Surgically resectable disease

Approach

Since the early 2000s, thoracic surgical oncology has witnessed the adoption of minimally invasive techniques. The benefits of video-assisted thoracic surgery (VATS) ( Fig. 5.1 ) and, more recently, robotic-assisted thoracic surgery (RATS) ( Fig. 5.2 ) are well documented—decreased length of stay, decreased postoperative pain, decreased blood loss, and less impairment in pulmonary function, with similar perioperative complication rates. A body of literature emerged in the early 2010s suggesting minimally invasively hilar lymphadenectomy to be potentially substandard, evidenced by a lower number of N1 lymph nodes and rate of nodal upstaging from N0 to N1 in patients undergoing VATS lobectomy compared with thoracotomy (see Fig. 5.1 ). ^{, , ,} Boffa and colleagues used the Society of Thoracic Surgeons’ General Thoracic Surgery Database (STS-GTSD) linked with Medicare data to evaluate long-term outcomes from >10,000 patients with stage I lung cancer who underwent lobectomy via open thoracotomy or minimally invasive approaches. In a propensity-matched cohort, they found perioperative and long-term survival in patients undergoing VATS resection to be noninferior to those patients with open thoracotomies. Furthermore, they found a shorter interval time between resection and initiation of adjuvant therapy in those undergoing VATS resection. Other recent studies, including one by Yang and colleagues using a propensity-matched cohort from the National Cancer Database, confirmed these findings. Minimally invasive resection of early-stage lung cancer should be considered safe, oncologically acceptable, and preferable to open thoracotomy due to equivalence in long-term outcomes and improved short-term toxicity and tolerability. Introduction of robotic techniques to thoracic surgery appears to have greatly facilitated the transition from open to minimally invasive approaches for early-stage tumors; to date, there is no significant evidence to suggest any difference in outcomes between VATS and RATS approaches.

Nodule localization

Nodule localization is an issue unique to minimally invasive procedures, due to the inability to directly palpate parenchyma that is afforded by thoracotomy. Small lesions, except when located on the periphery, are often unable to be palpated or visualized with the unaided eye. During video- or robotic-assisted resections, haptic feedback is further hindered, often leaving the surgeon unable to localize the targeted lesion and relying solely on preoperative imaging. This places the patients at risk for nondiagnostic, nontherapeutic resections that contain only lung parenchyma. A variety of technologies are available and others are newly emerging to help mitigate this problem and aid the surgeon in lesion localization, thereby facilitating minimally invasive resections. ^, Localization options include imaging accessories such as thoracoscopic ultrasound; physical localization with the use of microcoils, hook wires, and other fiducials; liquid or tattoo markers such as barium, methylene blue, or contrast used in conjunction with near-infrared imaging (NIR), synthetic fluorophores, and molecular targeting ( Fig. 5.3 ).

Each of these techniques has limitations, and no single method is used universally. Intraoperative thoracoscopic ultrasound is attractive, requiring no additional radiation exposure or separate procedures, but has a steep associated learning curve that has hindered widespread adoption. There is a growing body of evidence and support for preoperatively placed image-guided fiducials. ^, McGuire and colleagues recently reported their 5-year experience with CT-guided microcoil placement prior to thoracoscopic nodule resection. Successful nodule localization at the time of surgery was 100%. They did report a nearly 60% complication rate associated with coil placement, the majority of which were small pneumothoraces and lung hematomas; however, all were minor and self-limited in nature, not requiring additional intervention. While similar localization techniques have reached widespread popularity in breast surgery, and the advantages seemingly apparent, it remains to be seen if fiduciary-guided localization reaches the mainstream in thoracic surgery.

The use of navigational bronchoscopy with dye or fiducial placement is a popular localization technique. It can be performed in the same OR setting and has a decreased pneumothorax rate compared with percutaneous techniques, however requires upfront planning and capital expense associated with the navigational system. Further work is needed to improve time requirements and safety profiles while lowering barriers to entry for surgeons and radiologists.

