General Thoracic Surgery (General Thoracic Surgery (Shields)) [2 VOLUME SET]

Editors: Shields, Thomas W.; LoCicero, Joseph; Ponn, Ronald B.; Rusch, Valerie W.

Title: General Thoracic Surgery, 6th Edition

Copyright 2005 Lippincott Williams & Wilkins

> Table of Contents > Volume II > Section XVII - Other Tumors of the Lung > Chapter 120 - Secondary Tumors of the Lung

Chapter 120

Secondary Tumors of the Lung

Joe B. Putnam Jr.

Pulmonary metastases represent a particular manifestation of systemic metastases from primary malignant tumors. Although primary tumors can be locally controlled with surgery or irradiation, the selection of therapy for systemic metastases requires systemic-type therapies such as chemotherapy or other, targeted therapy. Metastases are commonly treated with chemotherapy as an initial modality. Local control modalities such as radiation therapy, or even surgery, may be used to treat or palliate the local symptoms resulting from metastases, particularly bony metastasis causing pain. Although metastases often represent systemic and uncontrolled tumor growth, which may herald rapid disease progression, patients with metastases isolated within the lung may have a more favorable tumor biology. These patients are more amenable to local treatment options, or combinations of local and systemic treatment options, than are patients with multiorgan metastases. Isolated pulmonary metastases should not be viewed as untreatable. Patients who have complete resection of all metastases have a longer survival than patients who have pulmonary metastases left behind. Long-term survival, greater than 5 years, may be expected in about 30% of all patients with resectable pulmonary metastases.

Most patients with metastatic disease to the lung, however, have unresectable metastases. Only a small minority of these patients are amenable to complete resection of pulmonary metastasis (i.e., physical removal or ablation of all abnormalities that can be visualized or palpated). Selection of therapy for patients with pulmonary metastases isolated to the lung requires multidisciplinary evaluation in all but the simplest of situations. Isolated solitary metastasis with a prolonged disease-free interval (greater than 12 months) may be resected with good results. Combinations of chemotherapy and surgery may be considered and may offer more patients the potential for optimal local and systemic control of their disease process. Enhancement of survival will require improved local control, systemic therapies, or regional drug delivery to the lungs.

HISTORICAL PERSPECTIVE

Early attempts of resection of pulmonary metastases have been described by Meade (1961) and Martini and McCormack (1998). Weinlechner (1882) and Kronlein (1884) reported resection of pulmonary metastasis (as an incidental procedure) while resecting a primary chest wall tumor. Resection of a pulmonary metastasis as a planned procedure was described by Divis (1927) and Torek (1930). Barney and Churchill (1939) reported one of the first long-term survivors of pulmonary metastasectomy after resection of a metastasis from a patient with hypernephroma (metastatic renal cell carcinoma). After nephrectomy for local control of the primary tumor, the patient underwent resection of the metastasis. The patient survived for 23 years and died from unrelated causes. Alexander and Haight (1947) reviewed the first large series (25 patients) of patients who had resection of metastases from carcinoma and sarcoma. They concluded that patients who would withstand the resection and in whom no other metastases were evident could should undergo resection. Mannix (1953) described for the first time resection of multiple pulmonary metastases from a patient with osteochondroma of the tibia. Only one nodule was identified on the preoperative chest radiograph. Few attempts were made at multiple or repetitive resections for pulmonary metastases until Martini and associates (1971) described the value of resecting multiple metastases and the associated survival advantage of multiple resections (multiple sequential operations) for treatment of osteogenic sarcoma. Selection criteria have been proposed by many; however, the author (JBP) and Roth (1990) noted that unresectability may be the sole exclusion criterion. Selection for resection is otherwise subjective and individualized to the patient. In the past 20 years, resection of solitary and multiple pulmonary metastases from numerous primary neoplasms have been performed with good long-term survival in 20% to 40% of patients, as shown by Pastorino and colleagues (1997).

P.1832

Autopsy studies have demonstrated that about one third of patients with cancer die with pulmonary metastases and that a small percentage die with metastases confined solely to the lungs. Metastases from osteogenic and soft tissue sarcomas commonly occur only in the lungs, as shown by Potter and colleagues (1985). Less commonly, patients with other solid organ neoplasms from melanoma, breast, or colon have isolated pulmonary metastases, but these metastases may represent favorable tumor biology and a treatable subset of such patients. In the absence of extrathoracic metastases, patients with isolated and resectable pulmonary metastases should undergo complete resection for treatment, prolonged survival, and possible cure of their disease. Even in the presence of extrathoracic metastases, selected individual patients with complete resection may have a survival advantage. Limitations on the number of metastases resected with benefit have not as yet been determined; however, the greater the number of radiographically identified metastases before resection, or the larger the number palpated in the operating room, the greater likelihood that micrometastases exist and that the lesions are unresectable. Early, and potentially aggressive, recurrence is likely. Multidisciplinary evaluation and selection of effective systemic therapy may theoretically treat micrometastatic disease, thus increasing the postdiagnosis of metastases survival beyond that of immediate resection. Still, local control of rapidly enlarging solitary metastasis may be needed. Progression of multiple metastases will inexorably reduce pulmonary reserve.

PATHOLOGY

Malignant tumors may metastasize by hematogenous, lymphatic, direct invasion, and aerogenous routes. Underlying tumor biology and host resistance determine mechanisms of spread, locations of metastases, and extent of growth. Hematogenous metastases are most frequently found in the capillary beds of the lung, liver, brain, and bone. Clumps of

P.1833

SYMPTOMS

Symptoms rarely occur from pulmonary metastases. Therefore, diagnosis of metastases is routinely made on radiographic imaging after primary tumor resection. Palliation for pain is rarely needed because the parietal pleura is infrequently involved by parenchymal metastases. A distinction must be made between pleural-based and parenchymal-based metastases before resection. Few (<5%) patients with metastases present with symptoms of dyspnea, pain, cough, or hemoptysis. Bocklage and colleagues (2001) noted that patients with metastases from angiosarcoma would present with these symptoms lasting from a few weeks to months. Rarely, patients with peripheral sarcomatous metastases may develop pneumothorax from disruption of the peripheral pulmonary parenchyma. Srinivas and Varadhachary (2000) suggested that patients with a primary malignancy and a pneumothorax should be evaluated for lung metastases.

DIAGNOSIS

Pulmonary metastases may appear as solitary or multiple nodules and as well-circumscribed or diffuse opacities, and they may be miliary or massive in appearance, as described by Snyder and Pugatch (1998). Still, the radiographic findings of patients with pulmonary nodules may be nonspecific and represent a wide spectrum of benign or malignant processes. The surgeon must consider other, more locally related diagnoses such as histoplasmosis, tuberculosis, or other malignant diseases such as lung cancer in these patients. There are no pathognomonic radiographic criteria for metastatic disease. Small, well-circumscribed peripheral nodules in the patient with a known primary malignancy may very well represent metastatic disease.

Chest radiographs are commonly obtained after primary tumor resection because they may demonstrate pulmonary parenchymal changes consistent with metastases (Fig. 120-2A). Routine chest radiographs represent an effective means of screening patients for pulmonary metastases. The chest radiograph itself is an effective screening tool. Fleming and colleagues (2001) found that less than 1% of patients with T1 primary extremity soft tissue sarcomas had pulmonary metastases detectable on chest radiographs [with selective use of chest computed tomography (CT)]. These authors found that routine chest CT for all patients with extremity T1 soft tissue sarcoma was not an efficient means of detecting occult metastases in this patient population. In one study, Ren and colleagues (1989) noted that chest radiographs identified only 48% of patients with metastases. Lien and co-workers (1988) showed that about half of patients with nonseminomatous testicular tumors have negative chest radiographs but abnormalities identified on CT scans.

Patients with known metastases on chest radiographs should undergo CT to identify the precise location of the known metastases and to identify other smaller, potentially occult metastases (Fig. 120-2A, B, C). CT will demonstrate nodules as small as 2 to 4 mm. When clinically correlated with the patient's age, prior history of malignancy, and prior treatment, a clinical diagnosis of pulmonary metastases can be made.

Patients without evidence of metastases on chest radiograph may have metastases demonstrated by CT. Today's high-resolution CT of the chest may achieve resolution of pulmonary abnormalities of 2 to 3 mm in diameter. Metastases may appear at this size, but more commonly, sequelae of infections such as granulomas or other pulmonary parenchymal changes may produce these small indeterminate lesions. In specific parts of the country, granulomatous disease from histoplasmosis is prevalent, and clinical correlation with the radiographic size and number of the lesions, location, physical and radiographic characteristics, and character must be considered. Resection may provide the most direct way to evaluate histology; however, a benign etiology is more common in the general population. Margaritora and associates (2002) noted that helical computed tomography (HCT) of the chest was more sensitive than high-resolution CT of the chest (81% vs. 75%). Sensitivity for lesions less than 6 mm in diameter was 48% to 62%. Margaritora and colleagues (2002) noted that CT of the chest only achieved a sensitivity of 48% for lesions less than 6 mm. When exploring patients with such findings, the surgeon must be prepared to palpate these small lesions carefully, which may lie deep within the lung parenchyma. The authors also noted that HCT was more sensitive than high-resolution CT. Commonly, indeterminate lesions are followed for changes on sequential CT scans. If these

P.1834

Magnetic resonance (MR) imaging may be as sensitive as CT scans for identifying pulmonary metastases but adds little additional information, as observed by Feuerstein (1992) and Wyttenbach (1998) and their associates. The resolution of an MR image for pulmonary metastases is not as sharp as CT of the chest. A short time inversion recovery sequence provided the best sensitivity for individual nodules (82%). MR imaging is not routinely recommended for evaluation of patients with pulmonary metastases limited to the pulmonary parenchyma, although newer techniques, such as three-dimensional (3D) volumetric interpolated breath-held whole-body MR imaging, as recommended by Lauenstein and colleagues (2002), may be valuable in selected patients. Walker and associates (2000) have used MR imaging as a screening tool with success for extrapulmonary metastases in patients with breast carcinoma. MR

P.1835

Benign granulomatous diseases may mimic metastases; however, in patients with a prior diagnosis of malignancy, new and multiple nodules are most likely metastases. Fine-needle aspiration or thoracoscopic wedge excision may be helpful for diagnosis or staging of pulmonary nodules in high-risk patients. Clinical stage I or II primary lung carcinoma may be indistinguishable from a solitary metastasis particularly if the original tumor was squamous cell carcinoma or adenocarcinoma. For these two histologies specifically, or in patients in whom a primary non small cell carcinoma (NSCLC) of the lung cannot be excluded, lobectomy and a systematic mediastinal lymph node dissection would be a procedure of choice. In patients with lymphangitic spread of cancer and dyspnea, biopsy may be required to differentiate neoplasm from infection.

¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET) can be used to identify patients with metastases. Its value may lie in a negative study as well as identifying lesions, which are positive, or FDG avid. Lucas and co-workers (1998) evaluated the results of FDG PET and chest CT in 62 patients who had been treated for pulmonary metastases (mean age, 51 years) with 15 types of soft tissue sarcoma. Local recurrence, distant recurrence, and pulmonary metastases were evaluated. The mean follow-up was 3 years. For local disease, FDG PET was 73.7% sensitive and 94.3% specific (14 true positive, 5 false negative). MR imaging was 88.2% sensitive and 96.0% specific. When FDG PET was used to identify lung metastases in 70 comparisons, sensitivity was 86.7%, and specificity was 100% (13 true positive, 2 false negative). CT of the chest had 100% sensitivity and 96.4% specificity. Other metastases (13 patients) were identified by FDG PET. The authors concluded that FDG PET could identify local and distant recurrence of tumor and other metastases and recommended that all three methods be used in a complementary fashion to identify the extent of disease initially and during follow-up. Franzius and colleagues (2001) compared FDG PET to helical chest CT to detect pulmonary metastases arising from malignant bone tumors. FDG PET had a sensitivity of 0.50, a specificity of 0.98, and an accuracy of 0.87, compared with spiral chest CT of 0.75, 1.00, and 0.94, respectively. The authors concluded that helical chest CT is superior to FDG PET in detecting pulmonary metastases from primary bone tumors. Hung and colleagues (2001) noted that the use of FDG PET for patients with a current cancer provided good information for regional as well as extraregional metastases. This observation was confirmed by Lonneux and colleagues (2002), who observed that whole-body FDG PET was superior to conventional imaging modalities in patients being evaluated for recurrent colorectal carcinoma, or recurrent breast carcinoma as found by Siggelkow and co-workers (2003). Veronesi and associates (2002) noted that glucose uptake (by FDG PET) and angiogenesis were independent biological features in patients with pulmonary metastasis from the various neoplasms and may suggest future antiangiogenic therapies.

The surgeon must select the radiographic imaging or scanning techniques that will provide the necessary, and complete, clinical information required for treatment-planning decisions. Woodard and colleagues (1998) suggested several factors that may influence the surgeon's choice of radiographic studies. These factors include: (a) identifying the size, location, and characteristics of pulmonary nodules or metastases, (b) characterizing the solitary squamous cell carcinoma (or adenocarcinoma) metastasis from a primary NSCLC, (c) evaluating for extrathoracic metastatic disease (other sites of hematogenous spread, metastasis to regional lymph nodes, or other tumors), and (d) evaluating the potential for local invasion.

Metastasis or Primary Bronchial Carcinoma

Pulmonary metastases from sarcomas or other distinctive nonpulmonary neoplasms are easy to diagnose. Solitary carcinomatous metastasis from breast or colon and squamous cell carcinoma metastasis from head and neck primary tumors are more difficult to distinguish from primary lung carcinoma. Patients with two or more pulmonary nodules can be considered to have metastases. Treatment may be similar. In tumors without a propensity for bilaterality (e.g., nonsarcomatous histology), a unilateral approach may be optimal.

Traditionally, a comparison of the primary neoplasm and the lung nodule using light microscopy has been the only method for determining origin of the lung nodule or neoplasm. Electron microscopy, as studied by Herrera and associates (1985), or specific molecular or genetic characteristics may identify more precisely the origin of these neoplasms. Monoclonal antibodies may assist in discriminating between primary bronchial adenocarcinoma and colon carcinoma metastatic to the lung, as described by Ghoneim and colleagues (1990). Amplified K-rasoncogene expression in a pulmonary metastasis from colon adenocarcinoma primary was noted by Slebos and co-workers (1991) and was also present in the primary tumor. A monoclonal antibody to identify colorectal carcinoma has been used by Flint and Lloyd (1992a, 1992b) in 46 patients. Cytology samples from patients with metastatic colon carcinoma and primary lung adenocarcinoma were examined. Unfortunately, the monoclonal antibody was not effctive in discriminating primary lung cancer from metastatic adenocarcinoma. Flow cytometry and DNA analysis have been used by Nomori (1991) and Salvati (1989) and their colleagues to describe primary carcinomas of the lung and to distinguish them from metastases. Identical p53 mutations within a nonthoracic primary tumor and a lung nodule of similar histology may suggest a pulmonary metastasis, as

P.1836

Lefor and co-workers (1986) developed algorithms for patients with squamous cell carcinoma of the head and neck region who developed pulmonary nodules after treatment to determine the nature of the new pulmonary nodule. Characteristics of metastases and of primary lung carcinoma were examined in an attempt to better direct subsequent therapy.

TREATMENT OF PULMONARY METASTASES

Most patients with pulmonary metastases have multiple sites of metastases or unresectable pleural or pulmonary metastases. In these patients, treatment is directed systemically for control of the disease and to palliate symptoms. Although radiation therapy or chemotherapy is frequently used, inconsistent response rarely leads to consistent control or cure. Chemotherapy as initial therapy for these systemic metastases and resection as salvage may provide better results than resection alone. In patients with control of the primary tumor and metastases confined to the lungs, resection of all visualized or palpable metastases may be considered. Complete resection of isolated pulmonary metastases is generally associated with improved patient survival regardless of primary histology.

Chemotherapy

Chemotherapy has not been used routinely for treatment of resectable pulmonary metastasis. However, with the exception of the patient with only one metastasis or with a few metastases and a long disease-free interval, occult micrometastases may commonly exist. For example, in sarcomas, control of the primary tumor may be achieved in various ways; however, later occurrence of pulmonary metastases from existing micrometastases at the time of the control of the primary results in decreased survival compared with patients in whom such occult metastases did not exist. Even with multiple resections, complete eradication of all micrometastases may be unachievable. Use of chemotherapy or other targeted therapies to assist in control of micrometastases may be valuable for systemic control, which may enhance the local control achieved by resection. The traditional measure of postresection survival and postresection disease-free survival may be inadequate when resection is considered as salvage after chemotherapy for pulmonary metastases. A more fitting measure of survival should include survival from diagnosis (including radiologic diagnosis) of metastases. The duration of chemotherapy, the extent of response, the histology of the primary malignancy, and the fitness of the patient all affect the timing of resection and potentially long-term outcomes.

During the past 20 years, the survival rate in patients with osteogenic sarcoma has improved from 20% to approximately 60% to 70%. Limb-sparing procedures have replaced amputation. Neoadjuvant chemotherapy with a variety of agents has been instituted. The incidence of pulmonary metastases in patients with primary osteogenic sarcoma treated with surgical resection and adjuvant chemotherapy has dramatically declined compared with treatment of the primary osteogenic sarcoma with surgery alone, as shown by Skinner (1992), Goorin (1991), and Pastorino (1991) and their co-workers. Hirota and colleagues (1999) have observed that newer agents are increasingly being incorporated into chemotherapeutic strategies. Nonetheless, Ferguson and colleagues (2001) confirmed that relapse still remains a significant problem in these patients. In their report, carboplatin as induction therapy was followed by resection and postoperative multidrug chemotherapy in 37 patients. No patient had a complete response. Patients with metastases confined only within the lungs were more likely to survive than were patients with distant bone metastases. Salvage treatment with resection alone for pulmonary metastasis generates an actuarial survival rate of only about 30%. Salvage chemotherapy with resection may be effective in prolonging survival in patients who develop pulmonary metastases from osteogenic sarcoma, as described by Marina (1992) and Pastorino (1992) and their colleagues. More effective systemic therapies, however, are necessary.

