SEARCH

SEARCH BY CITATION

Keywords:

  • pulmonary;
  • metastasectomy;
  • lung

INTRODUCTION

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES

The first pulmonary metastasectomy was reported in 1882 [1]. For a period of time, resection of metastatic disease to the lung was performed only for disease discovered incidentally, usually at the time of resection of chest wall neoplasms. Eventually, as chest imaging became available, planned resection of lung metastases began to be performed. In the first series of 12 patients [2], the criteria for pulmonary metastasectomy were defined and have since been updated [3]. The primary site of disease has to be either controlled or appear controllable, complete resection of lung metastatic disease has to be feasible, the patient has to be able to tolerate planned procedures, and no better alternative treatments can be available are the criteria for resection.

Several prognostic factors have been suggested from reviews of retrospective patient series. The most important analysis of these was performed by The International Registry of Lung Metastasis, which reported outcomes and associated prognostic factors for 5,206 patients of all histologies from a database which combined retrospectively and prospectively gathered data [4]. The 5-year overall survival for all histologies was 36%. Extended disease-free interval (DFI) (i.e., more than 3 years), a limited number of lung metastases (i.e., one nodule), and completeness of resection were found to be predictors of favorable outcomes.

Most recently, The European Society of Thoracic Surgeons (ESTS) formed the Lung Metastasectomy Working Group to produce guidelines based on all available evidence [5]. They concluded the level of evidence is insufficient for guidelines and noted that randomized controlled trials have not been completed. At present, pulmonary metastasectomy is offered to patients based on the observation that long-term survival can be seen after resection, while long-term survival with systemic therapy alone as treatment for patients with pulmonary metastases appears extremely rare.

DIAGNOSIS AND DETECTION

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES

Pulmonary metastases are usually asymptomatic and only 15–20% of patients present with cough, chest pain, hemoptysis, or spontaneous pneumothorax. Computed tomographic (CT) scans are the imaging modality of choice. A pulmonary metastasis is typically a peripheral, well-circumscribed nodule [6, 7]. Two-thirds of metastases are found in the peripheral third of the lung.

The accuracy of CT in the definition of lesions as metastases depends on the lesion size. While CT scans detect a greater number of pulmonary nodules than chest radiographs (CXR), it is with a decrease in specificity since small benign nodules are common [8, 9]. McCormack et al. [10] reviewed 144 patients with metastatic lesions. CT scan underestimated the number of malignant nodules in 25% of patients and differed from pathology in 42% of cases. CT scans do not reliably predict the number of lesions that will be discovered at thoracotomy with palpation of the lung. Pastorino [11] reported that 25% of 2,988 patients had more metastatic lesions detected at surgery than estimated by pre-operative imaging. For the patients (1,134) who underwent bilateral exploration, the imaging underestimated the number of lesions in 39% and overestimated the number in 25%. Cerfolio et al. [12] prospectively evaluated non-imaged malignant nodules at thoracotomy for patients who had 64-slice helical CT scans. Thirty-four percent of patients had 57 pulmonary nodules that were not imaged preoperatively. Fifty-six percent of these nodules were malignant. Ellis et al. [13] reported that a significant number of nodules were manually detected at thoracotomy that were not noted on pre-operative multidetector helical CT scans (mean 3.24 vs. 2.12).

Neither MRI nor 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) have proven helpful in the characterization of lung lesions in patients at risk for pulmonary metastases [14, 15]. In the prospective evaluation by Cerfolio et al. previously mentioned, 87% patients (132/152) had integrated PET-CT. Even with this high percentage of PET-CT, 34% of patients had lesions only detected at thoracotomy [12]. The limitations for detecting subcentimeter disease apply to PET-CT as well as CT alone. Reinhardt et al. [16] reported a 41% and 100% false negative rate for lesions 6–10 mm and less than 5 mm, respectively.

PET is being used to search for extrathoracic metastatic disease based on the glycolytic properties of each histology. Pastorino et al. [17] reported that 21% of patients (19/86) did not undergo pulmonary metastasectomy secondary to unexpected findings of FDG-PET. Dalrymple-Hay et al. [14] reported that 9% of patients with metastatic melanoma were excluded based on FDG-PET results [18-20].

