T tases in patients with soft-tissue sarcomas is well

Response to Chemotherapy Does Not Predict Survival After Resection of Sarcomatous Pulmonary etastases Louis A. Lanza, D, Joe B. Putnam, Jr, D, Robert S. Benjamin, D, and Jack A. Roth, D Departments of Thoracic Surgery and edicine, The University of Texas D Anderson Cancer Center, Houston, Texas Between and, 6 patients with pulmonary metastases from adult soft-tissue sarcomas were treated with Adriamycin (doxorubicin hydrochloride), Cytoxan (cyclophosphamide), and DTIC before metastasectomy. Thirty-eight thoracotomies were performed with postoperative complications in patients (/,.%) and one postoperative death (/,.6%). Two patients had benign lesions at thoracotomy and were excluded from further survival analysis. The median survival of the remaining patients after thoracotomy was.. months, and the actuarial -year survival was %. ive patients (0, %) achieved a clinically complete response with preoperative chemotherapy, but all had recurrence in the lung and underwent resection of pulmonary metastases. Seven patients (/,%) achieved a partial response and had residual disease resected at thoracotomy. Twelve patients (/,0% showed either no change or disease progression while receiving chemotherapy and were referred for resection. Postthoracotomy disease-free survival and postthoracotomy overall survival did not differ significantly between the three groups. One patient in the group showing no change or progression of disease while receiving chemotherapy is alive without recurrence months after initial pulmonary metastasectomy. Chemotherapy can be used for the initial treatment of pulmonary metastases from adult soft-tissue sarcomas. However, survival after resection of pulmonary metastases cannot be accurately predicted based on the clinical response to preoperative chemotherapy. (Ann Thoruc Surg ;:-) he efficacy of resection of isolated pulmonary metas- T tases in patients with soft-tissue sarcomas is well established [ -. avorable prognostic factors include fewer than five pulmonary nodules on preoperative computed tomographic scans or linear tomograms, a tumor doubling time of greater than 0 days, and a preoperative disease-free interval of greater than months [-. Although combination chemotherapy produces significant responses in many patients with metastatic softtissue sarcoma [&], few patients with metastatic disease achieve long-term survival with chemotherapy alone. Rosenberg and colleagues [ reported improved survival in patients with extremity soft-tissue sarcomas treated with adjuvant Adriamycin (doxorubicin hydrochloride) and Cytoxan (cyclophosphamide) after complete resection of the primary sarcoma. Others [], however, have failed to show a beneficial effect in this setting. In patients with metastatic soft-tissue sarcoma, a favorable response to chemotherapy administered before pulmonary metastasectomy may be predictive of prolonged survival after surgical resection of pulmonary metastatic disease. This study examines survival in patients with sarcomatous pulmonary metastases treated with preoperative chemotherapy followed by pulmonary metastasec- Accepted for publication Oct, 0. Address reprint requests to Dr Putnam, Holcombe Blvd, Box 0, Houston, TX 00-00. tomy. In this select population, the preoperative response to chemotherapy was not predictive of overall survival or disease-free survival after pulmonary resection. aterial and ethods Patient Population Between and, 6 patients were retrospectively identified after treatment at The University of Texas D Anderson Cancer Center. These patients were treated with combination chemotherapy followed by exploration and resection of pulmonary metastases, had a histologically confirmed diagnosis of soft-tissue sarcoma, had the primary tumor controlled, and had metastatic disease that was confined to the thorax and was deemed potentially resectable. All patients were part of a prospective study designed () to evaluate the efficacy of combination chemotherapy with single-dose cyclophosphamide and -day infusion of Adriamycin and dimethyltriazeno-imidazolecarboxamide (DTIC) in the treatment of patients with metastatic or inoperable soft-tissue sarcomas and () to maximize tumor reduction by surgical excision. There were male and female patients with a mean age of 0 years (range, to 6 years). The mean follow-up after initial resection was months (range, to 0 months). Thirteen patients (/6, 0%) had primary tumors of the extremities, patients (/6, %) had 0 by The Society of Thoracic Surgeons 000-//$.0