The results of preoperative chemotherapy (high-dose methotrexate, cisplatin, doxorubicin, and ifosfamide), followed by surgery and additional postoperative chemotherapy, have been examined. Goorin and colleagues (2002) found that the combination of etoposide and high-dose ifosfamide as an induction regimen for patients with pulmonary metastasis from osteosarcoma can be effective, despite significant myelosuppression, infection, and renal toxicity. Bacci and co-workers (1997) noted that, in 16 patients, chemotherapy was given, followed by simultaneous resection of the primary and metastatic tumors. Complete resection was accomplished in 15 patients. However, 5 patients died within a few months as a result of undetectable metastatic disease. Survival was strongly correlated with the chemotherapy effects (necrosis) in the primary tumor and in the metastases. Improved survival with combined-modality therapy (chemotherapy followed by salvage surgery) was achieved compared with historic results.

Chemotherapy alone may be insufficient. Jaffe and co-workers (2002) examined the role of chemotherapy in 31 patients with osteogenic sarcoma. Only three patients were cured with chemotherapy alone. Four patients underwent resection. No viable tumor was found in the resected mass. New therapies are needed for better treatments of osteogenic sarcoma. Until then, combinations of chemotherapy and local control (resection) will be needed. Glasser and associates (1992) noted that histologic response to chemotherapy (percentage of necrosis) was the only independent predictor of enhanced survival in a study of 279 patients with stage II osteogenic sarcoma.

P.1837

Based on the effects of chemotherapy in the treatment of primary sarcoma, the effective use of such chemotherapy as planned induction therapy before resection of metastatic disease remains more elusive. Lanza and associates (1991) examined the response of soft tissue sarcoma metastases that were treated with chemotherapy before surgery. Patients were graded as having complete, partial, or no response or progression from the chemotherapy. Survival could not be predicted based on response to chemotherapy alone.

An optimal therapeutic strategy may be to combine systemic and local control, particularly in those patients with recurrent disease (pulmonary metastasis). Chemotherapeutic agents with activity in primary soft tissue sarcoma are limited. According to Benjamin (1975) and Patel (1998) and their colleagues, doxorubicin and ifosfamide are the two most active chemotherapeutic agents for soft tissue sarcoma and have a positive dose-response profile. Resection of pulmonary metastases after optimizing response to chemotherapy may enhance overall local and systemic control resulting in improvements in overall and disease-free survival. This therapeutic model appears to provide a synergistic benefit over and above that which can be achieved with either surgery alone or with chemotherapy alone. Doxorubicin and ifosfamide had been used in a dose-intensive manner with resulting improved response rates, decreased time to progression, and improved survival, particularly in patients treated for higher-risk primary extremity soft tissue sarcoma. This regime for pulmonary metastasis from soft tissue sarcoma provides similar survival benefits to those for other malignancies. Patients with a biological response to chemotherapy before resection may have some benefit in receiving the same combination of chemotherapy after resection. After failure of doxorubicin, dacarbazine (DTIC), and ifosfamide, there are no additional drugs for the treatment of soft tissue sarcomas that are of established value. The other somewhat active agents (methotrexate, etoposide, and interferon- ) have response rates of about 10%.

Other characteristics may suggest effectiveness or lack of effectiveness for various chemotherapeutic agents. Dhaini and colleagues (2003) evaluated human P450 isoenzymes, specifically CYP3A4/5, which aid in the metabolism and detoxification of carcinogens and chemotherapy. The authors found that patients with distant metastases were more likely to have elevated expression of CYP3A4/5 within the primary tumor biopsy specimen than patients without metastatic disease (p = 0.0004). The authors concluded that high levels of this human cytochrome P450 isoenzyme may be a marker to predict metastases or limited survival in patients with primary osteogenic sarcoma. Cyclooxygenase II (COX-II) enzyme levels did not correlate with primary or metastatic disease and the survival, as noted by Dickens and colleagues (2003).

A recommended practice is to consider patients with soft tissue sarcoma with one or two isolated lung lesions and a long disease-free survival for immediate surgery. For patients with more than two lesions, chemotherapy (adriamycin, ifosfamide) could be used to assess a biological response. When maximal response has been achieved, resection can be performed and followed by additional chemotherapy. For unresectable metastases, chemotherapy may provide a response sufficient to allow surgical resection, after which additional chemotherapy may be considered. If chemotherapy is unsuccessful, surgery may be considered for palliation of symptoms. In marginal patients in whom chemotherapy provided only a minimal response or no change, surgery may be considered for local control of the metastases. Occasionally, metastases may grow to enormous size, compressing the heart and mediastinum (a tumor-thorax or tumor tamponade ) with the same consequences of tension pneumothorax or hemothorax. Chemotherapy is not commonly effective in this situation given the need for urgent mechanical intervention. A heroic attempt at resection may be required. Cardiopulmonary bypass for cardiac decompression and cardiopulmonary support may be required simply to manipulate the tumor within the thorax or mediastinum.

Radiation Therapy

Currently, radiation therapy is used for palliation of symptoms of advanced metastases (e.g., extensive pleural involvement, bone metastases). Radiation therapy is rarely used for treatment of pulmonary metastases. Prophylactic lung irradiation has been carried out in patients with osteogenic sarcoma. Burgers and colleagues (1988) reported that patients having prophylactic lung irradiation had similar rates of recurrence of pulmonary metastases as patients having postoperative adjuvant chemotherapy. More recently, Feigenberg and colleagues (2002) proposed whole-lung radiation therapy for patients with pulmonary metastases from giant cell tumors of bone. Whelan and co-workers (2002) suggested that lung radiation could be considered in selected patients for treatment of subclinical lung disease in patients with osteogenic or Ewing's sarcoma. The role of prophylactic lung irradiation in addition to current standard chemotherapy remains to be determined. Spunt and associates (2001) retrospectively reviewed the role of whole-lung irradiation in patients with pulmonary metastases from Ewing's sarcoma who did not respond completely to induction chemotherapy. Only eight patients received irradiation. The 5-year survival rate was 37% in the treated population. There are no randomized studies to suggest that the addition of whole-lung radiation therapy improves outcome in patients with Ewing's sarcoma.

Surgery

In selected patients with resectable pulmonary metastases and absence of extrathoracic metastases, complete resection

P.1838

Selection of Patients for Resection

Patients with isolated pulmonary metastases may be selected for resection. Clinical criteria have been proposed to identify and select patients who can benefit optimally from resection of their pulmonary metastases by McCormack (1978), Mountain (1984), and Pastorino (1997) and their co-workers as well as by the author (JBP) and Roth (1990) (Table 120-1). Unfortunately, most patients with metastases do not benefit from surgery because of one or more of the following reasons: (a) a biologically aggressive tumor characterized by extensive disease, (b) a short disease-free interval (DFI) between control of their primary tumor and identification of pulmonary metastases, and (c) rapid metastatic growth.

In patients being considered for resection, physical examination, radiographic examination, and physiologic assessment are performed to estimate the extent of resection and to determine whether the planned procedure may be safely performed. Cardiac and pulmonary assessment are emphasized. In patients with preoperative chemotherapy, or in those patients in whom pulmonary compromise is expected, a spectrum of pulmonary function tests are performed. These tests include spirometry with and without bronchodilators, diffusion capacity for carbon monoxide (Dlco), and oxygen consumption testing ([V with dot above]o₂max). Echocardiography and exercise stress test also may be needed. In the operating room, the chest radiographs and chest CT scans are displayed prominently. After bronchoscopy, a double-lumen endotracheal tube is placed and is used for anesthetic gas delivery. When a median sternotomy incision is used, sequential deflation of each lung aids in exposure and palpation of the pulmonary nodules. All nodules are resected with a margin of normal tissue. Nodules should not be shelled out because viable tumor cells remain on the periphery of the resected area. The margin should be adequate. Even when the margin is negative, microscopic cells may remain. Higashiyama and colleagues (2002) prospectively evaluated 51 patients with pulmonary metastases with an intraoperative lavage cytology technique for the surgical margin. They found that 11% of patients had a positive cytology at the margin despite having a rim of normal tissue. Additional tissue was then resected. Localized micrometastases may be present in some patients despite a macroscopically negative margin. This may contribute to subsequent local recurrence.

In general, the decision as to the adequacy of margin is the surgeon's alone. After resection, the lung parenchyma may become distorted around the nodule, thereby giving the illusion of a positive or close margin to the pathologist. Mediastinal lymph node metastases rarely occur from pulmonary metastases, as Udelsman and co-workers (1986) have shown.

Is there a limit to the number of metastases that can be resected with associated survival benefit? Several authors have been tempted to address this specific question including myself (1984), Girard (1994), Pastorino (1997), and Robert (1997) and respective colleagues. In general, only unresectability as defined by the thoracic surgeon should be considered as an absolute contraindication to resection. As the numbers of metastases increase, the potential for occult micrometastatic disease also increases. Although the surgeon may be able to extirpate all identifiable disease mechanically by visual inspection and by palpation, the surgeon typically cannot identify nor extripate microscopic disease. The biology of patients with excessive numbers of metastases (but yet still resectable ) is not changed by resection. Balancing the advantages of mechanical resection with the need for control of micrometastatic disease may be best accomplished in a multidisciplinary sarcoma center with a multidisciplinary conference of all potential treating physicians. Pneumonectomy may be a consideration simply for mechanical palliation of mediastinal compression from tumor-thorax as shown by the author (JBP) (1993) and Grunenwald (1997) and respective associates.

Surgical Techniques and Incisions

Surgical procedures for resection include single thoracotomy, staged bilateral thoracotomies, median sternotomy, the clamshell incision, minimally invasive techniques in selected

P.1839

Patients with sarcomas and unilateral nodules often have multiple and bilateral metastases discovered during the operation. Bilateral metastases may occur in 38% to 60% of patients with preoperative unilateral sarcomatous metastasis. Postresection survival rates after median sternotomy or bilateral staged thoracotomies and complete resection are similar. A median sternotomy is performed for the initial exploration and resection in patients with bilateral nodules and may be considered as an initial procedure in patients with pulmonary metastases from osteogenic or soft tissue sarcomas, or suspected bilateral metastases from any primary neoplasm in which wedge resections may be required. An exploration for unilateral or bilateral nodules as well as resection of these nodules may be accomplished through a median sternotomy incision.

Despite the previous discussion, for some unilateral disease, bilateral exploration may not be necessary. High-resolution CT scan may assist in this determination. Younes and colleagues (2002) evaluated the role of ipsilateral thoracotomy in patients with unilateral pulmonary metastasis for contralateral disease-free survival and overall survival. They noted that there was no significant difference in survival in patients who had recurrence in the contralateral lung compared with patients who had bilateral metastases on admission. They suggested that delaying the contralateral thoracotomy did not affect survival.

Laser-assisted Resection

Laser-assisted pulmonary resection, described by Kodama (1991), Branscheid (1992), Miyamoto (1992), Landreneau (1991), Mineo (1998), and Rolle (2002) and their associates, using the neodymium:yttrium-aluminum-garnet (Nd:YAG) laser, may provide a better means of resecting pulmonary metastases than the surgical stapler. Use of the laser may enhance preservation of lung parenchyma with less distortion. Bovie electrocautery may also spare lung parenchyma by removing the metastases with minimal distortion of remaining lung. Air leaks, if they occur, can be sealed by oversewing the parenchymal defect or by the use of fibrin glue. Disadvantages of laser resection may include longer operating time and potential for prolonged postoperative air leaks. Newer laser technologies have been developed. Rolle and colleagues (2002) describe a new 1,318-nm Nd:YAG laser for better and more precise incisions into the parenchyma with concurrent coagulation and sealing of the lung tissue. A 5-mm rim of tissue destruction is achieved.

In a prospective randomized trial conducted by Mineo and colleagues (1998), use of the Nd:YAG laser for resection of lung metastases was examined in 45 patients. The authors identified that the use of the laser reduced hospital stay, air leak, and tissue loss; however, a survival advantage was not proved. The use of the laser for resection of pulmonary

P.1840

Median Sternotomy and Clamshell Incision

For the median sternotomy incision, the patient is positioned supine with the entire anterior thorax exposed from the neck to the umbilicus and laterally to each anterior axillary line. The sternum is divided. The pulmonary ligament is divided on each side to mobilize the lung completely. The lungs are sequentially deflated and palpated. Metastases are identified and resected, and then the deflated lung is reinflated. The deflated right lung may be brought completely into the field, attached by only the hilar structures. Exposure of the left lower lobe may be more difficult than exposure of the other lobes because of the overlying heart. With appropriate gentle traction on the pericardium, the left lower lobe can be exposed quite readily and brought into the operative field. Various techniques to better visualize the lung may be used, such as surgical packs behind the hilum of the deflated lung to elevate the parenchyma or an internal mammary artery retractor to expose basilar tumors or posterior hilar left lower lobe masses. Relative contraindications to median sternotomy include obesity, chronic obstructive pulmonary disease, elevated hemidiaphragm (particularly on the left), and cardiomegaly. Patients with metastases involving the left hilum or the posterior or medial portions of the left lower lobe may benefit from bilateral staged thoracotomies, rather than median sternotomy. A median sternotomy in these situations may compromise completeness of resection and may injure lung parenchyma or create a need for greater pulmonary resection than would be otherwise required.

The clamshell incision, as described by Bains and co-workers (1994), is a modification of the median sternotomy incision. Originally, this approach developed from the early days of cardiac surgery and was later rediscovered for access to enhance bilateral sequential single-lung transplantation. A curvilinear incision is made under the breasts or pectoral muscles (Fig. 120-3A). The pectoral muscles are elevated to gain access to the fourth intercostal space bilaterally, whereupon the chest is entered and the incision carried to the sternum bilaterally. The most lateral aspect of the incision may curve superiorly toward the axilla (Fig. 120-3B). The sternum is divided transversely at the level of the fourth intercostal space with a Gigli or oscillating saw. After placement of a chest retractor for both the right and left thorax, the chest is opened, giving excellent exposure to right and left thorax, hilum, and mediastinum. Advantages of this approach include better exposure of the left hilum posteriorly and the left lower lobe. Disadvantages include a large, painful incision and some difficulty with sternal reconstruction and stabilization.

Thoracotomy

The posterolateral thoracotomy is a familiar and standard approach to pulmonary resection for carcinoma of the lung. Posterolateral thoracotomy (with or without sparing of the latissimus dorsi muscle) may provide better exposure for metastases located more medially, or more posteriorly near the hilum on the left side. In addition, for patients with bulky metastases, a posterolateral thoracotomy provides

P.1841

The vertical axillary thoracotomy may also be considered. Margaritora and colleagues (1999) describe their experience with staged axillary thoracotomy. Hospitalization was short (3.2 days). Operative trauma was minimal, as was postoperative pain. The interval between the two staged procedures was about 24 days.

Bilateral anterior thoracotomy as described by d'Amato and co-workers (2002) also may be used. Bilateral minithoracotomy with video assistance was used as an alternative to the other surgical approaches to the chest.

Video-assisted Thoracic Surgery

Video-assisted thoracoscopic resection using high-resolution video imaging may be helpful for diagnosis, staging, and resection of metastases. Its usefulness is limited, however, because metastases can be identified generally only on the surface of the lung or the outer 10% to 20%, depending on size. Metastases within the lung parenchyma may be undetectable with this technique. In one early report, Landreneau and associates (1992) have described minimal morbidity and no mortality in 61 patients who underwent 85 thoracoscopic pulmonary resections. Lesions were small (<3 cm) and in the outer one third of the lung parenchyma. Metastases in 18 patients were resected through thoracoscopy in this series. Video-assisted thoracic surgery (VATS) was the only procedure performed in these patients.

P.1842

Thoracoscopy may readily be used for diagnosis of metastatic disease, as stated by McCormack and co-workers (1993), but its use in treatment of metastatic disease is more controversial. In an elegant study, McCormack and colleagues (1996) prospectively evaluated VATS resection for treatment of pulmonary metastases. Patients were screened with CT, and VATS was performed on all patients. Under the same anesthetic, thoracotomy or median sternotomy was performed. The authors found more nodules at thoracotomy than had been noted at the VATS procedure. Limitations of the study were the inclusion of patients with multiple metastases or prior sarcoma histology and screening with older CT scans. VATS is not the standard approach for resection in patients with pulmonary metastases. However, VATS may be considered in highly selected patients with a solitary nodule and nonsarcomatous histology on high-resolution (spiral) chest CT scan. Patients with sarcomatous histology frequently (40% to 60%) have occult metastases, which may be palpated and resected with open thoracotomy.

More recently, Landreneau and colleagues (2000) recorded their experience in 80 patients with colorectal metastasis who underwent thoracoscopic resection of pulmonary metastasis. A single lesion was removed in 60 patients, and two or more lesions were removed in 20 patients. The overall 5-year survival rate was 30.8%. The authors required that all lesions identified on CT scan be identified at thoracoscopy or the minimally invasive approach would be abandoned. If location of the lesion compromised a complete resection, conversion to thoracotomy was performed. Accurate, high-resolution CT is critical for selection of patients for minimally invasive techniques as reported by Nakajima and colleagues (2001). Lin and colleagues (1999) also noted that the results appeared comparable to historical results by open thoracotomy. The need for high-resolution helical CT scanning was crucial for patient selection. In addition, conversion to an open procedure was recommended when preoperative lesions were not identified, or when surgical margins would be compromised.

To balance the need for palpation of the lung parenchyma in addition to minimizing trauma with thoracoscopy, Mineo and colleagues (2001) retrospectively evaluated transit xiphoid bilateral palpation during VATS for lung metastasis ectopy. Bilateral palpation was performed in 23 of 29 patients. Fifteen radiographically undetected lesions were identified 11 of which were malignant. The authors recommended that this technique be considered as a blended approach to minimize thoracic trauma while providing for palpation of the lung parenchyma.

At present, VATS can be advocated for diagnosis or staging of the extent of metastases, or for resection of metastases in highly selected patients (i.e., those with solitary, nonsarcomatous histology in peripheral location on high-resolution spiral CT scan).

In patients with solitary metastasis from solid tumor adenocarcinoma or squamous cell carcinoma, careful consideration must be given to excluding primary lung carcinoma, which would require lobectomy and systematic mediastinal lymph node dissection for optimal care. Complications of VATS may include not resecting all metastases, positive margins, or pleural seeding with extraction of the metastasis (Fig. 120-4) as shown by Walsh and Nesbitt (1995), as well as Ang (2003) and Downey (1996) and their co-workers. Follow-up on all patients is necessary at regular intervals because the likelihood of recurrence remains for a period of time.

