change outcome in a series of pediatric cases that included osteogenic sarcoma and soft tissue sarcomas [12].
The number of nodules found to have tumor has been found to affect survival in several reports. For osteogenic sarcoma, differing reports have documented better survival after metastasectomy for one versus two or more resected [25], two or fewer [29], and four or fewer [6]. Fewer than four resected metastases had a favorable impact for Ewing's sarcoma [30]. Whereas survival was prolonged for patients with two or fewer resected pulmonary metastases from soft tissue sarcoma [8], other studies have detected no difference for patients with four or fewer versus five or more [7] and even a resection benefit with 15 or fewer resected nodules [6]. In adult patients whose soft tissue sarcoma metastases could be resected completely, the number of nodules did not affect survival in Jablons and associates' [7] and Robinson and colleagues' [13] series. In a similar fashion, Girard and associates [14] found no association between nodule number and survival in osteogenic and soft tissue sarcoma. Both Heij and colleagues [12] (in a study of pediatric patients with osteogenic sarcoma, nephroblastoma, Ewing's sarcoma, and various other tumors) and van Geel and coworkers [11] in soft tissue sarcoma patients reported a nonsignificant trend toward a worse prognosis as the number of resected nodules increased. Our data suggest that in the patients with pediatric histology metastasectomy, three or more nodules positive for tumor at the initial thoracotomy negatively affected survival. Probably the most important prognostic predictor of survival was the ability to achieve complete resection of all nodules, and this concurs with other reports in the literature, including those that concentrate primarily on pediatric neoplasms [12]. Survival also was influenced by histology of the primary neoplasm, with osteogenic sarcoma being the most favorable and rhabdomyosarcoma being associated with a poor outcome. Moreover, the 5-year survival rate of 45% to 50% seen in this series compares favorably with that in the literature [31] and would seem to be stable at least out to 10 years.
Besides the number of patients, substantial long-term follow-up, and multiple histologies involved, this series is also unusual in its analysis of the prognostic factors associated with pulmonary progression free survival. A shortened pulmonary progression–free survival was noted in patients with rhabdomyosarcoma histology and those who had previously been found to have two or more positive nodules at thoracotomy. Additional factors that predicted shorter pulmonary progression–free survival were left location of nodules, age 14 years or greater at diagnosis, and multiple nodules on computed tomographic scan for patients having computed tomographic scan. These data, if substantiated in larger numbers of patients, potentially could predict those patients with the highest chance of early recurrence. It is that population of patients that may benefit the most from the addition of an effective adjuvant regimen after thoracotomy, if one becomes available.
As with all publications of this type, this study is limited due to data based on retrospective analyses. Prospective, randomized trials and improvements in adjuvant modalities as well as refinements in diagnostic and surgical techniques are required to further affect survival results. Larger numbers of patients subjected to similar analyses are necessary to define clearly patients who can be targeted for benefit from metastasectomy. It is only now that an international consortium is being organized to establish a database to analyze thousands of patients undergoing metastasectomy from 15 institutions. Until such data are available no single factor should be used to exclude an individual patient from resection. However, this information can be used to explain the limits of this therapy to the patient and family.
|
Footnotes
|
|---|
Presented at the Thirty-first Annual Meeting of The Society of Thoracic Surgeons, Palm Springs, CA, Jan 30–Feb 1, 1995.
Address reprint requests to Dr Pass, Thoracic Oncology Section, Surgery Branch, National Cancer Institute/NIH Building 10, Rm 2B07, Bethesda, MD 20892.
|
References
|
|---|
- Martini N, Huvos AG, Mike V, Marcove RC, Beattie EJ Jr. Multiple pulmonary resections in the treatment of osteogenic sarcoma. Ann Thorac Surg 1971;12:271–80.[Medline]
- Putnam JB Jr, Roth JA, Wesley MN, Johnston MR, Rosenberg SA. Survival following aggressive resection of pulmonary metastases from osteogenic sarcoma: analysis of prognostic factors. Ann Thorac Surg 1983;36:516–23.[Abstract]
- Meyer WH, Schell MJ, Jumar AP, et al. Thoracotomy for pulmonary metastatic osteosarcoma: an analysis of prognostic indicators of survival. Cancer 1987;59:374–9.[Medline]
- Flye MW, Woltering G, Rosenberg JA. Aggressive pulmonary resection for metastatic osteogenic and soft tissue sarcomas. Ann Thorac Surg 1984;37:123–7.[Abstract]
- Pastorino U, Gasparini M, Valente M, et al. Primary childhood osteosarcoma: the role of salvage surgery. Ann Oncol 1992;3:543–6.