0 LANZA ET AL SARCOATOUS PULONARY ETASTASES Ann Thorac Surg ;:- truncal sarcomas, and patients (/6, %) had primary sarcomas of the retroperitoneum. All patients received combination chemotherapy preoperatively for two to 6 cycles (mean number, six cycles). Chemotherapy consisted of Adriamycin (60 mg/m infused over 6 hours), Cytoxan (600 mg/m), and DTIC ( g/m over 6 hours). Two patients also received vincristine sulfate. Combination chemotherapy was continued postoperatively unless toxicity was prohibitive. Patients were stratified to one of three response categories according to standard response criteria as judged by the attending physician s evaluation of clinical response at the time of treatment: CR = complete response, no measurable disease; = partial response, greater than 0% reduction in volume of measurable disease; = no change or less than 0% reduction in volume of measurable disease; and = progressive disease. Twelve patients (/6,6%) whose disease either showed no clinically significant change () or progressed () while they were receiving chemotherapy were classified as /. ive patients (/6, %) who achieved a clinically complete response with chemotherapy (not pathologically confirmed) but later had recurrence (RE) and underwent metastasectomy were classified as CR/RE. Seven patients (/6, %) whose disease partially regressed with chemotherapy and who had residual disease resected were classified as. Patients undergoing thoracotomy met previously described criteria []. Briefly, the primary tumor was controlled, extrapulmonary metastases were absent, and the lesions were considered resectable. Thoracic exploration was performed through a median sternotomy or staged bilateral thoracotomies as previously described [. All gross disease was removed when technically possible, sparing maximal pulmonary parenchyma. Sfatisfical Analysis Postoperative survival and disease-free survival were estimated using the product-limit method of Kaplan and eier []. Statistical comparisons between groups were made using the log-rank test with survival estimated from the time of pulmonary metastasectomy. All p values reported are two tailed. Results The 6 patients had a total of thoracotomies. Seventeen patients (/6, 6%) underwent a single exploration, patients (/6, %) underwent two explorations, patient (/6, %) underwent three explorations, and l patient (/6,%) underwent four explorations. Postoperative complications occurred in patients (/,.%): air leaks for longer than days (n = ), bleeding necessitating transfusion (n = l), and postoperative myocardial infarction (n = ). One patient (/,.6%) died days after the fourth thoracotomy for a pulmonary metastasis invading the mediastinum. A portion of esophagus was resected, and the patient subsequently died of sepsis. Postmortem examination was not performed. Chemotherapy-related toxicity necessitating dose reduction devel- Y - z L J 0. p 0.- 0.- n. 0.- 0.- 0. TOTAL AIL 0 6 0 0 0 0 0 0 0 60 0 0 0 ONTHS ig. Overall actuarial survival of all patients after thoracotomy (n = 6). oped in patients (6, %). One patient who received high-dose Adriamycin died (disease free) of congestive heart failure months after thoracotomy. After thoracotomy, the median survival of the entire cohort of 6 patients was f. months, and the actuarial -year survival was.6% (ig ). Two (%) of the 6 patients were found to have fungal lesions (cryptococcosis, ; histoplasmosis, ) at thoracotomy and were excluded from further survival analysis. The remaining patients with histologically confirmed pulmonary metastases had a postthoracotomy median survival of. k. months and an actuarial -year survival of %. Three patients (/,.%) could not be made disease free at operation because of hilar node involvement with sarcoma (n = l), miliary sarcomatous metastases (n = l), or myocardial infarction after the first of staged thoracotomies for bilateral metastases (n = ). None of the patients was alive at months after thoracotomy despite aggressive additional treatment. In contrast, the median survival of the patients rendered disease free was 0 k 6. months, and the actuarial -year survival was % (p = 0) (ig ). ive patients with histologically proven metastatic disease (/, %) achieved a clinically complete response with preoperative chemotherapy alone (CR/RE group) (Tables, ). However, none of these patients were explored to confirm histological response to chemotherapy. They received a mean of cycles (range, to 6 cycles) of preoperative chemotherapy. All patients had recurrence in the lungs to months (mean time, 0. months) after achievement of complete response and