RESULTS OF RESECTION OF PULMONARY METASTASES

The results of resection for pulmonary metastasectomy require critical analysis of factors that may potentially influence survival. Analysis of results should be based on review or study of single primary histology (breast, colon, melanoma) or similar histology (e.g., soft tissue sarcomas) and sufficient numbers of patients. Prognostic indicators have been reviewed to assess their influence singularly and in combination on postresection survival in patients with pulmonary metastases and to assist clinically in describing appropriate patients for resection of pulmonary metastases. Age, gender, histology, grade, and location of the primary tumor, stage of primary tumor, disease-free interval between resection of the primary tumor and the appearance of the metastasis, number of nodules on preoperative radiologic studies, unilateral or bilateral metastases, tumor doubling time (TDT), and synchronous or metachronous metastases may be evaluated preoperatively. Postoperatively, extent of resection, technique of resection, nodal spread, number of metastases and location, re-resection, postthoracotomy disease-free survival, and overall survival may also be considered in selecting patients for resection of pulmonary metastasis.

Pastorino and associates (1997) reviewed the long-term results of resection of pulmonary metastasis based on an International Registry of Lung Metastases. This International Registry was established in 1991 based on 5,206 patients with pulmonary metastases and treatment collected from Europe, the United States, and Canada. Various clinical characteristics were compared in a retrospective yet consistent and controlled manner. Eighty-eight percent of these patients had complete resection. A solitary metastasis was resected in 2,383 patients; multiple lesions were resected in 2,726. Epithelial histology predominated (2,260 patients), followed by sarcoma (2,173), germ cell (363), and melanoma (328). With a median follow-up of 46 months, actuarial survival was 36% at 5 years, 26% at 10 years, and 22% at 15 years. For incomplete resection, actuarial survival was 15% at 5 years. The multivariate analysis showed several favorable prognostic indicators: resectable metastases, germ cell tumors, disease-free intervals (DFIs) of 36 months or greater, and a solitary metastasis. In this international and multiinstitutional study, the overall operative mortality rate was 1%; the mortality rate was 2.4% after incomplete resections and 0.8% after complete resections.

P.1843

The most frequently performed operation was unilateral thoracotomy (58% of patients). Bilateral exploration was performed through either bilateral synchronous or staged thoracotomy (11%) or median sternotomy (27%). Thoracoscopy was only performed in 2% of patients. Wedge resections (67%), segmentectomy (9%), lobectomy or bilobectomy (21%), and pneumonectomy (4%) also were performed. Only 26% of patients had four or more metastases. Only 9% had 10 or more metastases, and 3% had 20 or more. Multiple metastases were most commonly resected in sarcomas (64%), germ cell tumors (57%), epithelial tumors (43%), and melanoma (39%). Metastases to the mediastinal lymph nodes were uncommon. Three percent had redo surgery. Fifteen percent had two operations, 4% had three operations, and 1% had four or more operations. The maximum number of resections performed on a single patient was seven.

The authors proposed a system by which patients can be grouped into prognostic categories. These would include three parameters: (a) resectability, (b) DFI, and (c) number of metastases. In patients with resectable lesions, a DFI less than 36 months and multiple metastases were found to be independent risk factors. In resectable patients, therefore, three clinically distinct groups could be identified: (a) no risk factors, DFI of 36 months or longer, single metastasis; (b) one risk factor, DFI less than 36 months, or multiple metastases; and (c) two risk factors, DFI less than 36 months, and multiple metastases. Group 4 consisted of all the unresectable patients. The authors noted that median survival was 61 months for group 1, 34 months for group 2, 24 months for group 3, and 14 months for group 4. The discriminant power of the model was appropriate for epithelial tumors, soft tissue sarcomas, and melanomas.

The value of this International Registry of Lung Metastases lies in its large collection of patient characteristics. These clinically identifiable characteristics may be reexamined and analyzed for various hypotheses. The limitations of such a registry lie in not accounting for variables in the biological behavior of these metastases. This variable behavior may be explained by molecular characteristics on which the clinical characteristics are based. This clinical database has to been used to evaluate the value of resection of pulmonary metastases from various histologies, as well as other clinical and molecular characteristics that may be valuable in selecting patients for optimal care of the of their metastases.

Extended Resection of Pulmonary Metastases

Pneumonectomy or other extended resection of pulmonary metastases may be performed safely in selected patients with associated long-term disease-free survival. Less than 3% of all patients undergoing resection of pulmonary metastases require an extended resection. Pneumonectomy or en bloc resection of pulmonary metastases with chest wall or other thoracic structures, such as diaphragm, pericardium, or superior vena cava, have been performed in a small number of patients with good results, as noted by the author and associates (1993). Nineteen patients had a pneumonectomy, and 19 patients had other extended resection. The 5-year actuarial survival rate was 25%. The mortality rate was 5%, and these deaths occurred in patients having pneumonectomy, often after multiple prior wedge resections for metastases.

Pneumonectomy is rarely performed for resection of pulmonary metastases. In a French study by Spaggiari and colleagues (1998), 42 patients were treated over 10 years: 29 patients underwent pneumonectomy for sarcoma, 12 for carcinoma, and 1 for a lipoma. Most tumors were centrally located. Two postoperative deaths occurred. Four patients had major complications. Five patients (12%) had recurrences in the contralateral lung. The median survival time was only 6.25 months, and the 5-year survival rate was 16%. Given that the standard surgical mortality rate for operations for pulmonary metastases is less than 1%, mortality for pneumonectomy should be considered in planning operations for patients with large, centrally located metastases. Although mortality for pneumonectomy for pulmonary metastases corresponds to mortality for other histologies, the 5-year survival rate of only 16% should prompt strict preoperative selection criteria. The authors suggest that young patients, those with a long DFI, and those with normal carcinoembryonic antigen (CEA) levels (for patients with metastases from colorectal carcinoma) be considered for pneumonectomy for pulmonary metastases.

Koong and co-workers (1999) also examined the value of pneumonectomy by retrospective review of the International Registry of Lung Metastases. Of the 5,206 patients who were enrolled, 133 patients (2.6%) had undergone pneumonectomy for pulmonary metastases between 1962 and 1994. Eighty-four percent of these patients underwent complete resection, and the 30-day mortality rate was 3.6%. The 5-year survival rate was 20% with complete resection. For incomplete resection, the perioperative mortality rate was 19%, and most did not survive beyond 5 years. The authors identified favorable prognostic factors of (a) single metastasis, (b) negative mediastinal lymph nodes, and (c) complete resection (R0). The authors concluded that pneumonectomy may be performed safely with adequate long-term survival.

Larger or rapidly growing metastases may compress the mediastinum, or a uniquely positioned metastasis may impinge on or invade cardiac structures or great vessels. The use of cardiopulmonary bypass or other cardiovascular surgical techniques may allow resection of these metastases with palliation of symptoms and the potential for cure. Vaporciyan and colleagues (2002) reviewed a single institution experience of resection with cardiopulmonary bypass of metastatic noncardiac primary malignancies. Patients with inferior vena cava tumors were excluded. Nine patients with sarcomas required cardiopulmonary bypass because

P.1844

Intraatrial extension of sarcoma through the pulmonary vein is rare but also may be safely treated with pulmonary resection (pneumonectomy and resection of the tumor from the left atrium). Extracorporeal cardiopulmonary support is required, as noted by Heslin (1998) and Shuman (1984) and their associates.

Osteogenic Sarcoma

Goorin and associates (1991) and Huth and Eilber (1989) reported that pulmonary metastases from osteogenic sarcoma may occur in up to 80% of patients who relapse after treatment for their primary neoplasm, whether or not they receive adjuvant chemotherapy. CT is commonly used to identify patients with potential metastases. The positive predictive value for chest CT may be limited. Often, the surgeon finds twice the number of nodules that otherwise would be expected simply on the basis of preoperative CT of the chest, as described by Picci and colleagues (2001). Resection as initial therapy for solitary metastasis, resection as salvage after chemotherapy for these pulmonary metastases, and multiple repeat thoracotomy may all be considered in selecting an optimal therapeutic strategy.

Meyer and associates (1987) reported that because osteogenic sarcoma metastases are often isolated to the lungs, resection may render a significant number of patients disease free and enhance long-term survival. The 5-year survival rate may range up to 40%, as shown by Snyder (1991) and Belli (1989) and their co-workers. Patients may have benefit regardless of the time of identification of lung metastases. In one Japanese study, Tsuchiya and colleagues (2002) noted that a longer DFI was associated with improved 5-year survival. Still, 2-year survival from the time of identification of pulmonary metastases ranged from 24% to 33% for patients with lung metastases at initial presentation, those with lung metastases identified during preoperative chemotherapy, and those with lung metastases identified during postoperative chemotherapy. Patients with lung metastases that occurred or were identified after completion of chemotherapy had a 2-year overall survival of 40% and a 5-year survival rate of 31%. However, in a small study, Yonemoto and colleagues (1998) evaluated 117 patients with osteogenic sarcoma of the extremity; 9 patients had pulmonary metastases at presentation. Patients who were treated with chemotherapy and aggressive resection had a 5-year survival rate of 64%.

Carter (1991), Jaffe (1983), and the author (JBP) (1983) and respective associates have evaluated survival and prognostic factors in patients with pulmonary metastases from osteogenic sarcoma. In a series from the National Institutes of Health, Tthe author and colleagues (1983) evaluated 80 patients with osteogenic sarcoma of the extremity. Forty-three patients developed pulmonary metastases, and 39 patients underwent one or more thoracic explorations for resection of their metastases. The 5-year survival rate was 40%. Various prognostic factors were analyzed. Three or fewer nodules, longer DFI, resectable metastases, and the fewer metastases identified and resected were associated with longer postthoracotomy survival. Resection was not possible if more than 16 nodules were identified on preoperative tomograms. A multivariate analysis did not find any combination of factors to be more predictive than the number of nodules identified on preoperative tomograms. In a more recent study by Heij and co-workers (1994) of 40 children with osteogenic sarcoma, it was found that incomplete excision, lack of primary tumor control, and progression and development of metastases during treatment were all negative prognostic factors. Surprisingly, in resectable patients, the number of metastases, DFI, unilateral versus bilateral metastases, preoperative and postoperative adjuvant treatment, and the number of thoracotomies performed were not significant prognostic factors.

Chemotherapy may prevent or cure micrometastatic disease not amenable to surgery, as stated by Belli and associates (1989). Most patients with osteogenic sarcoma of the extremities are treated with neoadjuvant chemotherapy and limb salvage as proposed by Bacci (2001), Goorin (2002), and Ferrari (2003) and their associates. On the other hand, patients treated with adjuvant chemotherapy after local control (typically with a limb salvage procedure) will show a different pattern of systemic relapse. Likewise, such a therapeutic approach was not shown to decrease event-free survival by Goorin and colleagues (2003). All patients received 44 weeks of combination chemotherapy. They also reported that presurgical chemotherapy (for 10 weeks) did not improve event-free survival compared with immediate surgery. Voute and associates (1999) have suggested that combination chemotherapy such as cisplatin and doxorubicin, or cisplatin ifosfamide, and doxorubicin, would be active in patients with osteogenic sarcoma. Miniero and colleagues (1998) have suggested that high-dose chemotherapy and autologous peripheral blood stem cell transplantation could be considered as a promising regimen for patients with metastatic osteogenic sarcoma, particularly those who were not cured by conventional chemotherapy.

Chemotherapy may assist in treating newly diagnosed metastatic osteogenic sarcoma as recommended by Ferguson and associates (2001). The response to chemotherapy, as a biological prognostic factor, may be considered in future studies. In patients with osteogenic sarcoma, such response may be difficult to assess given the calcified matrix of these metastases.

P.1845

Soft Tissue Sarcomas

Soft tissue sarcomas comprise a family of nonossifying malignant neoplasms arising from mesenchymal connective tissues that can metastasize to the lungs, as described by Hoos and colleagues (2000). Potter and associates (1985) demonstrated that, as with osteogenic sarcomas, local recurrence is common (20%), and metastases are predominantly to lungs.

Billingsley and colleagues (1999) performed a multifactorial analysis of 994 patients with primary extremity soft tissue sarcoma treated at the Memorial Sloan-Kettering Cancer Center; 230 patients recurred. Of disease that recurred, 73% (169 of 273) of cases recurred initially within the lungs. Median survival after recurrence of metastases was 11.6 months. The multivariate analysis identified resection of metastatic disease, disease-free interval, the presence of local recurrence, and age more than 50 years as significant prognostic indicators. Primary tumor characteristics did not have an association with survival after resection of pulmonary metastases. In general, a longer disease-free interval (<6 months) and three or fewer metastases were associated with a higher overall 5-year survival. Temple and Brennan (2002) recently reviewed the role of pulmonary metastasectomy in the management of soft tissue sarcoma. Belal and colleagues (2001) reviewed 23 patients with soft tissue sarcomas treated with pulmonary metastasectomy. They also found that a longer DFI (<6 months) and fewer metastases (three or less) were associated with a higher overall 5-year survival.

In a retrospective study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group, van Geel and colleagues (1996) noted a 5-year overall survival rate of 38% after complete resection of pulmonary metastases. Negative margins, younger age (less than 40 years), and lower-grade tumors (grade 1 or 2) were associated with improved survival compared with patients without these characteristics. These authors also suggested that additional study would be needed before chemotherapy could be recommended as additional therapy.

In patients with histologically documented pulmonary metastases from soft tissue sarcomas treated at the National Cancer Institute, Jablons and associates (1989) showed that significant preoperative predictors of enhanced survival included TDT (<20 days), number of metastases on preoperative tomograms (less than four nodules), and DFI (<12 months). Predictive ability for better survival was improved when all three prognostic factors were combined. These patients represent the patients who will have the best response (i.e., prolonged postresection survival) to pulmonary metastasectomy. Casson and co-workers (1992) evaluated determinants of 5-year survival in 58 patients who had complete resection and who were followed until death or for a minimum of 5 years. Favorable prognostic factors included TDT greater than 40 days, unilateral disease, three or fewer nodules identified on preoperative tomograms, two or fewer metastases resected, and tumor histology (median survival, 33 months for malignant fibrous histiocytoma versus 17 months for all others). Using multivariate analysis, the number of nodules (at least four) was the most significant adverse prognostic indicator. The addition of tumor histology (malignant fibrous histiocytoma) improved the predictive ability of this model. Absolute 5-year survival rate was 25% (15 of 58 patients).

Resection of recurrent pulmonary metastases is associated with improved postresection survival compared with patients with unresectable metastases. Pogrebniak and co-workers (1991) evaluated 43 patients who had two or more resections. In 31 completely resectable patients (72%), median survival was 25 months, compared with not completely resectable or unresectable patients, who had a median survival of only 10 months. A longer DFI ( 18 months) was also associated with prolonged disease-free survival. Increased age and female gender were associated with an increased risk for death from disease in resected patients with recurrent pulmonary metastases, in contrast to initial isolated pulmonary metastases. Casson and associates (1991) noted that in 39 patients with recurrent pulmonary metastases from adult soft tissue sarcomas, resectable patients and those with only one metastasis had the best postresection survival. Chemotherapy for metastatic soft tissue sarcoma remains poor. Median survival ranges from 13 to 16 months, as reported by Weh (1990) and Elias (1989) and their colleagues. However, a multidisciplinary treatment plan of combination chemotherapy, surgery, and additional chemotherapy may provide improved survival over single-modality care, as noted by Rosen and co-workers (1994). A larger meta-analysis from the Sarcoma Meta-analysis Collaboration (1997) suggested that adjuvant chemotherapy (doxorubicin based) was associated with prolonged local and distant recurrence-free survival and that overall survival trended to improvement.

Colorectal Neoplasms

Colorectal metastases commonly spread to local or regional nodes, or to the liver through the portal venous system. Colorectal metastases may also occur initially as pulmonary metastases.

The distinction between a single metastasis from colorectal carcinoma and a primary NSCLC is typically made by histology and the specific markers. Molecular markers specific for metastases or lung carcinoma may aid in distinguishing a metastatic tumor from a primary lung carcinoma. Complete resection, where possible, should be considered. Screening of serum CEA should be performed in all patients with prior diagnosis of colorectal carcinoma. Although TTF-1 and SP-A are good markers of NSCLC, good markers for colorectal metastases to the lung are

P.1846

Patients with pulmonary metastases from colorectal carcinoma may be resected safely with low morbidity and mortality and long-term survival. Saito and colleagues (2002) described a 5-year survival rate up to 39.8% after resection of pulmonary metastases from colon carcinoma. Differences in age, gender, location, grade, and stage of the primary colorectal cancer are not associated with either improved or worsened survival after resection of these metastases. In a large series from the Mayo Clinic, McAfee and associates (1992) presented 139 patients who underwent resection of pulmonary metastases from colorectal carcinoma with an operative mortality rate of 1.4%. The overall 5-year survival rate was 30.5%, and the median follow-up was 7 years. Patients with a solitary pulmonary metastasis and those with a preoperative CEA level less than 4.0 ng/mL had better postthoracotomy survival than other patients. Of interest was that longer DFI and diameter of metastases of less than 3 cm were not associated with improved survival. More recently, Higashiyama and colleagues (2003) described a strong association between prethoracotomy serum carcinoembryonic antigen and survival. Patients with a high prethoracotomy serum CEA were more likely to have extrathoracic metastases. The authors recommended evaluation for extrathoracic metastases in patients with a high prethoracotomy serum CEA. Saito and collegues (2002) evaluated the role of resection in 165 patients with pulmonary metastases of colorectal carcinoma. The overall 5-year survival rate was 39.6%. Patients with hilar or mediastinal lymph node metastasis had only a 6% survival rate at 4 years. In patients with recurrent metastases, the 5-year survival rate was 52% from the time of the second thoracotomy. The authors noted that patients who had previously undergone resection of hepatic metastases had 5-year postthoracotomy survival similar to that in patients with pulmonary metastases as the first site of metastasis. The authors confirmed the prognostic advantage of a normal prethoracotomy CEA, no metastases within the hilar or mediastinal lymph nodes, and complete resection.