SURGICAL THERAPY FOR LUNG METASTASIS

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES

The complete resection of all disease is the goal of pulmonary metastasectomy. The volume of disease, the location of the lesions, and the performance status of the patient guide the surgical approach. A posterolateral thoracotomy has been the standard approach, but muscle-sparing thoracotomy, anterior thoracotomy, median sternotomy, “clamshell” thoracotomy (i.e., bilateral anterior thoracotomies with a transverse sternotomy), and “hemiclamshell” thoracotomy (i.e., unilateral anterior thoracotomy with partial sternotomy) have all been utilized. Each of these approaches allows palpation of all of the parenchyma in one or both hemithoraces.

More controversial because of an inability to thoroughly palpate the lung parenchyma is video-assisted thoracic surgery (VATS) approach for pulmonary metastasectomy [21-23]. McCormack et al. [22] prospectively evaluated the role of VATS for pulmonary metastasis resection. In 18 patients, all CT detected lesions were resected thoracoscopically followed by a standard thoracotomy for palpation and resection of additional lesions. Fifty-six percent of patients had additional malignant lesions after VATS resection. This study was reported in 1996 and significant improvements in CT scanning and development of integrated PET-CT have occurred. However, despite improvements in imaging, missed lesions remain a concern. Cerfolio et al. [12], Ellis et al. [13], and Kayton et al. [24] report the inability to detect all lesions with up-to-date CT scanners. The ESTS practice patterns were investigated to determine the approach for pulmonary metastasectomy [25]. Sixty-five percent thought palpation was necessary for adequate metastasectomy.

Even though a large number of metastases appear likely to be missed if thoracoscopic techniques are employed, studies that reported long-term outcomes do not report a difference in survival associated with this approach. Mutsaerts et al. [21] reported that of 20 patients with a single pulmonary metastasis who underwent either VATS or thoracotomy, the 5-year survivals were similar. Gossot et al. [26] concluded that VATS yielded a similar survival for 31 patients treated by VATS to 29 patients treated by thoracotomy for one or two sarcoma metastases. Nakas et al. [27] report no survival advantage among 27 patients undergoing VATS versus open resection for pulmonary metastases. These studies can be criticized for including only limited numbers of patients with very limited number of lesions.

The approach to bilateral lesions include median sternotomy, clamshell thoracotomy, sequential or simultaneous bilateral thoracotomies, or VATS procedures [28-30].

Resection of pulmonary metastases most often requires wedge resection of the lung parenchyma. Anatomic resections are infrequently required, but if a lesion only can be performed by a segmentectomy, lobectomy, or even pneumonectomy, performing such procedures is often appropriate.

PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES

Colorectal cancer is the disease for which pulmonary metastasectomy has been most commonly reported. Pulmonary metastases without other systemic metastases occur in 2–7% of patients with colon cancer and in 4–7% of patients with rectal cancer [31]. After resection, the 5-year survivals are 30–68%, and the 10-year survivals are 20–37% [32-35]. Melloni et al. [36] reported complete versus incomplete resection are associated with a 5-year survivals of 44% versus 0%, respectively. Onaitis et al. [33] reported >3 metastases, DFI < 1 year, age > 65, and male sex were adverse prognostic factors [37]. The survival of patients with bilateral lesions is not different than patients with ipsilateral lesions after adjusting for number of lesions [35, 38]. A solitary metastasis is associated with a better prognosis than multiple metastases (either unilateral or bilateral), but the complete removal of all disease is consistently most important prognostic factor. In addition, the experience of combining pulmonary metastasectomy with resection of extra-pulmonary sites of metastatic disease has been reported. Several reports suggested patients undergoing combined pulmonary and hepatic metastasectomy can experience prolonged survival [37, 39-51]. Headrick et al. [42] reported 5- and 10-year survivals of 30% and 16%, respectively. Zabaleta et al. report 3- and 5-year survival rates of 50% and 39% for 17 patients who underwent pulmonary metastasectomy after prior hepatic metastasectomy [42].

Bone and soft tissue sarcoma pulmonary metastasis occur in 25–70% of patients after treatment for localized disease and in 10% who are initially diagnosed with metastatic disease [43]. The lung is the site of failure in up to 90% of cases and was also the most common cause of death [44, 45]. The reported 5-year survival for patients undergoing pulmonary metastasectomy for sarcomas ranges from 23% to 50% [11, 46-49]. Completeness of resection of pulmonary disease is associated with long-term survival [11, 44, 48-50]. Billingsley et al. showed a 3-year actuarial survival of 46% versus 17% for patients who had complete versus incomplete resection, respectively. Kager et al. [46] also reported the importance of complete resection with a 5-year actuarial survival of 44% versus 0% in patients with complete versus incomplete resection, respectively. Prognostic factors have been reported in several studies. The histologies of liposarcoma, malignant fibrous histiocytoma, and malignant peripheral nerve tumor are associated with worse outcomes [44, 49]. Long DFI and minimal number of lesions are reported to be favorable prognostic factor [46].