Ann Thorac Surg ::- LANZAETAL SARCOATOUS PULONARY ETASTASES TOTAL AIL 0 6 RESECTED A NOT RESECTED o*'l o 0 0 0 0 0 60 0 ao ONTHS ig. Overall postthoracotomy actuarial survival of all patients with histologically proven pulmonary metastases by resectability at thoracotomy (n = ) (p = 0). were then referred for resection. The mean number of malignant nodules resected was. (range, to ). One patient could not be made disease free at operation because of a postoperative myocardial infarction after the first of staged thoracotomies for bilateral disease. This patient died months after thoracotomy. The postthoracotomy median survival of the remaining patients made disease free at operation was 6. f.6 months, and the postthoracotomy median disease-free survival was. f. months. None of the patients in this group were alive at years after thoracotomy. Three died of progressive pulmonary metastases, and died (clinically disease free) of congestive heart failure at months after thoracotomy. Seven of the patients with histologically proven metastatic disease (/, %) achieved a partial response after preoperative chemotherapy ( group) (see Tables, ). The mean number of preoperative chemotherapy cycles in this group was eight (range, to ). The patients then underwent thoracotomy for resection of residual disease. The mean number of malignant nodules resected was (range, to ). All patients were made disease free at thoracotomy. The postthoracotomy median survival of this group was 0 f. months, and the postthoracotomy median disease-free survival was *.6 months. All patients had recurrence after thoracotomy, and of the died of progressive metastases. One surviving patient remains disease free months after a second thoracotomy to resect recurrent disease. The other surviving patient is alive with documented disease recurrence months after the first thoracotomy. T Twelve patients with histologically proven metastases (/, 0%) experienced either no clinically apparent change (n = ) or disease progression (n = ) while receiving chemotherapy (/ group) (see Tables, ). These patients received a mean of.6 cycles (range, to ) of preoperative chemotherapy. The mean number of malignant nodules resected at thoracotomy in this group was. (range, to ). Two patients could not be made disease free at thoracotomy because of hilar involvement with sarcoma (n = ) or miliary metastases (n = ). These patients died months and months after thoracotomy, respectively, despite additional treatment. The postthoracotomy median survival of the remaining 0 patients was 6. f.6 months, and the postthoracotomy median disease-free survival was 6. f months. A single patient has not had recurrence after the initial thoracotomy and remains disease free at months. A second patient remains alive with disease at months after the initial thoracotomy. All other patients in this group (0/, %) died of progressive metastases. Postthoracotomy disease-free survival did not differ between the three patient groups (p = 0.) (ig ). Similarly, no significant differences were found in postthoracotomy actuarial survival between groups (p = 0.) (ig ) despite complete surgical resection of metastatic disease. Comment Tumor response to preoperative chemotherapy in patients with primary soft-tissue sarcomas may provide a biological assessment of tumor sensitivity to chemotherapeutic agents and indirectly predict survival. However, the clinical response to preoperative chemotherapy for pulmonary metastases from soft-tissue sarcomas in this select patient population was not indicative of survival after thoracotomy. In osteosarcomas, histological evidence of a response to preoperative chemotherapy, measured as percent tumor necrosis, has been found to be a significant prognostic factor [6]. This information has