Patients may have pulmonary metastases develop after resection of hepatic metastases. In these patients, complete resection of pulmonary metastases may be associated with improved survival. Labow and associates (2002) noted a 3-year survival rate of 60% in the resected group versus 31% in the nonresected group. Patients in this study met the standard criteria for resection of pulmonary metastases, and there was no concurrent extrapulmonary disease.

Patients with resection of colon metastases from the lung and the liver have a survival advantage with complete resection, as suggested by Murata (1998) and McCormack (1992) and their co-workers. Robinson and colleagues (1999) reported that in 48 patients with both liver and lung metastasis, 25 patients underwent resection, and 23 patients did not. Median survival was longer after resection of the last metastasis (either lung or liver) than in those individuals who did not undergo resection (16 months vs. 6 months; p< 0.001). They also noted that patients with metachronous resections survived longer than did patients with synchronous resections (70 months median survival compared with 22 months, p < 0.001). The authors noted that the ideal candidate for resection was younger than 50 years of age, had a solitary liver metastasis, and had had a 4-year interval between the colorectal cancer resection and occurrence of the pulmonary metastasis. The poorest patients for resection included those aged 70 years or older, those with multiple liver metastases, and those with synchronous disease. In a French study by Regnard and associates (1998), the authors examined 43 patients who had undergone complete resection of hepatic metastasis and then subsequently developed pulmonary metastases. The median survival time was 19 months, and the 5-year survival rate was estimated to be 11%. Patients with a CEA exceeding 0.5 ng/mL had a significantly lower probability of survival than did those with lower levels [<0.5 ng/mL (p = 0.0018)]. Rena (2002), Sakamoto (2001), and Inoue (2000) and their colleagues also noted that a normal prethoracotomy serum CEA was a favorable prognostic factor. Follow-up for these patients should include radiographic examinations and serum CEA as proposed by Ike and colleagues (2002).

In patients with colorectal metastases to both liver and lung, complete resection is generally associated with improved survival. Whether the liver metastases occur first and then lung metastases, or whether lung metastases occur first and then liver metastases, complete surgical resection is required when possible. Nagakura and colleagues (2001) retrospectively reviewed patients who underwent both hepatic and pulmonary resections. Patients who had sequentially detected metastases had a cumulative 5-year survival rate of 44%; in patients who had simultaneously detected pulmonary and hepatic metastases, the 5-year survival rate was 0%. They concluded that patients with simultaneous detection of hepatic and pulmonary metastases from colorectal carcinoma should not undergo resection. Unfortunately, the potential for improvement in survival with chemotherapy is limited, and a patient with resectable disease should be considered for mechanical extirpation of his or her metastases.

Other predictors of survival have been studied; the role of the adhesion molecule CD44 variant 9 and its relationship to pulmonary metastases from colorectal cancer were examined in 42 patients by Goi and colleagues (2002). The overall survival rate was 35%. Patients with elevated CD44 variant 9 had a higher rate of pulmonary metastases (88%) than patients in whom CD44 variant 9 was normal (the rate of metastases was only 42%).

P.1847

Poorer survival is found in patients who cannot be completely resected or who are denied operation because their disease is deemed to be unresectable. Of the total population of patients with colorectal metastases, those with completely resectable lung or hepatic metastases represent a small percentage and are the ones with the most favorable biology of the tumor. The surgeon can take advantage of this biologically favorable subset of patients. With sequential and complete resection of both lung and hepatic metastases, survival may be enhanced.

Breast Carcinoma

Patients with metastases from breast carcinoma have poor survival because breast metastases occur in multiple sites. Treatment is typically systemic by chemotherapy or other therapy, or palliative in nature. In a minority of patients, resection of isolated pulmonary metastases may be considered. Patanaphan and colleagues (1988) described 145 patients with metastatic breast carcinoma (145 of 558, or 26%); the major sites of metastases were bone (51%), lung (17%), brain (16%), and liver (6%). Overall median survival was 12 months for patients with lung metastases, who were mostly treated with either palliative chemotherapy or irradiation or both. Lanza and co-workers (1992) studied 44 women with a prior history of breast cancer who underwent pulmonary resection for new pulmonary lesions. Seven patients were excluded who had benign nodules (3 patients) or unresectable metastases (4 patients). In 37 resectable patients, the actuarial 5-year survival rate was 50% (Fig. 120-5). Disease-free interval exceeding 12 months was associated with a longer median survival time (82 months) and 5-year survival rate (57%) compared with patients with a DFI of less than 12 months (15 months median, 0% 5-year survival; p= 0.004). Estrogen receptor positive status tended to be associated with a longer postthoracotomy survival (p= 0.098). Other favorable prognostic factors included positive receptor status of the primary tumor (improved 3-year survival rate of 61%) compared with negative receptor status (38% 3-year survival rate). Bathe and coauthors (1999) described the distant sites of failure following ablation of liver and lung metastases. They recommended that adjuvant therapy with significant activity against visceral metastasis might enhance survival. Resection of solitary metastasis, according to Friedel and associates (1994), provided a 35% 5-year survival rate, as compared with 0% after resection of five or more metastases. Simpson and colleagues (1997) noted that favorable selection of patients has enabled survival to increase up to 62% at 5 years.

Staren and co-workers (1992) evaluated 33 patients treated with surgical resection of pulmonary metastases from breast carcinoma and compared the results to that of 30 patients treated primarily with systemic chemotherapy and hormonal therapy. Patients having complete resection of metastases had a longer median survival than did patients with medical therapy, particularly when single nodules were compared (58 months vs. 34 months median survival). The 5-year survival rate in patients treated with some surgical resection was 36%, compared with 11% in patients treated only with medical treatment. A review by Bodzin and associate (1998) confirms these findings. More recently, Friedel and colleagues (2002) reviewed the results of resection in 467 patients with isolated metastases from breast cancer. The 5-year survival rate was 38%. The authors noted that disease-free intervals of greater than or equal to 36 months were associated with a 5-year survival rate of 45%. Patients with solitary metastases did better than those with multiple metastases, although the data were not statistically significant.

Testicular Neoplasms

Nonseminomatous testicular tumors can be diagnosed by the occurrence of new pulmonary nodules identified on chest radiograph or by CT scan, as described by Lien (1988) and Tesoro-Tess (1987) and their colleagues. Metastatic testicular seminoma most commonly is identified as mediastinal nodal enlargement. CT scan therefore is more accurate in diagnosis of seminomatous metastases than are plain

P.1848

Cytoreductive surgery in patients with disseminated nonseminomatous germ cell tumors of the testis may be performed for removal of residual metastatic disease after chemotherapy. The response to chemotherapy has been maximized when there is no further reduction in size of the nodules. Most patients require retroperitoneal lymph node dissections (69%), although thoracotomies may be required in 18% of patients. Kulkarni (1991) and Carsky (1992) and their associates evaluated 80 patients with germ cell tumors and lung metastases treated with chemotherapy and subsequent surgery. In this series, 35% (28 patients) achieved complete response after chemotherapy; 45% (36 patients) with partial response underwent surgery for resection of residual metastases. The residual disease was in the abdomen (17 patients), the lungs (15 patients), or both (4 patients); 27 of 36 patients (75%) achieved complete response after both chemotherapy and surgery. Carter and colleagues (1987) noted that extensive pulmonary metastases (unresectable metastases) were a predictor of ultimate treatment failure. In contrast, Gels and associates (1997) reported a 10-year survival rate of 82.2% after resection of residual retroperitoneal and pulmonary tumors after chemotherapy. Morbidity after surgical resection was minimal.

Liu and colleagues (1998) evaluated the role of pulmonary metastasectomy for testicular germ cell tumors over a 28-year period. The typical patient was young (median age, 27 years). Preoperative tumor markers were normal in most patients, and patients with multiple metastases predominated. About one half the patients had synchronous presentation of their metastases. Complete resection was generally possible. Most of these patients had already undergone chemotherapy. Viable metastasis was present in 44% of the patients, and in the remainder, there was no viable tumor (mature teratoma and fibrosis or necrosis were equally represented). Twenty-five percent had metastasis to other sites after resection of their pulmonary metastasis. The overall 5-year survival rate was 68%, and for the patients diagnosed after 1985, the survival rate was 82%. The authors noted that extrathoracic metastasis (nonpulmonary visceral sites) as well as the presence of viable tumor in the resected specimens were adverse prognostic indicators. Patients with metastases outside the pulmonary parenchyma, elevated tumor markers, and a viable tumor had a worse prognosis. Parenchymal resection not only removed all identifiable disease but also provided a measure of the effectiveness of their chemotherapy treatment.

Schnorrer and co-workers (1996) described 28 patients with pulmonary metastases from germ cell or testicular neoplasms who were treated with bleomycin, etoposide, and cisplatin. An overall complete response was achieved in 21 patients (75%); in 11 of them, a complete response was achieved after chemotherapy alone. Resection of residual mass was necessary in 12 patients with normalized serum markers. Resection of the residual mass was recommended for histology and may modify subsequent treatment. The overall cure rate was 89.3%.

In one multiinstitutional study of 215 patients, Steyerberg and colleagues (1997) evaluated the potential to predict necrosis, mature teratoma or cancer in the residual pulmonary masses. Necrosis (54%) and mature teratoma (33%) predominated, whereas cancer occurred in 13%. The authors recommended that the retroperitoneal lymph node dissection (RPLND) be performed before thoracotomy because the pathology found during RPLND was a strong predictor of pathology at thoracotomy.

Gynecologic Neoplasms

Various authors, including Niwa (2002), Anderson (2001), Chauveinc (1999), Shiromizu (1999), and Bouros (1996) and their co-workers, have discussed the role of pulmonary resection, as well as other therapy, for the treatment of metastatic uterine and cervical malignancies. Fuller and colleagues (1985) from the Massachusetts General Hospital reviewed a 40-year experience of treating patients with pulmonary metastases from gynecologic cancer. The 5-year survival rate was 36%. Lesions less than 4 cm in diameter and a DFI exceeding 36 months were associated with prolonged survival. Shiromizu and colleagues (1999) confirmed that a smaller size (2.8 cm average) of the metastasis, a smaller number of metastases (one to three), and no lymph node metastasis were important favorable prognostic factors.

Levenback and co-workers (1992) reviewed 45 patients with pulmonary metastases from uterine sarcomas. Most patients (71%) had unilateral lesions, and 51% had only one lesion. The 5-year survival rate was 43%. Unilateral metastases or fewer numbers of metastases were not significantly associated with prolonged survival. Leitao and associates (2002) reviewed 41 patients with recurrent uterine leiomyosarcoma. Eighteen patients had distant metastases, and six had both local and distant metastases. Seventeen patients had local recurrence. Thoracic resection was performed in 13 patients. The authors noted that the disease-free interval and complete resection were predictors of improved survival. Kumar and colleagues (1988) reviewed 97 patients with metastatic gestational trophoblastic disease; chemotherapy was the treatment of choice. Selective thoracotomy in patients with solitary lung metastases reduced the treatment time and need for further aggressive chemotherapy. The overall 2-year survival rate after diagnosis was 65%. A DFI of less than 1 year was associated with poorer survival. Barter and associates (1990) studied 2,116 patients with primary cervical malignancy between 1969 and 1984 and found 88 patients (88 or 2,116, or 4.16%) with pulmonary lesions consistent with metastases. Prognosis was poor with chemotherapy only (median survival,

P.1849

Resection of pulmonary metastases from squamous cell carcinoma of the uterine cervix has also been described by Fuller (1985) and Seki (1992) and their coinvestigators.

The role of chemotherapy for treatment of endometrial cancer is evolving, although there is no standard chemotherapy regimen for patients with metastatic disease. Niwa and colleagues (2002) reported on patients who underwent chemotherapy with paclitaxel and carboplatin. The multiple lung metastases either disappeared or remained as scars after six courses. The patients have remained disease free for 28 and 7 months, respectively.

Renal Cell Carcinoma

Various series have examined the value of resection of pulmonary metastases from renal cell carcinoma. Several recent collected series have demonstrated the safety and efficacy of resection of metastatic renal cell carcinoma. Pfannschmidt and colleagues (2002) noted that complete resection of the pulmonary metastases, absence of primary tumor recurrence, and absence of other extrathoracic metastatic disease were associated with a 5-year survival rate of 36.9%. In patients with complete resection, the 5-year survival rate was 41.5%, compared with 22.1% for patients with an incomplete resection. Multivariate analysis demonstrated that the number of pulmonary metastases, the involvement of lymph nodes with regional metastases, and the length of the disease-free interval were overall predictors of survival. Similar findings were confirmed by Piltz (2002) and Friedel (1999) and their colleagues as well as by Fischer and Schmid (1999).

Schott and colleagues (1988) reported 39 patients (4.1%) with pulmonary metastases after nephrectomy for renal carcinoma in 938 patients. Patients with pulmonary metastases less than 2 cm in diameter and limited to one site had prolonged survival and DFI compared with other patients. Pogrebniak and associates (1992) from the National Cancer Institute reported 23 patients who underwent resection of pulmonary metastases from renal cell carcinoma, of which 18 had previous interleukin-2 based immunotherapy. The patients who were resected (15 of 23, or 65%) had a longer survival (mean, 49 months; median not yet reached) than did the unresected patients (median, 16 months; p= 0.02). Postresection survival did not depend on the number of nodules seen on CT, resected nodules, or the DFI. The 5-year survival rate after complete resection was 44% in one study of 50 patients undergoing resection of metastases from renal cell carcinoma. Twelve patients had repeat resection, achieving a 42% 5-year survival rate after second resection. Complete resection was the most important factor associated with 5-year survival, as described by Fourquier (1997) and van der Poel (1999) and their colleagues.

Melanoma

The overall biological behavior of melanoma cannot be predicted. Most commonly, pulmonary metastases occur in addition to other regional (lymphatic) or visceral sites, and overall long-term survival is poor. Immunotherapy has been used with some favorable results. In the rare patient who presents with isolated pulmonary metastases, resection may be associated with long-term survival, as noted by Ollila and Morton (1998). Current 5-year survival rates range from 4.5% to 25%. In a large series of 1,521 patients with American Joint Committee on Cancer stage IV melanoma, the 5-year survival rate was only 4% (median survival, 8.3 months), as reported by Barth and co-workers (1995). Hofmann and colleagues (2002) suggested that imaging procedures for routine follow-up in nonmetastatic melanoma were not cost effective. FDG PET scans may be effective for screening for extrathoracic metastases in patients with potentially isolated pulmonary metastases. Patients who underwent resection of radiologically isolated pulmonary metastases had a 5-year survival rate of 22.1%; and patients who underwent a PET scan preoperatively had significantly better 5-year survival, as noted by Dalrymple-Hay and colleagues (2002). Allen and Coit (2002) found that patients with early recurrence (within 1 year), multiple metastases, and incompletely resected metastases will have poor survival.

Gorenstein and associates (1991) evaluated 56 patients with histologically proven pulmonary metastases from melanoma. The overall postresection survival rate was 25% at 5 years (Fig. 120-6). Patients with earlier primary stage melanoma and patients with metastases to the lungs as the first site of metastases had longer postresection survival than did other patients. Neither location of the primary tumor, histology, thickness, Clark level, nodal metastases, metastasis doubling time, nor type of resection of the primary tumor was associated with improved postresection survival.

Harpole and co-workers (1992) and Lewis and Harpole (2002) evaluated pulmonary metastases in 945 patients in a population of 7,564 melanoma patients. Bilateral as well as multiple metastases were present in most of these patients. Multivariate predictors of survival included complete resection, DFI, chemotherapy, two or fewer metastases, negative lymph nodes, and histologic type. The 5-year survival for all 7,564 patients was 4%, in contrast to a 20% 5-year survival rate in patients with resection of the pulmonary metastases.

P.1850

Squamous Cell Carcinoma

Squamous cell carcinomas of the aerodigestive tract may occur in one or multiple areas. In patients with a primary squamous cell carcinoma outside the lung, lung metastases or a separate primary lung cancer may occur. Secondary lung neoplasms may be resected with survival benefit, as emphasized by Nibu and colleagues (1997). They found an overall 5-year survival rate of 32% in patients undergoing complete resection of squamous cell carcinoma metastases resulting from a primary tumor of the head and neck. A solitary pulmonary nodule correlated with improvement in survival. Finley and co-workers (1992) described factors associated with improved survival in patients with squamous cell carcinoma metastases from head and neck cancers. These included complete resection, control of primary, early stage of head and neck primary, one nodule on chest radiograph, and longer DFI (<2 years) from primary resection. Complete resection of all malignant disease was associated with a 5-year survival rate of 29%. The number of nodules was not significantly associated with survival. However, in eight patients with more than one nodule, median survival was 2 years, and there were no 5-year survivors. Therefore, the benefits of resection of multiple pulmonary metastases from head and neck primary squamous cell carcinoma are not completely clear. In another study of 44 patients, the 5-year survival rate after pulmonary resection was 43%.

Tan and colleagues (1999) reviewed the role of screening CT of the chest in patients with newly diagnosed advanced head and neck cancers. The authors noted that CT of the chest did not add to the sensitivity of the screening for pulmonary metastasis or second lung primary. In another study, Troell and Terris (1995) noted the chest radiograph had a sensitivity of 50% and a specificity of 94% for detection of pulmonary metastases. The authors recommended a chest radiograph as a gross screening examination followed by CT of the chest for an abnormal chest radiograph.

However, when a radiographic examination reveals a solitary pulmonary lesion after treatment of primary squamous cell carcinoma elsewhere in the body, the origin of the pulmonary lesion remains questionable. The lesion may represent a solitary metastasis, a primary bronchial carcinoma, or a benign process. In patients with a prior diagnosis of NSCLC, the lesion would be considered a metastasis (if similar histology and within 2 years of original resection), or perhaps a new primary (if greater than 2 years from original resection). The recommended treatment for such a solitary lesion is bronchoscopy, thoracic exploration, and an excisional biopsy. If a squamous cell carcinoma is identified, a lobectomy and systematic mediastinal lymph node dissection should be performed, particularly if there is any question that the lesion could be a second primary neoplasm. In patients with compromised pulmonary function, a sublobar resection may be required.

