Extent of surgical resection

The goal of pulmonary metastasectomy is the complete resection of the metastases while preserving as much pulmonary parenchyma as possible. Most of the pulmonary metastatic lesions are generally located at the periphery of the lung and easily accessible to wedge resection. In contrast to primary lung cancer that generally requires anatomic resection, wedge resection is as effective as anatomic resection to manage metastasis to the lung parenchyma^[10]. Pulmonary metastases are easily managed by wedge resections using a standard stapler. This approach has the advantage of achieving clear surgical margins and efficient aerostasis. Other techniques have been developed as an alternative to staplers to allow maximal sparing of lung parenchyma, especially when lesions are multiple^[24] (Table 17.4). Cautery resection (precision resection), initially described by Perelman, may be proposed for deeply located or peripheral lesions in patients who are not candidates for anatomic resection^[25]. The technique consists of nodule resection by cautery and individual ligation of small vessels and bronchi. Another approach uses lasers. This is usually performed by thoracotomy and allows a superficial limited excision without deformation of the pulmonary parenchyma. It can also allow complete resection for lesions located close to major bronchi and/or vessels^[21]. The 1,318-nm Nd:YAG laser is the only laser device capable of dissecting lung parenchyma due to its high absorption of water^[26,27]. The disadvantages of laser systems are the financial investment and staff education. Moreover, this expensive technique has not proven a clear benefit in comparison to cautery resection. The Ligasure Vessel Sealing System, which is an electrothermal bipolar tissue sealing system, and the Ultracision scalpel using ultrasonic waves promoting coagulation have been proposed recently for wedge resection, but data are lacking in thoracic surgery on lung parenchyma with these techniques^[28,29].

For large or central lesions, segmentectomy, lobectomy, or occasionally pneumonectomy may be required. Within the International Registry of Lung Metastases, pulmonary metastasectomy was achieved with wedge resection in 67%, segmentectomy in 9%, lobectomy in 21%, and pneumonectomy in 3%^[9]. Some authors consider pneumonectomy to be a relative contraindication to pulmonary metastasectomy that should be reserved for solitary centrally located tumors with a long tumor-free-interval without previous pulmonary resection. Resection extended to other structures (chest wall, great vessels) may be justified if complete resection can be achieved^[30,31].

Surgical approach

Surgical approach is determined by the number of metastases, their location, the functional reserves of the patient, and whether or not the disease is limited to one hemithorax. The choice of surgical approach should permit complete resection while saving as much pulmonary parenchyma as possible^[32]. There are different approaches, each with their advantages and disadvantages: thoracotomy (anterior or posterolateral), median sternotomy, bilateral transsternal thoracotomy (Clamshell incision), and thoracoscopy. The optimal approach still remains controversial (Table 17.5)^[10,32,33].

In a recent survey of the European Society of Thoracic Surgeons (ESTS), the approaches and preferences were recorded and included anterolateral thoracotomy (36.3%), video-assisted thoracoscopic surgery (28.8%), posterior muscle-sparing thoracotomy (26%), posterolateral thoracotomy (22.6%), axillary thoracotomy (17.2%), sternotomy (1.4%), bilateral staged thoracotomy (66.2%), single-stage sternotomy (26.9%), single-stage bilateral sequential thoracotomy (19.3%), bilateral staged versus single-stage VATS (12.4 vs 7.6%), and single-stage clamshell incision (7.6%)^[34].

The most commonly used approach is the lateral or posterolateral thoracotomy via the fourth or fifth intercostal space, which allows optimal exposure of the lung with the opportunity to perform bimanual palpation of the entire lung to find additional nodules not noticeable on the pre-operative CT scan. Postoperative pain and the inability to reach the contralateral lung are major disadvantages. Systematic one-stage bilateral thoracotomies provide no survival advantage when unilateral disease is evident on radiologic findings^[35]. Most centers propose staged posterolateral thoracotomies for patients with bilateral disease. Median sternotomy offers the advantage to reach both lungs, allowing bilateral resection^[36]. Postoperative pain is less important than a bilateral thoracotomy. However, access to the lower lobes, lymph node dissection, and anatomic resections are more difficult than with a classical thoracotomy. Clamshell incision (bilateral anterior thoracotomy with transverse sternotomy) allows excellent visualization and palpation of the lung but involves a sacrifice of two mammary arteries and important postoperative pain and is nowadays considered of historical interest^[10].