Stephens et al. [51] reported that patients with sarcoma who do not demonstrate a response to induction chemotherapy have a worse prognosis. The 5-year actuarial survival was 32% and 0% for patient without or with progressive disease, respectively. These results suggest that as multimodality therapy evolves, the treatment response to neoadjuvant therapy may serve to select patient for surgical resection of metastatic disease.

Rates of recurrences of sarcoma in the lungs after metastasectomy range from 45% to 83%. Weiser et al. [52] reported a series of 86 patients from Memorial Sloan Kettering after re-resection of pulmonary metastases of sarcoma. The 5-year survival after two or more pulmonary resections was 36%. The median survival was 51 months after complete resection versus 6 months after incomplete resection. Three nodules, lesions > 2 cm, and high-grade primaries were adverse factors. Resection of sarcomatous pulmonary metastases in addition to synchronous and metachronous extrathoracic disease has been reported [44, 53, 54]. Blackmon et al. [54] reported a series at MD Anderson Cancer Center of pulmonary metastasectomy with non-pulmonary metastases. A progressive reduction in survival was noted from patients with resected pulmonary metastases, to patients with resected pulmonary and extrapulmonary metastases, to non-resected pulmonary and extrapulmonary lesions. These studies show that complete resection of recurrent and extrathoracic disease is associated with long-term survival in some patients.

Although colorectal and sarcoma pulmonary metastasectomies comprise 50% of resections performed, resections for melanoma, renal cell carcinoma, head and neck tumors, and germ cell tumors have multiple reports outlining feasibility, outcomes, and prognostic factors.

Patients with metastatic melanoma have isolated lung metastasis in 1.9–11% of patients [55]. Petersen et al. [56] reported 1,720 patients with pulmonary metastasis from melanoma. Prognostic factors included nodular histology, number of lesions, extrathoracic disease, and pulmonary metastasectomy. For the 318 patients treated by resection, a survival advantage was observed, but the DFI and presences of extrathoracic metastases had a statistical interaction. The reported 5-year survivals after removal of pulmonary melanoma ranged from 4.5% to 38% [14, 57, 58]. Survival of patients undergoing metastasectomy is poor regardless of resection, but recent advances in systemic therapy may alter the current treatment paradigm. BRAF inhibitors, anti-CTLA-4 antibody (ipilimumab), and immune-based therapies may work as single agents, but also may be incorporated into multimodality therapy. Metastasectomy in combination with treatment with these novel compounds remains largely unexplored.

About 30% of patients with renal cell carcinoma will initially be diagnosed with stage IV disease and another 30–50% will develop metastases [59]. The 5-year survival is about 3% if metastatic disease is not resected [60]. The 5-year survival ranges from 36% to 53% after resection of lung only disease [60-63]. Complete resection, number of metastases, and lymph node involvement are prognostic factors [60, 62, 63]. Alt et al. [64] reported a cancer-specific 5-year survival of 73.6% and 19% for complete and incomplete pulmonary metastasectomy, respectively, in a report of 224 patients with isolated pulmonary metastasis.

Similar to melanoma, renal cell carcinoma has been historically resistant to chemotherapy, but newer agents have shown improved efficacy. The monoclonal antibody, bevacizumab, small molecule inhibitors, sorafenib and sunitinib, and mTOR inhibitors, temsirolimus and everolimus, are rapidly changing the treatment of renal cell carcinoma, but the role of these drugs with metastasectomy remains unknown. Regardless, these therapies with metastasectomy may rapidly redefine the treatment paradigm of patients with metastatic renal cell carcinoma.

Head and neck cancers are unusual in that determining whether a lesion is represents pulmonary metastatic disease or a new primary lung cancer can be difficult. About 10–40% of malignant lung nodules in patients with prior head and neck cancers will be found to be primary lung tumors [65]. Lung metastases occur in about 4–25% of patients with head and neck cancers [66, 67]. The 5-year survival in these patients is about 21–59% [66, 68, 69].

Germ cell tumors are unique in that post-treatment histology is critical for prognostic information. The survival has increased significantly with cisplatin-based chemotherapy, but residual pulmonary nodules are common [70, 71]. Surgical resection is diagnostic for viable malignancy, mature teratoma, and fibrosis. Residual viable tumor is the most important post-treatment prognostic factor. Liu et al. [70] report that the 10-year outcomes of patients with viable tumor cells was 43%, fibrosis was 86%, and mature teratoma was 84%. The 5-year survival ranges from 59% to 94% [72-74].