Lefor and colleagues (1986) attempted to correlate primary carcinomas of the head and neck with subsequent development of pulmonary metastases or second primary lung carcinomas. They used an algorithm that considered the DFI, histology, radiographic findings, and characteristics of the lung lesion as well as the identification of mediastinal lymphadenopathy. The authors recommended that indeterminate lesions be treated as primary lung carcinomas (e.g., lobectomy and mediastinal lymph node dissection) because this provides the best local control of the disease as well as the potential for cure. Leong and associates (1998) proposed

P.1851

Childhood Tumors

Primary tumors of childhood, such as hepatoma, neuroblastoma, hepatoblastoma, osteogenic sarcoma, Ewing's sarcoma, and rhabdomyosarcoma, commonly spread to the lungs; however, other sites of metastasis are frequent (with the exception of osteogenic sarcoma). Chemotherapy remains the major treatment modality for metastases in multiple sites in children. Pulmonary resection for metastases may be required for initial or salvage therapy, to document metastases in the lungs, to assess the tumor's response to chemotherapy or the viability of the remaining tumor, and to enhance postresection survival in these children with resectable metastases.

Hepatoblastoma

Hepatoblastoma may metastasize to the lungs in about 44% of patients, as described by Uchiyama (1999), Karnak (2002), and Perilongo (2000) and their associates. Improved survival in patients with hepatoblastoma requires a multidisciplinary treatment program including resection of the hepatoblastoma in the liver, combination chemotherapy (cisplatin based), and resection of isolated pulmonary metastases.

Combination chemotherapy is recommended by Katzenstein (2002) and Nishimura (2002) and their coinvestigators. Perilongo and colleagues (2000) described a prospective single-arm study of preoperative cisplatin and doxorubicin for patients presenting with metastases from neuroblastoma. Resection of the metastases follows. In this study, conducted by the International Society of Pediatric Oncology, 31 of 154 patients presented with metastases. The authors recommended this aforementioned therapeutic strategy for systemic and local control in these children. The International Society of Pediatric Oncology also reported 40 children with hepatocellular carcinoma; this study was published by Czauderna and colleagues (2002). Metastatic disease was identified in 31% of these patients and was associated with a worse survival compared with children with hepatoblastoma.

Wilms' Tumor

Patients with Wilms' tumor may present with pulmonary metastases at the time of diagnosis or as a relapse after initial treatment, as recorded by Macklis and colleagues (1991). Early diagnosis using CT may identify metastases in up to 36% of patients, as shown by Wilimas and co-workers (1988). Wootton-Gorges and colleagues (2000) suggested that plain chest radiograph alone appears adequate for the diagnosis or exclusion of pulmonary metastases in patients with Wilms' tumor. Pulmonary metastases may be resected safely from children with Wilms' tumor, as emphasized by de Kraker and associates (1990) and Di Lorenzo and Collin (1988). Green and colleagues (1991) described 211 patients entered in one of the three National Wilms' Tumor Studies whose initial relapse was in the lungs and who showed no survival advantage to resection of pulmonary metastases over treatment with chemotherapy and whole-lung irradiation. Combination therapy, including surgery, radiation, and chemotherapy, may be applied. Godzinski and co-workers (1999) noted that 4-year disease-free survival with pulmonary metastasis reached 83% with such therapy.

Ewing's Sarcoma

Ewing's sarcoma metastasizes preferentially to the lungs in children and may be resected as described by Bacci and colleagues (1995). Lanza and co-workers (1987) examined patients with resectable pulmonary metastases from Ewing's sarcoma. These patients had prolonged survival (actuarial 5-year survival, 15%; median, 28 months) compared with patients who were explored but found to have unresectable metastases (no survivors beyond 22 months; median survival, 12 months; p = 0.0047). Patients with four nodules or fewer had better survival than patients with more than four nodules. Lung irradiation may aid in prolonging survival, according to Spunt (2001) and Dunst (1993) and their associates. Spunt and colleagues (2001) noted a 5-year survival rate of 37.3%. They recommended whole-lung irradiation in good responders to chemotherapy. The European Intergroup Cooperative Ewing's Sarcoma Studies reported 114 patients with Ewing's sarcoma who underwent perioperative chemotherapy and local treatment for the primary tumor. Whole-lung radiation therapy (15 to 18 Gy) was given to 75 patients; 63% of first relapses involved the lung. Adverse risk factors included poor chemotherapy response of the primary tumor, bilateral metastases, and no lung irradiation, as reported by Paulussen and colleagues (1998a, 1998b).

Osteogenic Sarcoma

Osteogenic sarcoma metastasizes preferentially to the lungs and may be resected with associated survival benefit (see prior discussion on osteogenic sarcoma within this

P.1852

Recurrent Pulmonary Metastases

If pulmonary metastases recur in the lungs after initial complete resection, repeat resection (one or more times) can be performed. Patients may be selected by the same criteria as presented in Table 120-1. Groeger and associates (1997) noted that resection can again be accomplished safely, with prolonged postthoracotomy survival. In one report by Kandioler and colleagues (1998) of 396 operations in 330 patients, the authors identified a subgroup of 35 patients who had undergone reoperation for pulmonary metastasis. In this group, 5- and 10-year survival rates were 48% and 28%, respectively. The favorable prognostic factors included a DFI greater than 1 year. There was no survival advantage associated with histology, whether epithelial carcinoma, osteosarcoma, or soft tissue sarcoma. In this patient population, successful repeat surgical resection of pulmonary metastasis and survival advantage are probably related to a favorable biological behavior. The specific criteria for this favorable behavior are not yet known.

Several studies have reviewed results of multiple resections for recurrent pulmonary metastases. Rizzoni and associates (1986) described 29 patients with recurrent pulmonary metastases from soft tissue sarcomas with two or more resections of pulmonary metastases. Patients with favorable tumor biology (resectable metastases, longer TDT, three nodules or less, and DFI of more than 6 months) had longer survival. There was no operative mortality, and complications occurred in only 7.5%. Median survival was 14.5 months, and the overall 5-year survival rate was 22%. Resectable patients had a median survival of 24 months. Casson and colleagues (1991) confirmed these findings in 39 patients with adult soft tissue sarcomas. Thirty-four patients were resectable (median survival time, 28 months; 5-year survival rate, about 32%). Unresectable patients had a median survival of 7 months. Median survival after resection of a solitary recurrent metastasis was 65 months, as compared with patients with two or more nodules (14 months median; p = 0.01). Weiser and co-workers (2000) reviewed patients who relapsed after complete resection of isolated pulmonary metastases. The postthoracotomy disease-specific survival following re-resection was 42.8 months (estimated 5-year survival rate, 36%). The authors noted three independent prognostic factors associated with favorable outcomes: (a) one or two nodules, (b) size less than or equal to 2 cm, and (c) lower grade primary tumor histology. Patients without good prognostic factors had a median disease-specific survival of 10 months. The authors recommended that reexploration and resection of recurrent pulmonary metastases from sarcoma could be beneficial with improvements in survival. Patients with poor prognostic factors will have worse survival and should be considered for alternative or investigational therapies.

Repeat resection of pulmonary metastasis may salvage a subset of pediatric patients with sarcomatous histologies. These pediatric sarcomas include osteogenic sarcoma, nonrhabdomyosarcomatous soft tissue sarcoma, and Ewing's sarcoma. At the National Cancer Institute, Temeck and co-workers (1998) described 70 patients who underwent at least one reoperation between 1965 and 1995. Osteosarcoma predominated, with 36 patients. Single-wedge resection was the most common operation performed (84%). The authors noted that complete resection was the most important and favorable prognostic factor. Patients with complete resection had improved survival compared with those who were incompletely resected. Median survival was 2.25 years. In resectable patients, median survival was 5.6 years, compared with 0.7 year in unresectable patients (p < 0.0001). From this review, the authors concluded that an aggressive surgical approach in patients with small numbers of lesions, longer DFI, and the ability to obtain a complete resection is warranted and is associated with prolonged survival.

SURVIVAL ANALYSIS

Survival may be absolute or actuarial and is usually calculated from the time the surgical procedure is performed until death or until the date of last follow-up, as noted by Toledano and Olak (1998). For example, patients followed for a minimum of 5 years (survivors) or until death provide an absolute 5-year survival rate (patients alive/all patients studied); patients followed for varying periods of time (i.e., 2 to 7 years) may be evaluated by using an actuarial survival curve. Actuarial survival and disease-free survival may be estimated using the method of Kaplan and Meier (1958). Patients grouped into two or more populations are defined as meeting or not meeting objective criteria and are compared to evaluate differences in survival. Univariate analysis (comparisons between groups) may be made using the generalized wilcoxon test of Gehan (1965) or log-rank

P.1853

PROGNOSTIC INDICATORS

Predictors for improved survival have been studied retrospectively for various tumor types to identify selected patients who will benefit from pulmonary metastasectomy. These prognostic indicators are clinical, biological, and molecular criteria that describe the biological interaction between the metastases and the patient and their association with prolonged survival. These prognostic indicators may be used to identify patients who are most likely to benefit from resection of pulmonary metastases.

Analysis of prognostic indicators in groups of patients with pulmonary metastases from heterogeneous tumors describes prolonged survival in patients with resectable metastases. Resectable patients, longer DFI, longer TDT, fewer numbers of metastases, or solitary metastasis are prognostic indicators generally associated with prolonged postresection survival. Prognostic indicators should be studied in patients with the same primary tumor to define their association with postresection survival. A wide variability exists in the characteristics of pulmonary metastases from different primary neoplasms and the subsequent survival of patients with these metastases. The study of prognostic indicators from the same primary neoplasm yields the most precise information on association with postresection survival. Age and gender do not usually influence postthoracotomy survival and generally should not be considered prognostic factors.

Location and Stage of Primary Tumor

Postresection survival is not usually influenced by the specific anatomic location of the primary tumor. Postresection survival in patients with more advanced primary neoplasms does not usually differ from that in patients with earlier-stage disease. Although initial or primary stage may suggest the biological aggressiveness of the tumor, it has little impact on subsequent survival in patients with isolated pulmonary metastases.

Disease-free Interval

The initial DFI extends from resection of the primary tumor until pulmonary metastases or other metastases are detected. A short DFI may indicate a more virulent tumor with a poor prognosis. Metastases may be multiple and grow rapidly. A longer DFI may represent a less biologically aggressive tumor and correlates with a longer postresection survival. The DFI may also be defined as the time between resection of the pulmonary metastases and recurrence of metastases in the lungs or elsewhere. DFIs of greater than 12 months are usually associated with improved survival. A DFI longer than 36 months was an independent predictor of survival in the International Registry of Lung Metastases, as shown by Pastorino and colleagues (1997).

Number of Nodules on Preoperative Radiographic Studies

High-resolution CT has replaced linear tomograms as the examination of choice in patients with suspected pulmonary metastases. CT of the chest provides a sensitive and specific study for patients with pulmonary metastases. CT of the chest is quite sensitive but less specific than conventional linear tomography or chest radiograph. Nodules may or may not represent metastases. Theoretically, earlier detection and treatment of metastases could improve survival. Laterality (unilateral or bilateral) of pulmonary metastases does not directly influence postresection survival; the number of nodules is a more precise prognostic indicator.

Number of Metastases Resected

The number of metastases resected may be associated with disease-free survival and overall survival. Complete resection is needed. In general, the number of metastases resected exceeds the total number of nodules identified on preoperative radiographic studies. Careful palpation of lung parenchyma will identify more nodules than would be otherwise suspected based on preoperative studies, especially in patients with osteogenic or soft tissue sarcomas. These nodules may be benign or malignant and must be resected in order to have histologic confirmation of the status. Not all nodules on preoperative chest radiographic studies are malignant, as noted by Pogrebniak and associates (1988).

Tumor Doubling Time

TDT is based on original observations by Collins and co-workers (1956), as calculated by Joseph and colleagues (1971a, 1971b), and has been analyzed for multiple tumor types. TDT is calculated by measuring the same metastasis on similar studies (e.g., serial chest radiographs), which are separated by a minimum of 10 to 14 days. The most rapidly growing nodule is selected. The TDT can be easily calculated by plotting changing diameters of pulmonary metastases on semilogarithmic paper; however, graphical error may be present. A mathematical formula may be used to precisely calculate TDT:

P.1854

Errors may occur in the calculation of TDT because not all metastases grow at the same rate. Different growth rates between tumor nodules may reflect heterogeneity of metastases from the primary. The TDT may indirectly reveal the underlying biological nature of the metastases and, as such, influence the patient's postresection survival.

Pulmonary metastases initially grow exponentially, and with increased size, the growth rate diminishes. Gompertz (1832) described growth kinetics [expanded by Laird (1960)], which considered a gradual diminution in TDT with time and increased size of the metastasis. Whether the growth rate is linear, exponential, or gompertzian may be difficult to evaluate because radiographs demonstrate a 3D structure in two dimensions. New techniques such as 3D tumor volume calculations may better describe the TDT in selected patients, as described by Belshi (1997) and Eggli (1995) and their associates. In addition, the growth rate measured over a few weeks represents only a brief period in the lifetime of the metastasis. Although this growth rate is presumed to be linear, it may not always be linear, and TDT only reflects growth during the interval measured.

Resectability

Complete resection consistently correlates with improved postthoracotomy survival for patients with pulmonary metastases. If pulmonary metastases cannot be completely removed, the postthoracotomy survival rate is less than for patients who underwent complete resection.

Endobronchial or Nodal Metastases

Involvement of mediastinal lymph nodes from pulmonary metastases is rare. Udelsman and colleagues (1986) noted that patients with endobronchial metastases from adult soft tissue sarcomas have a short post-resection survival. Seven of 11 patients with endobronchial metastases lived 6 months or less. Jablons and co-workers (1989) found that survival is poor (5 months) in patients with mediastinal lymph node involvement from soft tissue sarcomas compared with that in patients without nodal metastases (31 months).

Multivariate Analysis of Prognostic Factors

Use of multivariate analysis may allow more accurate prediction of postresection survival and better patient selection. Separate prognostic variables may be combined to enhance the predictive value for survival. Jablons and associates (1989) found the DFI, gender, resectability, and truncal location in patients with pulmonary metastases from soft tissue sarcomas to be the best predictors of postthoracotomy survival. The author (JBP) and colleagues (1984) noted that DFI greater than 12 months, TDT greater than 20 days, and four nodules or fewer on preoperative full-lung tomograms as multivariate prognostic indicators were the best predictors of postthoracotomy survival in patients with pulmonary metastases from soft tissue sarcomas. Roth and co-workers (1985) compared prognostic indicators in patients with osteogenic sarcoma and soft tissue sarcoma. The TDT, number of metastases on preoperative full-lung tomograms, and DFI, when combined, improved predictive ability over any single indicator or pair of indicators.

NOVEL TREATMENT STRATEGIES

Thoracic surgeons take full advantage of the unique and fortuitous tumor biology represented by the patient with isolated and resectable pulmonary metastases. Despite better-refined and more aggressive resection techniques, enhanced selection of patients, and evolving multidisciplinary care, only a minority of patients with isolated pulmonary metastases undergo resection. Better therapy must include treatment for macroscopic disease as well as occult or microscopic, disease.

Various strategies have been proposed to treat metastases more effectively isolated to the lung. Systemic (intravenous) chemotherapy has been discussed in the context of specific histology. New and experimental techniques include identification of molecular markers would serve to identify metastases and their organ of origin, as well as their potential responsiveness to systemic chemotherapy. Identification of these characteristics may lead to specific gene replacement strategy. Directed regional drug delivery can be accomplished through various routes, including inhalation or isolated single- or double-lung perfusion. Although potentially limited in its usefulness in the treatment of multiple pulmonary metastases, radiofrequency ablation (RFA) of solitary lesions may be applied in selected patients to preserve lung function.

Molecular Characteristics for Targeted Therapy

Molecular events associated with pulmonary metastases have been identified in patients with osteogenic sarcoma. Amplification of the MDM2gene (the human homologue of a murine p53-binding protein) may regulate p53 protein function by inactivating the protein and deregulating or enhancing tumor growth. In one study, Ladanyi and associates (1993) noted no detectable MDM2 gene amplification in the primary osteogenic sarcoma, compared with its presence

P.1855

Changes in Ki-67 activity are associated with worse survival and tumor progression as shown by Hernandez-Rodriguez and colleagues (2001). They evaluated 38 patients with immunohistochemical analysis. Fifteen of 17 patients who expressed Ki-67 developed pulmonary metastases and had a higher mortality rate. The author recommended using Ki-67 as a prognostic molecular marker for pulmonary metastasis in patients with osteogenic sarcoma.

Reduced expression of KAI1/CD82 expression is associated with poor prognosis and metastasis. Arihiro and Inai (2001) reviewed the role of KAI1 and its relationship to metastases and prognosis. At least 67% of benign bone tumors and 36% of osteogenic sarcomas express this metastasis suppressor gene. Only one of four patients with osteogenic sarcoma metastatic to the lung was positive for KAI1/CD82.

Pollock and colleagues (1998) noted that in soft tissue sarcomas, alterations (mutations of the p53 gene, a tumor-suppressor gene) may provide for uncontrolled cell growth. Restoration of normal p53 ( wild-type ) levels to soft tissue sarcomas may provide more controlled cell growth or even programmed cell death (apoptosis). In one in vitro study, transduction of wild-type p53 into soft tissue sarcomas bearing mutated p53 genes altered the malignant potential of the tumor. After transduction, transfected cells expressed wild-type p53, decreased cell proliferation, and decreased colony formation in soft agar and demonstrated decreased tumor formation in severe combined immunodeficient mice in vivo. The ability to restore wild-type p53 function in soft tissue sarcoma in vitro and in these mice may ultimately be considered as future therapy for patients with soft tissue sarcomas.

Targets for gene transfer may include chemotherapy-resistant tumors or tumors with greater propensity for metastatic spread. Scotlandi and associates (1996) noted that overexpression of the MDR1 gene product P-glycoprotein is an important predictor of poor prognosis in osteosarcoma patients treated with chemotherapy. A rodent model of osteogenic sarcoma has been developed by Asai and associates (1998) with high propensity for pulmonary metastasis. In this metastatic tumor model, matrix metalloproteinase-2 (MMP-2) activity is increased, as well is of vascular endothelial growth factor (VEGF) messenger RNA (mRNA).