been used to modify the treatment regimens of poorly responding patients [. Similarly, patients with squamous cell carcinoma of the esophagus who show a response to preoperative chemotherapy have been found to have prolonged survival compared with nonresponders or controls in a prospective randomized trial [. Surgical resection of pulmonary metastases in patients with soft-tissue sarcomas provides improved postthoracotomy survival [l,,,. Despite complete resection, the majority of patients with pulmonary metastases from soft-tissue sarcomas will have recurrences in the lung, and many die of progressive pulmonary insufficiency. The development of combination chemotherapy with marked activity against soft-tissue sarcomas has improved overall survival in patients with disseminated disease [0,. The incidence of postoperative complications in this patient population (.% morbidity and.6% mortality) compares favorably with that in another reported series

LANZA ET AL SARCOATOUS PULONARY ETASTASES Ann Thorac Surg ;:- Table. Effect of Preoperative Chemotherapy Response on Postthoracotomy Survival of Patients With Adult Soft-Tissue Sarcomas" No. of Preop Postthoracotomy Postthoracotomy Patient Age Chemotherapy No. of Overall Survival Disease-ree No. (Y) Sex Cycles Response Resectable Nodules (mo) Survival (mo) 6 0 6 0 6 6 6 60 60 0 0 6 66 0 6 0 6 0 CWRE CWRE CR/RE CR/RE CWRE a Parentheses denote censored observation. Second of staged thoracotomies was not done because of a myocardial infarction. Patient is alive with no evidence of disease after second resection. Patient is alive with disease after first resection. Patient is alive with no evidence of disease after first resection. 'ungal disease only was found at thoracotomy. CIURE = complete responselrecurrence; = no change; = partial regression; = progression. No No No 0 Of Of 0 () ()' 0 (Id 6 ()' ()d 0 () 0 () 6 Ob 6 6 0 0 () 0 () [] of similar patients not receiving preoperative chemotherapy. Chemotherapy as described can be safely administered preoperatively in this patient population without significantly increasing perioperative risks. ew reports have examined the effect of preoperative chemotherapy on postoperative survival after resection of sarcomatous pulmonary metastases. Huth and associates [] reviewed the outcome of patients with sarcomas of Table. Effect of Preoperative Chemotherapy Response on Postthoracotomy Survival of Patients With alignant Histology ade Disease ree at Thoracotomy ean No. ean No. Postthoracotomy Postthoracotomy Actuarial of Preop of alignant edian edian Disease- 6-onth No. of Chemotherapy Nodules Survival" ree Survivalb Survival Group Patients Cycles Resected (mo) (mo) (%) Status CRIRE. 6..6.. 0 DWD; DNED..0.6 NED; AWD; DWD / 0.. 6..6 6..0 0 NEDINR'; AWD; DWD a Significance: p = 0.. Significance: p = 0.. This is the only patient alive without recurrence after the initial thoracotomy months ago. AWD = alive with disease; CWRE = komplete responselrecurrence; DNED = dead, no evidence of disease; DWD = dead with disease; NU = no changelprogression; NED = no evidence of disease; NR = no recurrence; = partial regression.

Ann Thorac Surg ;:- LANZAETAL SARCOATOUS PULONARY ETASTASES various histologies who underwent pulmonary metastasectomy including 0 patients with osteosarcoma (0/; 6.%). Twenty-one patients (/; %) received preoperative chemotherapy (histologies not given) and were retrospectively stratified by response. All patients who demonstrated a clinical response were alive without disease a median of months postoperatively. Ten patients had prolongation of the tumor doubling time to greater than 0 days. The median survival of these patients was months. Six patients showed no response to chemotherapy, and their postthoracotomy median survival was only. months. In this study, changes in tumor growth induced by preoperative chemotherapy favorably affected survival after metastasectomy. The histologies of the subset of patients who received preoperative chemotherapy were not stated and may have included histologies other than soft-tissue sarcomas. In addition, the chemotherapy protocol differed substantially from that administered to our patient population. The effects and outcome of these two differences are difficult to determine. Giritsky and co-workers [] examined the effect of preoperative chemotherapy on survival of pediatric patients with metastatic osteogenic sarcoma. These patients received intensive preoperative chemotherapy and preoperative or postoperative pulmonary irradiation before pulmonary metastasectomy. Three-year survival was.