Video-assisted thoracoscopic surgery has gained progressive interest for pulmonary metastasectomy^[37,38]. The potential benefits of the VATS approach are numerous: less pain, faster recovery, smaller incisions, decreased length of hospital stay, good visualization of the pleural space, less adhesions in the event of reintervention, and better compliance for adjuvant treatment^[39]. However, these advantages do not allow bimanual palpation of the entire lung to detect nodules that are not visible on preoperative radiologic exams. Some studies have demonstrated that an open approach with manual palpation allows identification of additional nodules in 16 to 46% of patients in comparison with the VATS approach^[19–21,40]. These series were conducted with old-generation CT scans with thick slices of more than 5 mm. Even if there are no prospective, randomized studies comparing the thoracoscopic approach to open bimanual thoracotomy, numerous series have demonstrated comparable survival rates between the two approaches for patients with less than three lesions^[41–45]. These results appear to be partly explained by the fact that there is no evidence that the absence of immediate detection of nodules of <5 mm influences tumor disease^[46]. Microscopic metastases, undetectable by VATS, can often be resected later if they become detectable on a follow-up CT with no deletary effect on patient survival. In fact, as long as patients can regularly be screened by repeat high-resolution CT scans, new nodules can be detected and additional resections performed (more easily than following a previous thoracotomy). Moreover, the resection of recurrent metastases does not interfere with overall survival in case of metachronous metastases. Many curable patients benefit from repeated resection, and repeated thoracoscopies are better tolerated than repeated thoracotomies. Finally, VATS resection is less traumatic and significantly decreases the immune response with a possible effect on the progression of the disease. Several adjunctive procedures have been suggested to help in the localization of nodules during VATS approach. Needle localization, methylene blue injection, and sonographic evaluation have been used to identify nodules not easily palpable on the visceral pleural surface^[47–49]. However, these maneuvers may help for the resection of solitary radiologically detectable lesions and will not allow for detection of lesions less than 5 mm.

Even without any randomized, controlled trial or meta-analysis available, an ESTS workgroup proposed some recommendations on an optimal surgical approach: for bilateral lesions, staged thoracotomy with an interval of 3 to 6 weeks and an interval CT are recommended. A VATS approach seems appropriate for diagnostic procedures and is not still accepted as a therapeutic modality. At this time, there is no alternative to bimanual palpation.

In conclusion, the optimal approach for pulmonary metastasectomy continues to evolve, and the patient must be informed of the different surgical and interventional options with their advantages and limitations. Given continued advancement in both imaging and operative technology, the role of VATS is expected to grow.

Lymph node involvement

Although mediastinal lymph node dissection or sampling is currently performed during pulmonary resection for primary lung cancer, many surgeons do not routinely perform this procedure during lung metastasectomy. In a recent ESTS survey, 56% of surgeons performed mediastinal sampling, 13% completed dissection, and 32% did no lymph node biopsy^[34]. There are two distinct clinical scenarios: (1) lymph node involvement discovered on preoperative imaging or invasive mediastinal staging and (2) incidental lymph node involvement discovered during surgery after mediastinal sampling or complete dissection. When systematic or sampling lymph node dissection is performed, lymph node metastases are identified in 12 to 32%^[50–52]. Systematic mediastinal lymph node dissection has been reported to have a significant difference in survival between patients with positive lymph nodes versus negative lymph nodes. Pfannschmidt reported 245 patients who underwent systematic lymph node dissection during pulmonary metastasectomy and found hilar or mediastinal lymph node involvement in 32%^[52]. Median survival was 64 months for patients without nodal metastases and 33 and 21 months for patients with hilar and mediastinal metastases, respectively. Other series have reported that hilar or mediastinal lymph node involvement has worse survival in various primary tumors, including colorectal, renal or head and neck cancers^[53–56]. The morbidity after complete mediastinal lymphadenectomy has been reported to be low and does not require much extra operative time^[57]. Even though survival benefit after lymph node dissection is unclear, most authors consider that complete mediastinal lymphadenectomy or sampling should be recommended during pulmonary metastasectomy to achieve accurate staging and guide additional chemotherapy^[16].