Extent of resection

For patients with nonprohibitive lung function, removal of the involved lobe in its entirety, with negative margins, has long been accepted as the oncological standard of care for patients with early-stage NSCLC. As the only randomized trial comparing lobectomy to sublobar resection for stage I NSCLC fully reported to date, the results of the Lung Cancer Study Group trial, published in 1994, formed the basis of that standard. Limited resection was associated with an increased rate of local recurrence. Despite the findings of the Lung Cancer Study Group, multiple retrospective and cohort studies have identified sublobar resection to have acceptable oncological outcomes, with most measurable endpoints for T1a tumors to be equivalent between anatomical segmentectomy and lobectomy. A 2018 meta-analysis pooled results from 11 high-quality studies examining postoperative morbidity and 15 studies examining 30-day mortality, finding no significant difference in perioperative morbidity (odds ratio [OR], 1.18; 95% [CI], 0.76−1.83; P = 0.47) or 30-day mortality (OR, 1.39; 0.54−3.60; P = 0.5) when comparing sublobar resection with lobectomy. Paired with an increased understanding that early-stage, and even T1, lung cancer is a heterogeneous pathology with disparate outcomes, the question of both surgical outcomes and longer-term oncological appropriateness of sublobar resection are being revisited.

The CALGB/Alliance trial 140503 is a multicenter, randomized, prospective noninferiority trial comparing lobectomy with sublobar resection for patients with T1aN0 tumors. General thoracic surgeons at 69 academic- and community-based medical centers spanning the United States, Canada, and Australia randomized 697 patients with tumors ≤2 cm to lobar or sublobar resection. Of note, and importantly, sublobar resection technique (wedge resection or anatomical segmentectomy) was left to the discretion of the operating surgeon. Nearly 60% underwent wedge resection alone. A large portion, almost 80%, were performed with minimally invasive techniques. Post-hoc analysis examining perioperative morbidity and mortality found no significant differences in perioperative morbidity or 90-day mortality. After a median follow-up of 7 years, sublobar resection was found to be noninferior to lobar resection for the primary endpoint of disease-free survival (HR 1.01). Overall survival at 5 years was also similar between sublobar resection (80.3%) and lobar resection (78.9%). No significant differences existed between the groups with regards to the incidence of locoregional or distant recurrence, and interestingly at 6 months there was also only a minimal difference in the percent of predicted forced expiratory volume in 1 second favoring the sublobar group by 2%. Important questions remain as to the broad applicability of the data, in particular in light of the recurrence rates at 5 years in both cohorts exceeding 35%, suggesting that perhaps considerations beyond size, such as tumor biology, may be critical in helping guide extent of resection.

The Japanese Oncology Group and Western Japanese Oncology Group trial (JCOG0802/WJOG4607L) similarly randomized 1100 patients at 70 institutions with tumors <2 cm (exclusive of pure ground-glass opacities; consolidation-to-tumor ratio >0.5) to lobectomy or segmentectomy. Accrual completed in 2014. The baseline clinicopathological criteria were well balanced between the groups. There were no 30- or 90-day mortalities observed, and postoperative morbidity was similar between the two groups. Both non-inferiority and superiority were confirmed in overall survival favoring segmentectomy (HR 0·663; 95% CI 0·474–0·927). Local recurrence was similar between both groups at 5 years; however, the segmentectomy group was more likely to recur locoregionally (10.5% vs 5.4%). The segmentectomy cohort experienced a lesser reduction in median forced expiratory volume in 1 second; however, this did not reach the predefined threshold for clinical significance.

Mediastinal lymph node evaluation

The presence of pathologically involved lymph nodes in patients with early-stage lung cancer has remained the most important predictor of prognosis. Accurate lymph node evaluation is an essential component of staging and treatment of early stage NSCLC. There is growing appreciation of a significant quality gap in both preoperative and intraoperative nodal staging in NSCLC.