Patients with tumors that exhibit overexpression of P-glycoprotein (a multidrug transport protein produced by the MDR1 gene) develop resistance to chemotherapeutic agents. In these patients, the MDR phenotype is not de novo more aggressive (i.e., more metastatic); however, the poor outcome of patients with the MDR phenotype related to P-glycoprotein overexpression is related to the cells' lack of response to cytotoxic drugs. A doxorubicin-sensitivity assay recommended by Kumta and colleagues (2001) may be a better determinant of improved chemotherapy responsiveness and subsequent clinical outcome than P-glycoprotein. One potential mechanism of resistance to doxorubicin may be in the ability to regulate MDR1 RNA levels after administration of doxorubicin, as shown by Abolhoda and coinvestigators (1999).

Attenuation of insulinlike growth factor (IGF) activation may also improve survival, as proposed by Beech and colleagues (2003). Activation of the IGF-1 receptor decreases systemic response to doxorubicin. In vitro studies demonstrated that activation of the IFG-1 receptor enhanced tumor resistance to doxorubicin. Inhibition of IGF-1 receptor activation may be an effective adjunct to conventional chemotherapy.

Increased expression of ErbB-2 or gene amplification has been associated with poor survival in patients with osteogenic sarcoma, as shown by Onda and colleagues (1996). In this study, 42% of patients with osteogenic sarcoma had metastases that expressed erbB-2 and correlated with early development of pulmonary metastasis and poor survival. ErbB-2, therefore, may enhance tumor growth and promote metastases. These authors recommended that erbB-2 be considered a prognostic factor for patients with osteosarcoma. Murine models for suicide gene therapy treatment of osteogenic sarcoma cells have been developed by Seto and colleagues (2002).

Differences between the primary tumor and the metastases from osteosarcoma patients may exist. Akatsuka and colleagues (2002) examined the relationship between Erb-B2 and survival in patients with osteogenic sarcoma. Eighty-one patients with osteogenic sarcoma in the extremities were treated with resection and chemotherapy; 61% of the patients had high levels of expression of Erb-B2. Patients with higher levels of expression had improved disease-free survival and overall survival. Patients with decreased levels had a worse prognosis.

Other gene products have been proposed for transfer into pulmonary metastases. Benjamin and colleagues (2001) have proposed treatment of pulmonary metastases using bolus intravenous injections of Ad-OC-E1 in patients who had exhausted all chemotherapy regimens of greater effectiveness. The adenoviral vector Ad-OC-E1a (OCaP1) contains a murine osteocalcin (OC) promoter to regulate the production of the adenoviral E1a protein.

Lung Perfusion

Novel drug delivery systems may enhance chemotherapy treatment effects by increasing drug concentration in lung tissues and minimizing systemic effects and toxicity of such treatment, as noted by Van Schil (2002) and the author and associates (2002). In many patients, surgery has been used as sole therapy or as salvage treatment after maximal chemotherapy response has been achieved. Systemic toxicity may limit the amount of chemotherapy given to an individual patient. Regional drug delivery to the lungs minimizes

P.1856

Preclinical studies by Weksler and coinvestigators (1993, 1994) in rodents with experimental pulmonary metastases from a methylcholanthrene-induced syngeneic sarcoma have shown that chemotherapy may be delivered regionally to pulmonary tissue in significantly higher concentrations than by systemic delivery. Minimal to no systemic toxicity was noted. In this model, isolated single-lung perfusion with doxorubicin was safe and effective. This simple microsurgical technique was performed in rats. After left thoracotomy, the pulmonary artery and pulmonary vein were isolated and clamped. The lung was flushed before infusing doxorubicin. The infusion occurred over 10 minutes. Then, the drug was flushed out before removing the cannulas and restoring circulation (Fig. 120-7). A perfusion concentration of 255 mg/L caused less general toxicity than a systemic dose equivalent to 75 mg/m². The extraction ratio was 58%, and pulmonary tissue concentration of doxorubicin was 25-fold higher than with the systemic dose. The technique was also effective. Nine of 10 animals treated at 320 mg/L had complete eradication of metastases from an implanted methylcholanthrene-induced sarcoma. Other chemotherapeutic agents have been used, including gemcitabine by Van Putte and associates (2003), paclitaxel by Schrump and co-workers (2002), and melphalan by Van Putte and colleagues (2002). Brooks and coinvestigators (2001) described effective in vivo delivery of gene products using a herpes viral factor. They confirm that such treatment in a rat model of sarcomatous pulmonary metastasis would reduce tumor burden.

Alternative techniques in animal models of chemoembolization of the lung with carboplatin and degradable starch microspheres have been studied by Schneider and colleagues (2002). These degradable microspheres allow for higher concentrations to the lung and lung tissue during the degradation phase of the treatment.

Previous clinical studies of lung perfusion by Pass (1996) and Johnston (1995) and their associates have shown higher drug concentrations in pulmonary tissue, although clinical tumor response has been mixed. Johnston and colleagues (1995) described a continuous perfusion of the lungs with Adriamycin (single lung, continuous perfusion) as a safe technique and subsequently applied their technique clinically. Drug concentrations in normal lung and tumor generally increased with higher drug dosages. Two of eight patients had major complications: one patient developed pneumonia and sternal dehiscence; one patient developed respiratory failure 4 days after lung perfusion. No objective responses occurred (none of four patients with sarcomas). Continuous perfusion with a pump circuit offers some theoretical advantages but may be mechanically complex. Pass and co-workers (1996) examined isolated single-lung perfusion with tumor necrosis factor- , interferon- , and moderate hyperthermia for patients with unresectable pulmonary metastases. No hospital deaths occurred, and a short-term (<6-month) decrease in nodule size was noted in 3 of 15 patients.

Other small clinical studies have recently been reported. Schroder and colleagues (2002) reported on four patients with sarcoma metastases treated with hyperthermia (41 C) isolated lung perfusion with cisplatin (70 mg/m²). There was no systemic toxicity noted. One patient developed noncardiogenic edema of the lung, and the systemic cisplatin levels were continuously low. In a phase one study, Burt and colleagues (2000) examined the role of isolated lung perfusion with a dose escalation of doxorubicin. The authors noted that drug concentrations in the treated lung correlated with the drug delivered. There was no cardiac or systemic toxicity. There were minimal or undetectable blood levels of the drug. One patient had no ventilation or perfusion in the lung after receiving 80 mg/m² of doxorubicin. No partial or complete responses were noted, although one patient developed stable disease for a period of time. The authors recommended a maximum power to dose of 40 mg/m² in this isolated lung perfusion model. Similar findings were noted by the author and colleagues (2000b).

Inhalational Therapy

Inhalational therapy or inhalational administration of chemotherapy, gene products, or other biological compounds [e.g., interleukin-2 (IL-2)] has been actively investigated

P.1857

Biological Modifiers

Liposome-encapsulated muramyl tripeptide (L-MTP-E) activates macrophages to become tumoricidal. This strategy may be valuable in patients with chemoresistant tumors. After treatment, plasma levels of cytokines are increased. Monocyte-mediated cytotoxicity is increased. L-MTP-E therapy may be therapeutically effective, as shown by Kleinerman and associates (1995a, 1995b).

Investigations of DOTAP:cholesterol liposomes, protamine sulfate, and plasmid DNA construct (LPD) by Whitmore and colleagues (1999) have shown increased antitumor activity by the lipopolyplex alone as well as the LPD. The mechanism of action suggests a systemic proinflammatory cytokine response.

Radiofrequency Ablation

Radiofrequency ablation provides controlled thermal destruction of tumors or other tissue and has been examined in several recent reviews. Radiofrequency ablation has been applied to malignant liver tumors as described by Curley (2003), the kidney by Finell and associates (2003) as well as by Matin (2003), and the lung by Dupuy and colleagues (2002).

The safe and effective use of interstitial thermal therapy (radiofrequency ablation) for treatment of lung neoplasms was examined in a preclinical model that reproducibly created thermal lesions in normal lung parenchyma. The author and coinvestigators (2000a) noted that these lesions were affected by conductive heat loss through air, blood flow, and bronchi. Use of radiofrequency ablation techniques is an experimental technique for local control of pulmonary metastasis. Limitations of radiofrequency ablation for pulmonary metastases include potential injury to vascular or bronchial structures, generation of heat that must be dissipated, and failure to control or ablate all viable tumor, as described by Steinke and colleagues (2003).

One small phase 1 study of radiofrequency ablation of primary and metastatic lung tumors has been reported by Yang and colleagues (2002). Pulmonary metastases have variable consistency and may be difficult to penetrate with the large needle necessary to carry the ablation tines. This technique remains experimental at this time. Complications included massive hemorrhage, as described by Vaughn and associates in 2002.

CONCLUSIONS

Surgery for treatment of pulmonary metastases will not benefit a significant number of patients. Surgery attempts to control mechanically the biological sequelae of the primary malignancy. Isolated and resectable metastases to the lungs represent a unique biology among the host, the primary neoplasm, and the metastases. Cure in most patients represents a serendipitous occasion in which the host biology, spread of tumor, and surgical resection remove all tumor, including micrometastases.

Patients, however, typically have long-term survival associated with complete resection of all pulmonary metastases. Is the associated long-term survival the result of surgery or the result of the unique biology of the tumor and its metastases? This question remains unanswered. Complete resection remains the crucial characteristic associated with long-term survival, regardless of the primary tumor histology. Patients with resectable pulmonary metastases should undergo resection to render them disease free with the potential for long-term survival and cure.

Various prognostic indicators may define the biological nature of the metastases, predict postresection survival, and assist the clinician in selecting patients who will benefit from surgery. No single criterion in resectable patients consistently and reliably predicts which patients will have enhanced long-term survival after resection; however, unresectable patients do poorly, despite adjuvant therapy. Current molecular biological techniques may improve the ability to select patients for surgery or other treatment based on observations of survival in patients with certain molecular characteristics. Use of additional adjuvant therapy may allow for prolonged survival or cure.

The fundamental biology of the neoplastic and metastatic process is unchanged by surgery. Other novel therapies, including identification of molecular characteristics for targeted therapy such as gene transfer or regional delivery of drugs, may provide more effective therapy for the patients. The best results of treatment of pulmonary metastases await

P.1858

REFERENCES

Abolhoda A, et al: Activation of MDR1 gene expression in human metastatic sarcoma after in vivo exposure to doxorubicin. Clin Cancer Res 5:3352, 1999.

Akatsuka T, et al: ErbB2 expression is correlated with increased survival of patients with osteosarcoma. Cancer 94:1397, 2002.

Alexander J, Haight C: Pulmonary resection for solitary metastatic sarcoma and carcinoma. Surg Gynecol Obstet 85:129, 1947.

Allen PJ, Coit DG: The role of surgery for patients with metastatic melanoma. Curr Opin Oncol 14:221, 2002.

Anderson TM, et al: Pulmonary resection in metastatic uterine and cervical malignancies. Gynecol Oncol 83:472, 2001.

Ang KL, et al: Intrapleural tumor dissemination after video-assisted thoracoscopic surgery metastasectomy. Ann Thorac Surg 75:1643, 2003.

Arihiro K, Inai K: Loss of KAI1/CD82 expression in bone and soft tissue tumors is not associated with lung metastasis. Pathol Res Pract 197:627, 2001.

Asai T, et al: Establishment and characterization of a murine osteosarcoma cell line (LM8) with high metastatic potential to the lung. Int J Cancer 76:418, 1998.

Bacci G, et al: Metachronous pulmonary metastases resection in patients with Ewing's sarcoma initially treated with adjuvant or neoadjuvant chemotherapy. Eur J Cancer 31A:999, 1995.

Bacci G, et al: Osteogenic sarcoma of the extremity with detectable lung metastases at presentation. Results of treatment of 23 patients with chemotherapy followed by simultaneous resection of primary and metastatic lesions. Cancer 79:245, 1997.

Bacci G, et al: Pattern of relapse in patients with osteosarcoma of the extremities treated with neoadjuvant chemotherapy. Eur J Cancer 37:32, 2001.

Bains MS, et al: The clamshell incision: an improved approach to bilateral pulmonary and mediastinal tumor. Ann Thorac Surg 58:30, 1994.

Barbareschi M, et al: CDX-2 homeobox gene expression is a reliable marker of colorectal adenocarcinoma metastases to the lungs. Am J Surg Pathol 27:141, 2003.

Barney JD, Churchill EJ: Adenocarcinoma of the kidney with metastasis to the lung cured by nephrectomy and lobectomy. J Urol 42:269, 1939.

Barter JF, et al: Diagnosis and treatment of pulmonary metastases from cervical carcinoma. Gynecol Oncol 38:347, 1990.

Barth A, Wanek LA, Morton DL: Prognostic factors in 1,521 melanoma patients with distant metastases. J Am Coll Surg 181:193, 1995.

Bathe OF, et al: Metastasectomy as a cytoreductive strategy for treatment of isolated pulmonary and hepatic metastases from breast cancer. Surg Oncol 8:35, 1999.

Beech DJ, et al: Insulin-like growth factor-I receptor activation blocks doxorubicin cytotoxicity in sarcoma cells. Oncol Rep 10:181, 2003.

Belal A, et al: Pulmonary metastatectomy for soft tissue sarcomas: is it valuable? J Cardiovasc Surg 42:835, 2001.

Belli L, et al: Resection of pulmonary metastases in osteosarcoma. A retrospective analysis of 44 patients. Cancer 63:2546, 1989.

Belshi R, et al: Automatic three-dimensional expansion of structures applied to determination of the clinical target volume in conformal radiotherapy. Int J Radiat Oncol Biol Phys 37:689, 1997.

Benjamin R, et al: A phase I/II dose escalation and activity study of intravenous injections of OCaP1 for subjects with refractory osteosarcoma metastatic to lung. Hum Gene Ther 12:1591, 2001.

Benjamin RS, Wiernik P, Bachur N: Adriamycin: a new effective agent in the therapy of disseminated sarcomas. Med Pediatr Oncol 1:63, 1975.

Billingsley KG, et al: Multifactorial analysis of the survival of patients with distant metastasis arising from primary extremity sarcoma. Cancer 85:389, 1999.

Bocklage T, et al: Extracutaneous angiosarcomas metastatic to the lungs: clinical and pathologic features of twenty-one cases. Mod Pathol 14: 1216, 2001.

Bodzin GA, Staren ED, Faber LP: Breast carcinoma metastases. Chest Surg Clin N Am 8:145, 1998.

Bouros D, et al: Natural history of patients with pulmonary metastases from uterine cancer. Cancer 78:441, 1996.

Branscheid D, et al: Does ND-YAG laser extend the indications for resection of pulmonary metastases? Eur J Cardiothorac Surg 6:590, 1992.

Brooks AD, et al: Specific organ gene transfer in vivo by regional organ perfusion with herpes viral amplicon vectors: implications for local gene therapy. Surgery 129:324, 2001.

Burgers JM, et al: Osteosarcoma of the limbs. Report of the EORTC-SIOP 03 trial 20781 investigating the value of adjuvant treatment with chemotherapy and/or prophylactic lung irradiation. Cancer 61:1024, 1988.

Burt ME, et al: Isolated lung perfusion for patients with unresectable metastases from sarcoma: a phase I trial. Ann Thorac Surg 69:1542, 2000.

Carsky S, et al: Germ cell testicular tumours with lung metastases: chemotherapy and surgical treatment. Int Urol Nephrol 24:305, 1992.

Carter GE, et al: Reassessment of the role of adjunctive surgical therapy in the treatment of advanced germ cell tumors. J Urol 138:1397, 1987.

Carter SR, et al: Results of thoracotomy in osteogenic sarcoma with pulmonary metastases. Thorax 46:727, 1991.

Casson AG, et al: Efficacy of pulmonary metastasectomy for recurrent soft tissue sarcoma. J Surg Oncol 47:1, 1991.

Casson AG, et al: Five-year survival after pulmonary metastasectomy for adult soft tissue sarcoma. Cancer 69:662, 1992.

Chauveinc L, et al: Uterine sarcomas: the Curie Institut experience. Prognosis factors and adjuvant treatments. Gynecol Oncol 72:232, 1999.

Collins VP, Loeffler RK, Tivey H: Observations on growth rates of human tumors. AJR Am J Roentgenol 76:988, 1956.

Cox DR: Regression models and life-tables. J R Stat Soc B 34:187, 1972.

Curley SA: Radiofrequency ablation of malignant liver tumors. Ann Surg Oncol 10:338, 2003.

Czauderna P, et al: Liver Tumors Study Group of the International Society of Pediatric Oncology: Hepatocellular carcinoma in children: results of the first prospective study of the International Society of Pediatric Oncology group. J Clin Oncol 20:2798, 2002.

d'Amato T, et al: Bilateral anterior minithoracotomy with video assistance for lung volume reduction surgery and pulmonary metastasectomy. Surg Endosc 16:364, 2002.

Dalrymple-Hay MJ, et al: Pulmonary metastatic melanoma the survival benefit associated with positron emission tomography scanning. Eur J Cardiothorac Surg 21:611, 2002.

de Kraker J, et al: Wilms' tumor with pulmonary metastases at diagnosis: the significance of primary chemotherapy. International Society of Pediatric Oncology Nephroblastoma Trial and Study Committee. J Clin Oncol 8:1187, 1990.

Dhaini HR, et al: Cytochrome P450 CYP3A4/5 expression as a biomarker of outcome in osteosarcoma. J Clin Oncol 21:2481, 2003.

Dickens DS, et al: Cyclooxygenase-2 expression does not correlate with outcome in osteosarcoma or rhabdomyosarcoma. J Pediatr Hematol Oncol 25:282, 2003.

Di Lorenzo M, Collin PP: Pulmonary metastases in children: results of surgical treatment. J Pediatr Surg 23:762, 1988.

Divis G: Ein Beitrag zur operativen Behandlung der Lungengeschwultse. Acta Chir Scand 62:329, 1927.

Downey RJ, McCormack P, LoCicero J 3rd: Dissemination of malignant tumors after video-assisted thoracic surgery: a report of twenty-one cases. The Video-Assisted Thoracic Surgery Study Group. J Thorac Cardiovasc Surg 111:954, 1996.

Dunst J, Paulussen M, Jurgens H: Lung irradiation for Ewing's sarcoma with pulmonary metastases at diagnosis: results of the CESS-studies. Strahlenther Onkol 169:621, 1993.

Dupuy DE, et al: Clinical applications of radio-frequency tumor ablation in the thorax. Radiographics 22:S259, 2002.

Eggli KD, et al: Three-dimensional quantitation of pediatric tumor bulk. Pediatr Radiol 25:1, 1995.