- Putnam JB, Roth JA, Wesley MN, Johnston MR, Rosenberg SA. Analysis of prognostic factors in patients undergoing resection of pulmonary metastases from soft tissue sarcomas. J Thorac Cardiovasc Surg 1984;87:260–8.[Abstract]
- Jablons DM, Steinberg SM, Roth J, et al. Metastasectomy for soft tissue sarcoma: further evidence for efficacy and prognostic indicators. J Thorac Cardiovasc Surg 1989;97:695–705.[Abstract]
- Casson AG, Putnam JB, Natarajan G, et al. Five year survival after pulmonary metastasectomy for adult soft tissue sarcoma. Cancer 1992;69:662–8.[Medline]
- Saltzman DA, Synder CL, Ferrell KL, Thompson RC, Leonard AS. Aggressive metastasectomy for pulmonary sarcomatous metastases: a follow-up study. Am J Surg 1993;166:543–7.[Medline]
- Gadd MA, Casper ES, Woodruff JM, McCormack PM, Brennan MF. Development and treatment of pulmonary metastases in adult patients with extremity soft tissue sarcoma. Ann Surg 1993;218:705–12.[Medline]
- Van Geel AN, Van Coevorden F, Blankensteign JD, et al. Surgical treatment of pulmonary metastases from soft tissue sarcoma: a retrospective study in the Netherlands. J Surg Oncol 1994;56:172–7.[Medline]
- Heij HA, Vos A, Dekraker J, Voûte PA. Prognostic factors in surgery for pulmonary metastases in children. Surgery 1994;115:687–93.[Medline]
- Robinson MH, Sheppard M, Moskovic E, Fisher C. Lung metastasectomy in patients with soft tissue sarcoma. Br J Radiol 1994;67:129–35.[Abstract/Free Full Text]
- Girard P, Baldeyrou P, LeChevalier T, et al. Surgical resection of pulmonary metastases up to what number? Am J Respir Crit Care Med 1994;149:469–76.[Abstract]
- Matthay RA, Arroliga AC. Resection of pulmonary metastases. Am Rev Respir Dis 1993;148:1691–6.[Medline]
- Martini N, McCormack PM. Pulmonary resection in sarcoma metastases. Dev Oncol 1988;55:197–200.
- Kaplan EL, Meier P. Non-parametric estimation from incomplete observation. J Am Stat Assoc 1958;53:457–81.
- Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chem Rep 1966;50:163–70.[Medline]
- Cox D. Regression models and lifetables. J R Stat Soc 1972;34:187–202.
- Roth JA, Pass HI, Wesley MN, White D, Putnam JB, Seipp C. Comparison of median sternotomy and thoracotomy for resection of pulmonary metastases in patients with adult soft tissue sarcomas. Ann Thorac Surg 1986;42:134–8.[Abstract]
- Roth JA, Putnam JB, Wesley MN, Rosenberg SA. Differing determinants of prognosis following resection of pulmonary metastases from osteogenic and soft tissue sarcoma patients. Cancer 1985;55:1361–6.[Medline]
- Vogt-Moykopf I, Bulzebruck H, Merkle NM, Probst G. Results of surgical treatment of pulmonary metastases. Eur J Cardiothorac Surg 1988;2:224–32.[Abstract]
- Telander RL, Pairolero PC, Pritchard DJ, Sim FH, Gilchrist GS. Resection of pulmonary metastatic osteogenic sarcoma in children. Surgery 1978;84:335–41.[Medline]
- Lienard D, Rocmans P, Lejeune FJ. Resection of lung metastases from sarcomas [Abstract]. Eur J Surg Oncol 1989;15:530–4.[Medline]
- Goorin AM, Delorey MJ, Lack EE, et al. Prognostic significance of complete surgical resection of pulmonary metastases in patients with osteogenic sarcoma: analysis of 32 patients. J Clin Oncol 1984;2:425–31.[Abstract]
- Pastorino U, Valente M, Gasparini M, et al. Median sternotomy and multiple lung resections for metastatic sarcomas. Eur J Cardiothorac Surg 1990;4:477–81.[Abstract]
- Tabone MD, Kalifa C, Rodary C, Raquin M, Couanet D. Osteosarcoma recurrences in pediatric patients previously treated with intensive chemotherapy. Proc Annu Meet Am Soc Clin Oncol 1994;13:A1682.
- Carter SR, Grimer RJ, Sneath RS, Matthews HR. Results of thoracotomy in osteogenic sarcoma with pulmonary metastases. Thorax 1991;46:727–31.[Abstract/Free Full Text]
- Spanos PK, Payne WS, Ivins JC, Pritchard DJ. Pulmonary resection for metastatic osteogenic sarcoma. J Bone Joint Surg 1976;58A:624–8.
- Lanza LA, Miser JS, Pass HI, Roth JA. The role of resection in the treatment of pulmonary metastases from Ewing's sarcoma. J Thorac Cardiovasc Surg 1987;94:181–7.[Abstract]
- Pass HI. Treatment of metastatic cancer to the lung. In: DeVita V, Hellman S, Rosenberg A, eds. Principles and practice of oncology. Philadelphia: Lippincott, 1993:2186.
This article has been cited by other articles:
|
|
|
|
 
F. Chen, R. Miyahara, T. Bando, K. Okubo, K. Watanabe, T. Nakayama, J. Toguchida, and H. Date
Prognostic factors of pulmonary metastasectomy for osteosarcomas of the extremities
Eur. J. Cardiothorac. Surg.,
December 1, 2008;
34(6):
1235 - 1239.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Kempf-Bielack, S. S. Bielack, H. Jurgens, D. Branscheid, W. E. Berdel, G. U. Exner, U. Gobel, K. Helmke, G. Jundt, H. Kabisch, et al.
Osteosarcoma Relapse After Combined Modality Therapy: An Analysis of Unselected Patients in the Cooperative Osteosarcoma Study Group (COSS)
J. Clin. Oncol.,
January 20, 2005;
23(3):
559 - 568.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. P. Staddon, R. Lackman, K. Robinson, J. B. Shrager, and M. Warhol
Osteogenic Sarcoma Presenting with Lung Metastasis
Oncologist,
April 1, 2002;
7(2):
144 - 153.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. K. Temeck, L. H. Wexler, S. M. Steinberg, L. L. McClure, M. A. Horowitz, and H. I. Pass
Reoperative pulmonary metastasectomy for sarcomatous pediatric histologies
Ann. Thorac. Surg.,
September 1, 1998;
66(3):
908 - 913.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