% and compares favorably with that of other reported -I a.o TOTAL AIL 0 CR/RE 0. A P R 0. 0. 0.6 0. z 0. P 0. I 0. - 0.. 0 0 / 0 0 0 0 0 0 60 0 0 ONTHS ig. Overall postthoracotomy disease-free survival of all patients with malignant histology who were made to have no evidence of disease at thoracotomy (n = ) (p = 0.). (CWRE = complete response to chemotherapylrecurrence; I = no changelprogression while on chemotherapy; = partial regression with chemotherapy.).o 0. 0. 0. 0.6 z 0. oc 0. 0 0. 0. 0. TOTAL AIL 0 CR/RE A P R 0 0 / 0 0 0 0 0 0 60 ONTHS ig. Overall postthoracotomy actuarial survival of all patients with malignant histology who were made to have no evidence of disease at thoracotomy (n = ) (p = 0.). (CWRE = complete response to chemotherapylrecurrence; I = no changelprogression while on chemotherapy; = partial regression with Chemotherapy.) series. The value of preoperative chemotherapy, pulmonary metastasectomy, and possibly pulmonary irradiation in patients with pulmonary metastases from osteogenic sarcoma is unclear, as the results were based on comparison with historical controls and can be considered only tentative. Belli and colleagues [] observed only one complete response, one partial response, and three disease stabilizations among patients with osteosarcoma metastatic to the lungs who were treated with combination chemotherapy before metastasectomy. All patients had received previous chemotherapy during treatment of the primary tumor. In this review, the clinical response to bulk disease in patients who had received previous chemotherapy was poor. A survival advantage was not observed for the responding patients. In our patients, the response to preoperative chemotherapy was not a reliable predictor of postthoracotomy survival. utational models for the development of drug resistance predict that an inverse relationship exists between curability and tumor burden []. Thus, the earliest possible use of appropriate chemotherapy may have the potential for improving cure rates for solid tumors. ost patients in this series who responded to chemotherapy received prolonged courses of chemotherapy before operation. Surgical intervention was undertaken only when there was evidence of disease recurrence or progression or when a complete response could not be achieved. We

LANZA ET AL SARCOATOUS PULONARY ETASTASES Ann Thorac Surg ;:- are now evaluating a short course of preoperative chemotherapy (two or three cycles) followed by additional chemotherapy after resection. The response to preoperative chemotherapy has proved to be a useful prognostic indicator in other solid tumors [6,. The clinical response to chemotherapy represents complex interactions between malignant cell subsets and cytotoxic agents. Disparate responses have been noted between the primary tumor and its metastases in patients receiving preoperative chemotherapy [6]. Thus, the clinical response to chemotherapy given preoperatively may not reliably predict survival outcome in all histologies, as micrometastases may respond differently than bulk disease. urther studies will be helpful in defining these complex interactions. Two of the 6 patients were found to have unresectable metastases at thoracotomy. Both were in the / gr~up. The usefulness of preoperative chemotherapy in improving resectability or survival could not be determined from our data. In conclusion, preoperative chemotherapy can be used safely in patients with pulmonary metastases from softtissue sarcomas. A high clinical response rate can be achieved in this population using chemotherapy combinations that include Adriamycin, Cytoxan, and DTIC. However, the clinical response to preoperative chemotherapy cannot be used to accurately predict overall survival or disease-free survival after complete surgical resection of metastatic disease. Aggressive surgical resection of metastatic disease in patients with pulmonary metastases from soft-tissue sarcomas can result in longterm survival and should be performed for all patients with potentially resectable pulmonary metastases