Elias A, et al: Response to mesna, doxorubicin, ifosfamide, and dacarbazine in 108 patients with metastatic or unresectable sarcoma and no prior chemotherapy. J Clin Oncol 7:1208, 1989.

Enk AH, et al: High dose inhalation interleukin-2 therapy for lung metastases in patients with malignant melanoma. Cancer 88:2042, 2000.

Feigenberg SJ, et al: Whole-lung radiotherapy for giant cell tumors of bone with pulmonary metastases. Clin Orthop 401:202, 2002.

Ferguson WS, et al: Presurgical window of carboplatin and surgery and multidrug chemotherapy for the treatment of newly diagnosed metastatic or unresectable osteosarcoma: Pediatric Oncology Group Trial. J Pediatr Hematol Oncol 23:340, 2001.

Ferrari S, et al: Postrelapse survival in osteosarcoma of the extremities: prognostic factors for long-term survival. J Clin Oncol 21:710, 2003.

P.1859

Feuerstein IM, et al: Pulmonary metastases: MR imaging with surgical correlation a prospective study. Radiology 182:123, 1992.

Finell A, Rewcastle JC, Jewett MA: Cryotherapy and radiofrequency ablation: pathophysiologic basis and laboratory studies. Curr Opin Urol 13:187, 2003.

Finley RK 3d, et al: Results of surgical resection of pulmonary metastases of squamous cell carcinoma of the head and neck. Am J Surg 164:594, 1992.

Fischer CG, Schmid H: Operative therapy in disease progression and local recurrence of renal cell carcinoma. Urol Int 63:10, 1999.

Fleming JB, et al: Utility of chest computed tomography for staging in patients with T1 extremity soft tissue sarcomas. Cancer 92:863, 2001.

Flint A, Lloyd RV: Colon carcinoma metastatic to the lung. Cytologic manifestations and distinction from primary pulmonary adenocarcinoma. Acta Cytol 36:230, 1992a.

Flint A, Lloyd RV: Pulmonary metastases of colonic carcinoma. Distinction from pulmonary adenocarcinoma. Arch Pathol Lab Med 116:39, 1992b.

Fourquier P, et al: Lung metastases of renal cell carcinoma: results of surgical resection. Eur J Cardiothoracic Surg 11:17, 1997.

Franzius C, et al: FDG-PET for detection of pulmonary metastases from malignant primary bone tumors: comparison with spiral CT. Ann Oncol 12:479, 2001.

Friedel G, Linder A, Toomes H: The significance of prognostic factors for the resection of pulmonary metastases of breast cancer. Thorac Cardiovasc Surg 42:71, 1994.

Friedel G, et al: Resection of pulmonary metastases from renal cell carcinoma. Anticancer Res 19:1593, 1999.

Friedel G, et al: International Registry of Lung Metastases, London England: Results of lung metastasectomy from breast cancer: prognostic criteria on the basis of 467 cases of the International Registry of Lung Metastases. Eur J Cardiothorac Surg 22:335, 2002.

Fuller AF Jr, Scannell JG, Wilkins EW Jr: Pulmonary resection for metastases from gynecologic cancers: Massachusetts General Hospital experience, 1943 1982. Gynecol Oncol 22:174, 1985.

Gehan EA: A generalized Wilcoxon test for comparing arbitrarily singly-censored samples. Biometrika 522:203, 1965.

Gels ME, et al: Thoracotomy for postchemotherapy resection of pulmonary residual tumor mass in patients with nonseminomatous testicular germ cell tumors: aggressive surgical resection is justified. Chest 112: 967, 1997.

Ghoneim AH, et al: Monoclonal anti-CEA antibodies in the discrimination between primary pulmonary adenocarcinoma and colon carcinoma metastatic to the lung. Mod Pathol 3:613, 1990.

Girard P, et al: Surgical resection of pulmonary metastases. Up to what number? Am J Respir Crit Care Med 149:469, 1994.

Glasser DB, et al: Survival, prognosis, and therapeutic response in osteogenic sarcoma. The Memorial Hospital experience. Cancer 69:698, 1992.

Godzinski J, et al: The role of preoperative chemotherapy in the treatment of nephroblastoma: the SIOP experience. Societe Internationale d'Oncologie Pediatrique. Semin Urol Oncol 17:28, 1999.

Goi T, et al: Evaluation of clinicopathological factors and the correlation between the adhesion molecule CD44 variant 9 expression and pulmonary metastases from colorectal cancers. Int Surg 87:130, 2002.

Gompertz B: On the nature of the function expressive of the law of human mortality, and on a new mode of determining the value of life contingencies. Phil Trans Roy Soc London 123:513, 1832.

Goorin AM, et al: Changing pattern of pulmonary metastases with adjuvant chemotherapy in patients with osteosarcoma: results from the multiinstitutional osteosarcoma study. J Clin Oncol 9:600, 1991.

Goorin AM, et al: Phase II/III trial of etoposide and high-dose ifosfamide in newly diagnosed metastatic osteosarcoma: a pediatric oncology group trial. J Clin Oncol 20:426, 2002.

Goorin AM, et al: Pediatric Oncology Group: presurgical chemotherapy compared with immediate surgery and adjuvant chemotherapy for nonmetastatic osteosarcoma. Pediatric Oncology Group Study POG-8651. J Clin Oncol 21:1574, 2003.

Gorenstein LA, et al: Improved survival after resection of pulmonary metastases from malignant melanoma. Ann Thorac Surg 52:204, 1991.

Green DM, et al: The role of surgical excision in the management of relapsed Wilms' tumor patients with pulmonary metastases: a report from the National Wilms' Tumor Study. J Pediatr Surg 26:728, 1991.

Groeger AM, et al: Survival after surgical treatment of recurrent pulmonary metastases. Eur J Cardiothorac Surg 12:703, 1997.

Grunenwald D, et al: Completion pneumonectomy for lung metastases: is it justified? Eur J Cardiothorac Surg 12:694, 1997.

Harpole DH, et al: Analysis of 945 cases of pulmonary metastatic melanoma. J Thorac Cardiovasc Surg 103:743, 1992.

Heij HA, et al: Prognostic factors in surgery for pulmonary metastases in children. Surgery 115:687, 1994.

Hernandez-Rodriguez NA, et al: Ki-67: a proliferative marker that may predict pulmonary metastases and mortality of primary osteosarcoma. Cancer Detect Prev 25:210, 2001.

Herrera GA, Alexander CB, Jones JM: Ultrastructural characterization of pulmonary neoplasms. II. The role of electron microscopy in characterization of uncommon epithelial pulmonary neoplasms, metastatic neoplasms to and from lung, and other tumors, including mesenchymal neoplasms. Surv Synth Pathol Res 4:163, 1985.

Hershey AE, et al: Inhalation chemotherapy for macroscopic primary or metastatic lung tumors: proof of principle using dogs with spontaneously occurring tumors as a model. Clin Cancer Res 5:2653, 1999.

Heslin MJ, et al: Preoperative identification and operative management of intraatrial extension of lung tumors. Ann Thorac Surg 65:544, 1998.

Higashiyama M, et al: Intraoperative lavage cytologic analysis of surgical margins as a predictor of local recurrence in pulmonary metastasectomy. Arch Surg 137:469, 2002.

Higashiyama M, et al: Surgery for pulmonary metastases from colorectal cancer: the importance of prethoracotomy serum carcinoembryonic antigen as an indicator of prognosis. Jpn J Thorac Cardiovasc Surg 51:289, 2003.

Hirano J, et al: Prognosis for the cases after resection of pulmonary metastasis of osteosarcoma. Kyobu Geka 56:4, 2003.

Hirota T, et al: Combined multimodal therapy for osteosarcoma neoadjuvant chemotherapy. Gan To Kagaku Ryoho 26:1068, 1999.

Hofmann U, et al: Primary staging and follow-up in melanoma patients monocenter evaluation of methods, costs and patient survival. Br J Cancer 87:151, 2002.

Hoos A, Lewis JJ, Brennan MF: Soft tissue sarcoma: prognostic factors and multimodal treatment. Chirurg 71:787, 2000.

Huland E, et al: Overview of interleukin-2 inhalation therapy. Cancer J Sci Am 6:12, 2000.

Hung GU, et al: Value of 18F-fluoro-2-deoxyglucose positron emission tomography in the evaluation of recurrent colorectal cancer. Anticancer Res 21:1375, 2001.

Huth JF, Eilber FR: Patterns of recurrence after resection of osteosarcoma of the extremity. Arch Surg 124:122, 1989.

Ike H, et al: Results of aggressive resection of lung metastases from colorectal carcinoma detected by intensive follow-up. Dis Colon Rectum 45:468, 2002.

Imachi M, et al: Pulmonary metastasis from carcinoma of the uterine cervix. Gynecol Oncol 33:189, 1989.

Inoue M, et al: Surgery for pulmonary metastases from colorectal carcinoma. Ann Thorac Surg 70:380, 2000.

Jablons D, et al: Metastasectomy for soft tissue sarcoma. Further evidence for efficacy and prognostic indicators. J Thorac Cardiovasc Surg 97: 695, 1989.

Jaffe N, et al: Osteogenic sarcoma: alterations in the pattern of pulmonary metastases with adjuvant chemotherapy. J Clin Oncol 1:251, 1983.

Jaffe N, et al: Can cure in patients with osteosarcoma be achieved exclusively with chemotherapy and abrogation of surgery? Cancer 95:2202, 2002.

Jia SF, et al: Eradication of osteosarcoma lung metastases following intranasal interleukin-12 gene therapy using a nonviral polyethylenimine vector. Cancer Gene Ther 9:260, 2002.

Johnston MR: Median sternotomy for resection of pulmonary metastases. J Thorac Cardiovasc Surg 85:516, 1983.

Johnston MR, Minchen RF, Dawson CA: Lung perfusion with chemotherapy in patients with unresectable metastatic sarcoma to the lung or diffuse bronchioloalveolar carcinoma. J Thorac Cardiovasc Surg 110:368, 1995.

Joseph WL, Morton DL, Adkins PC: Prognostic significance of tumor doubling time in evaluating operability in pulmonary metastatic disease. J Thorac Cardiovasc Surg 61:23, 1971a.

Joseph WL, Morton DL, Adkins PC: Variation in tumor doubling time in patients with pulmonary metastatic disease. J Surg Oncol 3:143, 1971b.

Kandioler D, et al: Molecular genetic differentiation between primary lung cancers and lung metastases of other tumors. J Thorac Cardiovasc Surg 111:827, 1996.

Kandioler D, et al: Long-term results after repeated surgical removal of pulmonary metastases. Ann Thorac Surg 65:909, 1998.

P.1860

Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457, 1958.

Karnak I, et al: Pulmonary metastases in children: an analysis of surgical spectrum. Eur J Pediatr Surg 12:151, 2002.

Kaste SC, et al: Metastases detected at the time of diagnosis of primary pediatric extremity osteosarcoma at diagnosis: imaging features. Cancer 86:1602, 1999.

Katzenstein HM, et al: Treatment of unresectable and metastatic hepatoblastoma: a pediatric oncology group phase II study. J Clin Oncol 20: 3438, 2002.

Kleinerman ES: Biologic therapy for osteosarcoma using liposome-encapsulated muramyl tripeptide. Hematol Oncol Clin North Am 9:927, 1995a.

Kleinerman ES, et al: Efficacy of liposomal muramyl tripeptide (CGP 19835A) in the treatment of relapsed osteosarcoma. Am J Clin Oncol 18:93, 1995b.

Kodama K, et al: Surgical management of lung metastases. Usefulness of resection with the neodymium:yttrium-aluminum-garnet laser with median sternotomy. J Thorac Cardiovasc Surg 101:901, 1991.

Koong HN, Pastorino U, Ginsberg RJ: Is there a role for pneumonectomy in pulmonary metastases? International Registry of Lung Metastases. Ann Thorac Surg 68:2039, 1999.

Koshkina NV, et al: Paclitaxel liposome aerosol treatment induces inhibition of pulmonary metastases in murine renal carcinoma model. Clin Cancer Res 7:3258, 2001.

Kronlein RU: Ueber Lungenchirirugie. Berlin Klin Wschr 9:129, 1884.

Kulkarni RP, et al: Cytoreductive surgery in disseminated non-seminomatous germ cell tumours of testis. Br J Surg 78:226, 1991.

Kumar J, Ilancheran A, Ratnam SS: Pulmonary metastases in gestational trophoblastic disease: a review of 97 cases. Br J Obstet Gynaecol 95:70, 1988.

Kumta SM, et al: Clinical significance of P-glycoprotein immunohistochemistry and doxorubicin binding assay in patients with osteosarcoma. Int Orthop 25:279, 2001.

Labow DM, et al: Isolated pulmonary recurrence after resection of colorectal hepatic metastases is resection indicated? Cancer J 8:342, 2002.

Ladanyi M, et al: MDM2 gene amplification in metastatic osteosarcoma. Cancer Res 53:16, 1993.

Laird AK: Dynamics of tumor growth. Br J Cancer 18:490, 1960.

Landreneau RJ, et al: Neodymium:yttrium-aluminum garnet laser-assisted pulmonary resections. Ann Thorac Surg 51:973, 1991.

Landreneau RJ, et al: Thoracoscopic resection of 85 pulmonary lesions. Ann Thorac Surg 54:415, 1992.

Landreneau RJ, et al: Therapeutic video-assisted thoracoscopic surgical resection of colorectal pulmonary metastases. Eur J Cardiothorac Surg 18:671, 2000.

Lanza LA, et al: The role of resection in the treatment of pulmonary metastases from Ewing's sarcoma. J Thorac Cardiovasc Surg 94:181, 1987.

Lanza LA, et al: Response to chemotherapy does not predict survival after resection of sarcomatous pulmonary metastases. Ann Thorac Surg 51: 219, 1991.

Lanza LA, et al: Long-term survival after resection of pulmonary metastases from carcinoma of the breast. Ann Thorac Surg 54:244, 1992.

La Quaglia MP: Osteosarcoma. Specific tumor management and results. Chest Surg Clin N Am 8:77, 1998.

Lauenstein TC, et al: Three-dimensional volumetric interpolated breath-hold MR imaging for whole-body tumor staging in less than 15 minutes: a feasibility study. AJR Am J Roentgenol 179:445, 2002.

Lefor AT, et al: Multiple malignancies of the lung and head and neck. Second primary tumor or metastasis? Arch Surg 121:265, 1986.

Leitao MM, et al: Surgical resection of pulmonary and extrapulmonary recurrences of uterine leiomyosarcoma. Gynecol Oncol 87:287, 2002.

Leong PP, et al: Distinguishing second primary tumors from lung metastases in patients with head and neck squamous cell carcinoma. J Natl Cancer Inst 90:972, 1998.

Levenback C, et al: Resection of pulmonary metastases from uterine sarcomas. Gynecol Oncol 45:202, 1992.

Lewis CW Jr, Harpole D: Pulmonary metastasectomy for metastatic malignant melanoma. Semin Thorac Cardiovasc Surg 14:45, 2002.

Lien HH, et al: Computed tomography and conventional radiography in intrathoracic metastases from non-seminomatous testicular tumor. Acta Radiol 29:547, 1988.

Lin JC, et al: Diagnostic and therapeutic video-assisted thoracic surgery resection of pulmonary metastases. Surgery 126:636, 1999.

Liu D, et al: Pulmonary metastasectomy for testicular germ cell tumors: a 28-year experience. Ann Thorac Surg 66:1709, 1998.

Lonneux M, et al: FDG-PET improves the staging and selection of patients with recurrent colorectal cancer. Eur J Nucl Med Mol Imaging 29:915, 2002.

Lucas JD, et al: Evaluation of fluorodeoxyglucose positron emission tomography in the management of soft-tissue sarcomas. J Bone Joint Surg Br 80:441, 1998.

Macklis RM, Oltikar A, Sallan SE: Wilms' tumor patients with pulmonary metastases. Int J Radiat Oncol Biol Phys 21:1187, 1991.

Mannix EP: Resection of multiple pulmonary metastases fourteen years after amputation for osteochondroma of tibia. Apparent freedom from recurrence three years later. J Thorac Surg 26:544, 1953.

Margaritora S, et al: Staged axillary thoracotomy for bilateral lung metastases: an effective and minimally invasive approach. Eur J Cardiothorac Surg 16[Suppl 1]:S37, 1999.

Margaritora S, et al: Pulmonary metastases: can accurate radiological evaluation avoid thoracotomic approach? Eur J Cardiothorac Surg 21:1111, 2002.

Marina NM, et al: Improved prognosis of children with osteosarcoma metastatic to the lung(s) at the time of diagnosis. Cancer 70:2722, 1992.

Martini N, McCormack PM: Evolution of the surgical management of pulmonary metastases. Chest Surg Clin N Am 8:13, 1998.

Martini N, et al: pulmonary resections in the treatment of osteogenic sarcoma. Ann Thorac Surg 12:271, 1971.

Matin SF: Laparoscopic approaches to urologic malignancies. Curr Treat Options Oncol 4:373, 2003.

McAfee MK, et al: Colorectal lung metastases: results of surgical excision. Ann Thorac Surg 53:780, 1992.

McCormack P, et al: Pulmonary resection in metastatic carcinoma. Chest 73:163, 1978.

McCormack PM, et al: Lung resection for colorectal metastases: 10-year results. Arch Surg 127:1403, 1992.

McCormack PM, et al: Accuracy of lung imaging in metastases with implications for the role of thoracoscopy. Ann Thorac Surg 56:863, 1993.

McCormack PM, et al: Role of video-assisted thoracic surgery in the treatment of pulmonary metastases: results of a prospective trial. Ann Thorac Surg 62:213, 1996.

Meade RH: A History of Thoracic Surgery. Springfield, IL: Charles C. Thomas, 1961.

Meyer WH, et al: Thoracotomy for pulmonary metastatic osteosarcoma. An analysis of prognostic indicators of survival. Cancer 59:374, 1987.

Mineo TC, et al: The value of the Nd:YAG laser for the surgery of lung metastases in a randomized trial. Chest 113:1402, 1998.

Mineo TC, et al: Transxiphoid bilateral palpation in video-assisted thoracoscopic lung metastasectomy. Arch Surg 136:783, 2001.

Miniero R, et al: Feasibility of high-dose chemotherapy and autologous peripheral blood stem cell transplantation in children with high grade osteosarcoma. Bone Marrow Transplant 22[Suppl 5]:S37, 1998.