When mediastinal lymph node involvement is suspected in preoperative findings, more formal mediastinal evaluation should be performed, including mediastinoscopy, EBUS, or EUS. Pulmonary metastasectomy is questionable for preoperative histologically proven mediastinal involvement, and some authors propose to exclude patients with positive mediastinal lymph nodes from pulmonary metastasectomy^[16].

Resection of recurrent metastases

Recurrence of pulmonary metastasis is a common situation occurring in approximately 50% of cases. Within the International Registry of Lung Metastases, recurrent disease was found in 53%, and patients who underwent a second metastasectomy had a survival at 5 years of 44%^[9]. A longer time interval between the first metastasectomy and recurrent metastases appears to have a more favorable prognosis. Surgical resection of recurrent pulmonary metastases has been studied particularly in patients with colorectal cancer, with 5-year survivals ranging from 29 to 85%^[58–60]. These results suggest that the biology of the tumor is more important than the strategy of resection, biology allowing some lesions that are not detected by CT scans to be resected later without significant impact on overall survival. Conversely, in the case of biologically aggressive tumors, complete resection of all pulmonary lesions may not improve prognosis. No controlled studies are available to define the optimal imaging protocol after metastasectomy. Even if palpation at the time of metastasectomy was performed, regular follow-up is recommended. The interval for follow-up is not defined, but an ESTS workgroup’s recent recommendation was chest CT in the postoperative period, which will be repeated every 6 months for the first 2 years, then yearly^[15]. If the lung has not been palpated or the tumor doubling time is short, a more frequent radiologic monitoring should be discussed.

Other ablative techniques

Alternative, less aggressive therapies are currently under investigation. Radiologic ablation under guidance has been proposed using three different techniques: radiofrequency^[61], microwave^[62], and cryoablation^[63]. These CT-guided techniques have some potential limitations. It is impossible to ensure complete destruction of the lesion and resection margins cannot be obtained with histologic evidence. These techniques are still under investigation and may present serious complications (pneumothorax, hemoptysis, hemorrhage, bronchopleural fistula) and are usually reserved for nonoperable patients or patients who choose them.

Colorectal cancer

Lung metastases develop in 5 to 15% of colorectal cancer patients, and pulmonary metastasectomy is a treatment option in 1 to 2% of all patients with colorectal carcinoma^[64]. Colorectal cancer is currently the most common primary tumor for patients with potentially resectable pulmonary metastases. Pulmonary metastases may develop along four different types of scenarios: (1) synchronous with the primary colorectal cancer; (2) development of lung metastases in a patient who had previously developed liver metastases, (3) development of isolated lung metastases; and (4) recurrent pulmonary metastases after previous lung metastasectomy. Multiple studies have investigated the outcome of colorectal cancer patients who underwent resection of lung metastasis, with 5-year survival varying from 24 to 68% and with median survival of 18 to 67 months^[64–67]. Several survival prognostic factors have been identified: (1) disease-free interval between colonic tumor and lung metastasis identification of >12 months, (2) CEA preoperatively normal, (3) absence of hilar or mediastinal lymph node infiltration, and (4) single pulmonary metastasis. Patients presenting with both hepatic and pulmonary metastases have a 5-year survival rate after lung metastasectomy ranging from 11 to 61%^[64]. These results are comparable with patients who underwent only pulmonary metastasectomy. Nevertheless, none of these factors is an absolute contraindication to pulmonary resection. Repeated pulmonary resections also have encouraging results, with survival at 5 years between 29 and 85%^[58–60]. With the development of new chemotherapeutic agents, the progression-free survival and overall survival have been dramatically increased in stage IV colorectal cancer, and pulmonary metastasectomy is currently questioned. Cancer Research UK has began recently a randomized, controlled trial PulMiCC (Pulmonary Metastasectomy in Colorectal Cancer) in patients with pulmonary metastases comparing treatment with chemotherapy alone versus surgery combined with chemotherapy^[68]. Moreover, in metastatic colorectal cancer, additional systemic chemotherapy should be considered. The exact role of combined liver and lung metastasectomy in the context of modern chemotherapeutic modalities is also currently under discussion.