regardless of response to preoperative chemotherapy. References. Creagan ET, leming TR, Edmonson JH, et al. Pulmonary resection for metastatic non-osteogenic sarcoma. Cancer ;: 0-.. ountain C, curtrey J, Hermes KE. Surgery for pulmonary metastasis: a 0-year experience. Ann Thorac Surg ;:-0.. Rizzoni W, Pass HI, Wesley, et al. ultiple pulmonary metastasectomies in patients with soft tissue sarcoma. Arch Surg 6;:-.. Putnam JB, Roth JA, Wesley N, et al. Analysis of prognostic factors in patients undergoing resection of pulmonary metastases from soft tissue sarcomas. J Thorac Cardiovasc Surg ;60-.. Joseph WL, orton DL, Adhers PC. Prognostic significance of tumor doubling time in evaluating operability in pulmonary metastatic disease. J Thorac Cardiovasc Surg ;6: -. 6. Joseph WL. Criteria for resection of sarcoma metastatic to the lung. Cancer Chemother Rep ;:-.. Jablons D, Sternberg S, Roth JA, et al. etastasectomy for soft tissue sarcoma: further evidence for efficacy and prognostic indicator. J Thorac Cardiovasc Surg ;6-0.. Benjamin RS, Baker LH, Rodriguez V, et al. The chemotherapy of adult soft tissue sarcomas in adults. Proceedings of the st annual clinical conference on cancer, D Anderson Hospital and Tumor Institute, Nov 6. Current concepts in the management of primary bone and soft tissue sarcomas. Chicago: Year Book edical Publishers, :0-.. Bowden EC, Amato D, Enterline HT, et al. Randomized comparison of Adriamycin regimens for treatment of metastatic soft tissue sarcomas [Abstract]. Proceedings of the American Society of Clinical Oncology, :. 0. Pinedo H, Vendrik CPJ, Bramwell VHC, et al. Re-evaluation of the CYVADIC regimen for metastatic soft tissue sarcomas [Abstract]. Proceedings of the American Association of Cancer Research and the American Society of Clinical Oncology, :6.. Benjamin RS, Yap BS. Infusion chemotherapy for soft tissue sarcomas. In: Baker LH, ed. Soft tissue sarcomas. Boston: artinus Nijhoff Publishers, :0-.. Rosenberg SA, Tepper J, Glatstein E, et al. Prospective randomized evaluation of adjuvant chemotherapy in adults with soft tissue sarcomas of the extremities. Cancer ;:.. Edmonson JH, leming TR, hens JC, et al. Randomized study of systemic chemotherapy following complete excision of nonosseous sarcomas. Proceedings of the American Society of Clinical Oncology, :.. Roth JA, Pass HI, Wesley N, White D, Putnam JB, Seipp L. Comparisons of median sternotomy and thoracotomy for resection of pulmonary metastases in patients with adult soft-tissue sarcomas. Ann Thorac Surg 6;:-.. Kaplan EL, eier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc ;:-. 6. Raymond AK, Chawla SP, Carvasco CH, et al. Osteosarcoma chemotherapy effect: a prognostic factor. Semin Diagn Pathol ;:-6.. Winkler K, Beron G, Delleng G, et al. Neoadjuvant chemotherapy of osteosarcoma: results of a randomized cooperative trial (COSS-) with salvage chemotherapy based on histological tumor response. J Clin Oncol ;6:-.. Roth JA, Pass HI, lanagan, et al. Randomized clinical trial of preoperative chemotherapy and postoperative adjuvant chemotherapy with cisplatin, vindesin, and Adriamycin for carcinoma of the esophagus. J Thorac Cardiovasc Surg ;6:-.. artini N, ccormack P, Barnes S, et al. Surgery for solitary and multiple pulmonary metastases. NY State J ed ;:-. 0. Yap B, Baker LH, Sinhovers JG, et al. Cyclophosphamide, vincristine, Adriamycin, and DTIC (CYVADIC) combination chemotherapy for treatment of advanced sarcomas. Cancer Treat Rep 0;6:-.. Blum RH, Carson J, Wilson RE, et al. Successful treatment of metastatic sarcomas with cyclophosphamide, Adriamycin, and DTIC (CAD). Cancer 0;6:-6.. Huth J, Holmes EC, Vernon SE, et al. Pulmonary resection for metastatic sarcoma. Am J Surg 0;0:-6.. Giritsky AS, Etcubanos E, ark JBD. Pulmonary resection in children with metastatic osteogenic sarcoma: improved survival with surgery, chemotherapy, and irradiation. J Thorac Cardiovasc Surg ;:-6.. Belli L, Scholl S, Livartowski A, et al. Resection of pulmonary metastases in osteosarcoma. Cancer ;6:6-0.. Goldie JH. The rationale for the use of preoperative chemotherapy. rog Clin Biol Res ;0:-. 6. Nachman J, Simon A, Dean L, et al. Disparate histologic response in simultaneously resected primary and metastatic osteosarcoma following intravenous neoadjuvant chemotherapy. J Clin Oncol ;:-0.