Miyamoto H, et al: Application of the Nd-YAG laser for surgical resection of pulmonary metastases. Kyobu Geka 45:56, 1992.

Mountain CF, McMurtrey MJ, Hermes KE: Surgery for pulmonary metastasis: a 20-year experience. Ann Thorac Surg 38:323, 1984.

Murata S, et al: Resection of both hepatic and pulmonary metastases in patients with colorectal carcinoma. Cancer 83:1086, 1998.

Nagakura S, et al: Simultaneous detection of colorectal carcinoma liver and lung metastases does not warrant resection. J Am Coll Surg 193: 153, 2001.

Nakajima J, et al: Thoracoscopic surgery and conventional open thoracotomy in metastatic lung cancer. Surg Endosc 15:849, 2001.

Nibu K, et al: Surgical treatment for pulmonary metastases of squamous cell carcinoma of the head and neck. Am J Otolaryngol 18:391, 1997.

Nishimura S, et al: High-dose chemotherapy in children with metastatic hepatoblastoma. Pediatr Int 44:300, 2002.

Niwa K, et al: Complete remission of uterine endometrial cancer with multiple lung metastases treated by paclitaxel and carboplatin. Int J Clin Oncol 7:197, 2002.

Nomori H, et al: Tumor cell heterogeneity and subpopulations with metastatic ability in differentiated adenocarcinoma of the lung. Histologic and cytofluorometric DNA analyses. Chest 99:934, 1991.

Ollila DW, Morton DL: Surgical resection as the treatment of choice for melanoma metastatic to the lung. Chest Surg Clin N Am 8:183, 1998.

Onda M, et al: ErbB-2 expression is correlated with poor prognosis for patients with osteosarcoma. Cancer 77:71, 1996.

P.1861

Pass HI, et al: Isolated lung perfusion with tumor necrosis factor for pulmonary metastases. Ann Thorac Surg 61:1609, 1996.

Pastorino U, et al: The contribution of salvage surgery to the management of childhood osteosarcoma. J Clin Oncol 9:1357, 1991.

Pastorino U, et al: Primary childhood osteosarcoma: the role of salvage surgery. Ann Oncol 3[Suppl 2]:S43, 1992.

Pastorino U, et al: Long-term results of lung metastasectomy: prognostic analyses based on 5206 cases. J Thorac Cardiovasc Surg 113:37, 1997.

Patanaphan V, Salazar OM, Risco R: Breast cancer: metastatic patterns and their prognosis. South Med J 81:1109, 1988.

Patel SR, et al: Results of two consecutive trials of dose-intensive chemotherapy with doxorubicin and ifosfamide in patients with sarcomas. Am J Clin Oncol 21:317, 1998.

Paulussen M, et al: Primary metastatic (stage IV) Ewing tumor: survival analysis of 171 patients from the EICESS studies. European Intergroup Cooperative Ewing Sarcoma Studies. Ann Oncol 9:275, 1998a.

Paulussen M, et al: Ewing's tumors with primary lung metastases: survival analysis of 114 (European Intergroup) Cooperative Ewing's Sarcoma Studies patients. J Clin Oncol 16:3044, 1998b.

Perilongo G, et al: Hepatoblastoma presenting with lung metastases: treatment results of the first cooperative, prospective study of the International Society of Paediatric Oncology on childhood liver tumors. Cancer 89:1845, 2000.

Pfannschmidt J, et al: Prognostic factors for survival after pulmonary resection of metastatic renal cell carcinoma. Ann Thorac Surg 74:1653, 2002.

Picci P, et al: Computed tomography of pulmonary metastases from osteosarcoma: the less poor technique. A study of 51 patients with histological correlation. Ann Oncol 12:1601, 2001.

Piltz S, et al: Long-term results after pulmonary resection of renal cell carcinoma metastases. Ann Thorac Surg 73:1082, 2002.

Pogrebniak HW, et al: Resection of pulmonary metastases from malignant melanoma: results of a 16-year experience. Ann Thorac Surg 46:20, 1988.

Pogrebniak HW, et al: Reoperative pulmonary resection in patients with metastatic soft tissue sarcoma. Ann Thorac Surg 52:197, 1991.

Pogrebniak HW, et al: Renal cell carcinoma: resection of solitary and multiple metastases. Ann Thorac Surg 54:33, 1992.

Pollock R, et al: Wild-type p53 and a p53 temperature-sensitive mutant suppress human soft tissue sarcoma by enhancing cell cycle control. Clin Cancer Res 4:1985, 1998.

Potter DA, et al: Patterns of recurrence in patients with high-grade soft-tissue sarcomas. J Clin Oncol 3:353, 1985.

Putnam JB Jr, Roth JA: Prognostic indicators in patients with pulmonary metastases. Semin Surg Oncol 6:291, 1990.

Putnam JB Jr, et al: Survival following aggressive resection of pulmonary metastases from osteogenic sarcoma: analysis of prognostic factors. Ann Thorac Surg 38:516, 1983.

Putnam JB Jr, et al: Analysis of prognostic factors in patients undergoing resection of pulmonary metastases from soft tissue sarcomas. J Thorac Cardiovasc Surg 87:260, 1984.

Putnam JB Jr, et al: Extended resection of pulmonary metastases: is the risk justified? Ann Thorac Surg 55:1440, 1993.

Putnam JB Jr, Thomsen SL, Siegenthaler MP: Therapeutic implications of heat-induced lung injury. SPIE Proceedings. Matching the Energy Source to the Clinical Need: A Critical Review 75:139, 2000a.

Putnam JB Jr, et al: Early results of isolated single lung perfusion for treatment of unresectable sarcomatous metastases. American Association for Thoracic Surgery 80th Annual Meeting, Toronto, Canada, April 30 May 3, 2000b.

Putnam JB Jr: New and evolving treatment methods for pulmonary metastases. Semin Thorac Cardiovasc Surg 14:49, 2002.

Regnard J F, et al: Surgical treatment of hepatic and pulmonary metastases from colorectal cancers. Ann Thorac Surg 66:214, 1998.

Ren H, et al: Computed tomography of inflation-fixed lungs: the beaded septum sign of pulmonary metastases. J Comput Assisted Tomogr 13: 411, 1989.

Rena O, et al: Pulmonary resection for metastases from colorectal cancer: factors influencing prognosis. Twenty-year experience. Eur J Cardiothorac Surg 21:906, 2002.

Rizzoni WE, et al: Resection of recurrent pulmonary metastases in patients with soft-tissue sarcomas. Arch Surg 121:1248, 1986.

Robert JH, et al: Factors influencing long-term survival after lung metastasectomy. Ann Thorac Surg 63:777, 1997.

Robinson BJ, et al: Is resection of pulmonary and hepatic metastases warranted in patients with colorectal cancer? J Thorac Cardiovasc Surg 117: 66, 1999.

Rolle A, et al: Lobe-sparing resection of multiple pulmonary metastases with a new 1318-nm Nd:YAG laser first 100 patients. Ann Thorac Surg 74:865, 2002.

Rosen G, et al: Thoracotomy in the management of metastatic soft-tissue sarcomas in adults. Chest Surg Clin N Am 4:67, 1994.

Roth JA, et al: Differing determinants of prognosis following resection of pulmonary metastases from osteogenic and soft tissue sarcoma patients. Cancer 55:1361, 1985.

Roth JA, et al: Comparison of median sternotomy and thoracotomy for resection of pulmonary metastases in patients with adult soft-tissue sarcomas. Ann Thorac Surg 42:134, 1986.

Saito Y, et al: Pulmonary metastasectomy for 165 patients with colorectal carcinoma: a prognostic assessment. J Thorac Cardiovasc Surg 124: 1007, 2002.

Sakamoto T, et al: Pulmonary resection for metastases from colorectal cancer. Chest 119:1069, 2001.

Salvati F, et al: DNA flow cytometric studies of 66 human lung tumors analyzed before treatment. Prognostic implications. Chest 96:1092, 1989.

Sarcoma Meta-analysis Collaboration: Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: meta-analysis of individual data. Lancet 350:1647, 1997.

Schneider P, et al: Chemoembolization of the lung improves tumor control in a rat model. Clin Cancer Res 8:2463, 2002.

Schnorrer M, et al: Management of germ cell testicular cancer with pulmonary metastases. Neoplasma 43:47, 1996.

Schott G, Weissmuller J, Vecera E: Methods and prognosis of the extirpation of pulmonary metastases following tumor nephrectomy. Urol Int 43:272, 1988.

Schroder C, et al: Technique and results of hyperthermic (41 C) isolated lung perfusion with high-doses of cisplatin for the treatment of surgically relapsing or unresectable lung sarcoma metastasis. Eur J Cardiothoracic Surg 22:41, 2002.

Schrump DS, et al: Pharmacokinetics of paclitaxel administered by hyperthermic retrograde isolated lung perfusion techniques. J Thorac Cardiovasc Surg 123:686, 2002.

Scotlandi K, et al: Multidrug resistance and malignancy in human osteosarcoma. Cancer Res 56:2434, 1996.

Seki M, et al: Surgical treatment of pulmonary metastases from uterine cervical cancer. Operation method by lung tumor size. J Thorac Cardiovasc Surg 104:876, 1992.

Seto M, et al: Suppression of tumor growth and pulmonary metastasis in murine osteosarcoma using gene therapy. Oncol Rep 9:337, 2002.

Shingu K, et al: Intratracheal macrophage-activating lipopeptide-2 reduces metastasis in the rat lung. Am J Respir Cell Mol Biol 28:316, 2003.

Shiromizu K, et al: A clinicopathological study of postoperative pulmonary metastasis of uterine cervical carcinomas. J Obstet Gynaecol Res 25:245, 1999.

Shuman RL: Primary pulmonary sarcoma with left atrial extension via left superior pulmonary vein. En bloc resection and radical pneumonectomy on cardiopulmonary bypass. J Thorac Cardiovasc Surg 88:189, 1984.

Siggelkow W, et al: The value of positron emission tomography in the follow-up for breast cancer. Anticancer Res 23:1859, 2003.

Simpson R, et al: Pulmonary resection for metastatic breast cancer. Aust N Z J Surg 67:717, 1997.

Skinner KA, et al: Surgical treatment and chemotherapy for pulmonary metastases from osteosarcoma. Arch Surg 127:1065, 1992.

Skubitz KM, Anderson PM: Inhalational interleukin-2 liposomes for pulmonary metastases: a phase I clinical trial. Anti-Cancer Drugs 11:555, 2000.

Slebos RJ, et al: Allele-specific detection of K-ras oncogene expression in human non-small-cell lung carcinomas. Int J Cancer 48:51, 1991.

Snyder BJ, Pugatch RD: Imaging characteristics of metastatic disease to the chest. Chest Surg Clin N Am 8:29, 1998.

Snyder CL, et al: A new approach to the resection of pulmonary osteosarcoma metastases. Results of aggressive metastasectomy. Clin Orthop 270:247, 1991.

Spaggiari L, et al: Pneumonectomy for lung metastases: indications, risks, and outcome. Ann Thorac Surg 66:1930, 1998.

Spunt SL, et al: Selective use of whole-lung irradiation for patients with Ewing sarcoma family tumors and pulmonary metastases at the time of diagnosis. J Pediatr Hematol Oncol 23:93, 2001.

P.1862

Srinivas S, Varadhachary G: Spontaneous pneumothorax in malignancy: a case report and review of the literature. Ann Oncol 11:887, 2000.

Staren ED, et al: Pulmonary resection for metastatic breast cancer. Arch Surg 127:1282, 1992.

Steinke K, et al: Percutaneous pulmonary radiofrequency ablation: difficulty achieving complete ablations in big lung lesions. Br J Radiol 76: 742, 2003.

Steyerberg EW, et al: Residual pulmonary masses after chemotherapy for metastatic nonseminomatous germ cell tumor. Prediction of histology. ReHiT Study Group. Cancer 79:345, 1997.

Tan L, et al: Role of screening chest computed tomography in patients with advanced head and neck cancer. Otolaryngol Head Neck Surg 120:689, 1999.

Temeck BK, et al: Reoperative pulmonary metastasectomy for sarcomatous pediatric histologies. Ann Thorac Surg 66:908, 1998.

Temple LK, Brennan MF: The role of pulmonary metastasectomy in soft tissue sarcoma. Semin Thorac Cardiovasc Surg 14:35, 2002.

Tesoro-Tess JD, et al: Reliability of diagnostic imaging after orchiectomy alone in follow-up of clinical stage I testicular carcinoma: excessive cost with potential risk. Lymphology 20:161, 1987.

Thomas DG: Exact and asymptotic methods for the combination of 2 2 tables. Comput Biomed Res 8:423, 1975.

Toledano AY, Olak J: Statistical techniques and analysis in thoracic surgery. In: Franco KL, Putnam JB Jr, eds. Advanced Therapy in Thoracic Surgery. 1st Ed. Hamilton, Ontario, Canada: B.C. Decker, 1998, p. 146.

Torek F: Removal of metastatic carcinoma of the lung and mediastinum: suggestions as to technic. Arch Surg 21:1416, 1930.

Troell RJ, Terris DJ: Detection of metastases from head and neck cancers. Laryngoscope 105:247, 1995.

Tsuchiya H, et al: Effect of timing of pulmonary metastases identification on prognosis of patients with osteosarcoma: the Japanese Musculoskeletal Oncology Group study. J Clin Oncol 20:3470, 2002.

Uchiyama M, et al: A study of therapy for pediatric hepatoblastoma: prevention and treatment of pulmonary metastasis. Eur J Pediatr Surg 9: 142, 1999.

Udelsman R, et al: Endobronchial metastases from soft tissue sarcoma. J Surg Oncol 32:145, 1986.

van der Poel HG, et al: Metastasectomy in renal cell carcinoma: a multicenter retrospective analysis. Eur Urol 35:197, 1999.

van Geel AN, et al: Surgical treatment of lung metastases: the European Organization for Research and Treatment of Cancer-Soft Tissue and Bone Sarcoma Group study of 255 patients. Cancer 77:675, 1996.

Van Putte BP, et al: Single-pass isolated lung perfusion versus recirculating isolated lung perfusion with melphalan in a rat model. Ann Thorac Surg 74:893, 2002.

Van Putte BP, et al: Isolated lung perfusion with gemcitabine in a rat: pharmacokinetics and survival. J Surg Res 109:118, 2003.

Van Schil PE: Surgical treatment for pulmonary metastases. Acta Clin Belg 57:333, 2002.

Vaporciyan AA, et al: Resection of advanced thoracic malignancies requiring cardiopulmonary bypass. Eur J Cardiothoracic Surg 22:47, 2002.

Vaughn C, Mychaskiw G 2nd, Sewell P: Massive hemorrhage during radiofrequency ablation of a pulmonary neoplasm. Anesth Analg 94:1149, 2002.

Veronesi G, et al: Fluoro-deoxi-glucose uptake and angiogenesis are independent biological features in lung metastases. Br J Cancer 86:1391, 2002.

Voute PA, et al: A phase II study of cisplatin, ifosfamide and doxorubicin in operable primary, axial skeletal and metastatic osteosarcoma. European Osteosarcoma Intergroup (EOI). Ann Oncol 10:1211, 1999.

Walker R, et al: STIR magnetic resonance imaging as a whole-body screening tool for metastases in patients with breast carcinoma: preliminary clinical experience. J Magn Reson Imaging 11:343, 2000.

Walsh GL, Nesbitt JC: Tumor implants after thoracoscopic resection of a metastatic sarcoma. Ann Thorac Surg 59:215, 1995.

Weh HJ, et al: Chemotherapy of metastatic soft tissue sarcoma with a combination of adriamycin and DTIC or adriamycin and ifosfamide. Onkologie 13:448, 1990.

Weinlechner JW: Zur Kasuistik der Tumoren an der Brustwand and deren Behandlung (Resektion der rippen, Eroffnung der Brusthohle, partielle Entfernung der Lungen). Wien Med Wochenschr 32:589, 1882.

Weiser MR, et al: Repeat resection of pulmonary metastases in patients with soft-tissue sarcoma. J Am Coll Surg 191:184, 2000.

Weksler B, et al: Isolated single lung perfusion in the rat: an experimental model. J Appl Physiol 74:2736, 1993.

Weksler B, et al: Isolated single lung perfusion with doxorubicin is effective in eradicating soft tissue sarcoma lung metastases in a rat model. J Thorac Cardiovasc Surg 107:50, 1994.

Weksler B, Burt M: Isolated lung perfusion with antineoplastic agents for pulmonary metastases. Chest Surg Clin N Am 8:157, 1998.

Whelan JS, et al: A systematic review of the role of pulmonary irradiation in the management of primary bone tumours. Ann Oncol 13:23, 2002.

Whitmore M, Li S, Huang L: LPD lipopolyplex initiates a potent cytokine response and inhibits tumor growth. Gene Ther 6:1867, 1999.

Wilimas JA, et al: Significance of pulmonary computed tomography at diagnosis in Wilms' tumor. J Clin Oncol 6:1144, 1988.

Williams MP, Husband JE, Heron CW: Intrathoracic manifestations of metastatic testicular seminoma: a comparison of chest radiographic and CT findings. AJR Am J Roentgenol 149:473, 1987.

Woodard PK, Dehdashti F, Putman CE: Radiologic diagnosis of extrathoracic metastases to the lung. Oncology (Huntingt) 12:431, 1998.

Wootton-Gorges SL, et al: Chest radiography versus chest CT in the evaluation for pulmonary metastases in patients with Wilms' tumor: a retrospective review. Pediatr Radiol 30:533, 2000.

Wyttenbach R, Vock P, Tschappeler H: Cross-sectional imaging with CT and/or MRI of pediatric chest tumors. Eur Radiol 8:1040, 1998.

Yang SC, et al: Radiofrequency ablation of primary and metastatic lung tumors: analysis of an ablate and resection study. American Association for Thoracic Surgery 82nd Annual Meeting 210, 2002.

Yonemoto T, et al: Prognosis of osteosarcoma with pulmonary metastases at initial presentation is not dismal. Clin Orthop 349:194, 1998.

Younes RN, Gross JL, Deheinzelin D: Surgical resection of unilateral lung metastases: is bilateral thoracotomy necessary? World J Surg 26:1112, 2002.