Soft tissue sarcoma

Between 20 and 50% of patients with soft tissue sarcoma will develop pulmonary metastases. Frequently, the lung is the only metastatic site of sarcoma, and death usually occurs due to uncontrolled intrathoracic disease. Many studies have suggested a survival benefit in patients who undergo pulmonary metastasectomy. These tumors are generally poorly chemosensitive, and pulmonary metastasectomy is recommended as long as complete surgical resection is feasible. Survival at 5 years varies from 29 to 52%^[69–73]. The principal prognostic factor of survival is the completeness of resection. Poor prognostic factors found are (1) high-grade tumor, (2) tumor of >5 cm, (3) multiple metastases, (4) a bilateral disease and (5) a short disease-free interval.

Osteosarcoma

Treatment of osteosarcoma is nowadays multidisciplinary. It includes the combination of chemotherapy and surgery. If lung metastases are synchronous, chemotherapy is started, and lung metastasectomy is performed after resection of the primary tumor. Postoperatively, surgery is usually combined with systemic chemotherapy. The 5-year survival rate varies from 7 to 65%^[74–76]. Completeness of resection, long disease-free survival interval, and number of metastases have been described as prognostic factors for survival. Repeat pulmonary metastasectomy for osteosarcoma is also considered a viable option, and remissions or possible cure may be obtained even in patients who experience two or more lung relapses.

Renal cell carcinoma

Several studies have shown survival benefit after pulmonary metastasectomy in renal carcinoma^[77]. Two studies have shown a 5-year survival of 45 and 42% after complete resection in comparison with 8 and 22% in the case of incomplete resection^[55,78]. Hilar or mediastinal lymph node invasion indicated poorer prognosis. A recent study published by the Mayo Clinic showed a 5-year survival of 74% after pulmonary metastasectomy compared to 19% for incomplete resection^[77]. Current indications for systemic chemotherapy following metastasectomy are questionable in this nonchemosensitive disease. To date, targeted therapy have some effectiveness for palliation.

Melanoma

More than 30% of patients with a malignant melanoma will develop lung metastases, associated with a poor prognosis^[79]. In solitary pulmonary lesions, pulmonary metastasectomy may offer a benefit on survival, with 5-year survival ranging from 21 to 35%. Complete resection is also an important prognostic factor of survival, with 5-year survival of 21% and median survival of 19 months in comparison with 13% and 11 months, respectively, for incomplete resection^[80]. Disease-free interval of less than 12 months and invasion of thoracic lymph nodes confer poorer prognosis. Systemic chemotherapy following surgery has not proven efficient.

Nonseminomatous testicular germ cell tumor

Pulmonary metastases are frequently encountered in patients with nonseminomatous testicular germ cell tumors^[81,82]. Resection of the residual pulmonary lesion is indicated after chemotherapy when there is normalization of tumor markers (beta-HCG, alpha-fetoprotein). Residual pulmonary lesions after chemotherapy may result in three distinct lesions which cannot be differentiated radiologically: (1) necrotic lesions, (2) viable tumor, and (3) mature teratoma. Nevertheless, complete resection of viable tumor or mature teratoma is associated with a 5-year survival if from 79 to 94%. Viable tumor and an absence of tumor marker normalization after chemotherapy are associated with poorer prognosis. Pulmonary metastasectomy may also be indicated in cases of poor response to chemotherapy when complete resection is feasible.