A few reports detail patient outcomes after resection of histologies with very limited experience which include esophageal, gastric, hepatocellular, urothelial transitional cell, cervical, and adrenocortical carcinoma [75-79]. These reports represent a highly selected group of patients, but what is clear is that the DFI and number of lesions are fairly consistent prognostic factors and that long-term survival is obtainable in some patients.

ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES

Two treatments are alternatives to surgical resection of pulmonary metastases: stereotactic body radiation therapy (SBRT) and radiofrequency ablation (RFA). SBRT delivers radiation to a tumor with minimal damage to surrounding normal tissue. Several studies report high local control with low toxicity. Peripheral tumors that are small and solitary are the best targets for SBRT [80, 81]. Rusthoven et al. [80] reported a phase I–II trial with dose escalation, one to three lesions, and no metastasis greater than 7 cm. Median survival was 19 months and 2-year survival was 39%. RFA delivers thermal energy to induce coagulation necrosis. The experience with RFA is quite limited, but local recurrences have been reported to approach 50%. Pennathur et al. [82] reported patients undergoing RFA with mixed histologies. Six of these patients underwent RFA as an adjunct to open thoracotomy based on the locations of lesions. The 2-year survival was 68% for patients who underwent RFA during thoracotomy.

Experimental approaches include isolated lung perfusion (ILP) and transpulmonary chemoembolization [83, 84]. ILP has been developed based on experience with limb perfusion to treat malignant melanoma or sarcoma. Johnston et al. [85] reported this procedure in 1983. ILP requires cannulation of the pulmonary artery and veins to deliver chemotherapy directly to the lung. Clinical trials have revealed significant toxicity in phase I reports [84]. Transpulmonary chemoembolization delivers chemotherapy to the lung by percutaneous catheterization of pulmonary arteries. Vogl et al. [86] reported 106 lung metastases treated with this method.

FUTURE DIRECTIONS

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES

Despite the many hundreds of reports on addressing the feasibility and efficacy of resection of metastatic disease to the lung, there is essentially no prospective data, and therefore, benefit cannot be assumed. Prior attempts at trials evaluating efficacy of pulmonary metastasectomy have failed because of unwillingness on the part of either surgeons or patients to randomize patients to a non-surgical arm. In an novel trial design, Treasure et al. [87] is currently attempting to clarify the question as to whether metastasectomy has benefit in the Pulmonary Metastasectomy for Colorectal Cancer (PulMiCC) trial. This trial opened for enrollment in March of 2010 and is attempting to enroll patients with colorectal metastases who are in the “zone of uncertainty” regarding appropriateness of therapeutics. Patients are seen as falling into one of three groups. The first group would be most likely to benefit from surgery and would include such patients as those with a single lung lesion many years after resection of the primary site. At the other extreme are patients who are very unlikely to benefit from surgery. These patients have many lesions with a short DFI and would be treated with systemic therapy alone. These two groups would be enrolled but not randomized and treated at the physician's discretion. The third group is those who the treating physicians have equipoise about whether surgery or medical therapies are most appropriate. These patients have a limited number of lesions with an intermediate DFI time (e.g., one or two lesions at 2–4 years after resection). These patients are being enrolled and randomized to surgery versus medical therapies. Hopefully, this trial will help clarify which patients, if any, with pulmonary metastases from a colon primary benefit from resection.

The majority of advancements in metastatic disease have been the development of newer systemic agents with responses in diseases that have traditionally been resistant to chemotherapy. While these agents have created excitement in both patients and the oncologic community, most of these agents are tested with progression-free survival end-points. Unfortunately, most patients treated with these new agents do experience progressive disease even if median survival has been increased. The role of these agents in conjunction with metastasectomy remains largely unexplored and represents an area in which well-designed clinic trails of multimodality therapy could answer questions about the optimal management of patients with pulmonary metastases.

REFERENCES

  1. Top of page
  2. INTRODUCTION
  3. DIAGNOSIS AND DETECTION
  4. SURGICAL THERAPY FOR LUNG METASTASIS
  5. PULMONARY METASTASECTOMY FOR SPECIFIC HISTOLOGIES
  6. ALTERNATIVE LOCAL TREATMENTS FOR PULMONARY METASTASIS
  7. FUTURE DIRECTIONS
  8. REFERENCES