Video-assisted thoracoscopic lobectomy is feasible for selected patients with clinical N2 non-small cell lung cancer

Few studies have evaluated the usefulness of video-assisted thoracoscopic surgery (VATS) for advanced-stage lung cancer. We aimed to evaluate the feasibility of VATS for treating clinical N2 (cN2) lung cancer. A retrospective cohort analysis was performed with data from 268 patients who underwent lobectomy for cN2 disease from 2007 to 2016. Using propensity score-based inverse probability of treatment weighting (IPTW), perioperative and long-term survival outcomes were compared. We performed VATS and open thoracotomy on 121 and 147 patients, respectively. Overall, VATS was preferred for patients with peripherally located tumors (p < 0.001). After IPTW-adjustment, all preoperative information became similar between the groups. Compared to thoracotomy, VATS was associated with shorter hospitalization (7.7 days vs. 9.1 days, p = 0.028), despite equivalent complete resection rates (92.6% vs. 90.5%, p = 0.488) and dissected lymph nodes (mean, 31.9 vs. 29.4, p = 0.100). On IPTW-adjusted analysis, overall survival (50.5% vs. 48.4%, p = 0.127) and recurrence-free survival (60.5% vs 44.6%, p = 0.069) at 5 years were also similar between the groups. Among selected patients with resectable cN2 disease and peripherally located tumors, VATS is feasible, associated with shorter hospitalization and comparable perioperative and long-term survival outcomes, compared with open thoracotomy.

perioperative data. Table 2 presents the operative profiles of the two groups. The mean numbers of harvested LNs (31.9 vs. 29.4, p = 0.100) and the rates of complete resection (92.6% vs. 90.5%, p = 0.488) were not significantly different between the VATS and thoracotomy groups. The mean duration of hospital stay was shorter in the VATS group, regardless of IPTW adjustment (before adjustment: 7.3 days vs. 9.2 days, p = 0.001; after adjustment: 7.7 days vs. 9.1 days, p = 0.028). Additionally, the mean time to start adjuvant chemotherapy was also shorter in the VATS group (before adjustment: 36.7 days vs. 43.4 days, p = 0.036; after adjustment: 36.9 days vs. 43.1 days, p = 0.033). The rate of adjuvant chemotherapy was similar between the two groups overall (53.7% vs. 52.4%, p = 0.924); however, it became different after IPTW adjustment (54.5% vs. 46.9%, p = 0.033).
Early mortality (30-day or in-hospital) occurred in one patient each in the VATS (0.8%) and thoracotomy (0.7%) groups ( Table 3). The early death in the VATS group was caused by an aggravation of interstitial pulmonary fibrosis, and the early death in the thoracotomy group was caused by postoperative pneumonia. In terms of postoperative complications, there were no significant differences between the VATS and thoracotomy groups ( Table 3, Cohort 1). When patients were divided into three groups (VATS, planned thoracotomy, and thoracotomy conversion), the rate of pneumonia was the lowest in the VATS group (3.1%), whereas it was similar between the thoracotomy (7.5%) and conversion groups (8.7%).
Survival analysis. In the overall cohort before IPTW adjustment, 61 (50.4%) of the 121 patients in the VATS group and 81 (55.1%) of the 147 patients in the thoracotomy group had died at the end of the follow-up period; their 5-year overall survival (OS) rates were 49.6% and 52.6%, respectively. Recurrence events occurred in 47 and 56 patients in the VATS and thoracotomy groups, respectively; their 5-year recurrence-free survival (RFS) rates were 52.5% and 50.9%. The pattern of recurrence was not different between the two groups (Table S1).  www.nature.com/scientificreports/

Discussion
We evaluated the perioperative and long-term outcomes of NSCLC patients who received VATS versus open lobectomy for the treatment of clinical stage IIIA cancer due to suspected N2 node metastasis. The VATS approach was associated with a shorter length of hospitalization and time to initial postoperative chemotherapy compared with the open thoracotomy approach. Furthermore, the rate of complete resection, mean number of dissected LNs, and rate of postoperative complications were similar between the two groups. Importantly, on both IPTW-unadjusted and IPTW-adjusted analysis, the VATS group showed no significant observed difference in 5-year OS and a marginally better RFS, compared with the thoracotomy group. According to multivariable Cox analysis, the VATS approach was not a significant prognostic factor of OS and RFS among patients with cN2 disease. www.nature.com/scientificreports/ In our institution, upfront surgical resection followed by adjuvant therapy has been prospectively performed for patients with resectable N2 disease for 20 years 15 . Initially, open thoracotomy was usually performed for these patients. Later, the accumulation of clinical experience and advances in instrument development facilitated the application of VATS for patients with cN2 disease, and in 2016, the rate of VATS implementation for cN2 disease increased to 73.3% (22/30). However, compared with open thoracotomy, VATS lobectomy tended to be performed less frequently for male patients as well as those with a smoking history, squamous cell carcinoma, large tumors, and centrally located tumors (Table 1). These clinical findings are linked to each other solely by the central tumor factor. Thus, we concluded that VATS lobectomy for patients with cN2 disease was preferred for patients with peripherally located tumors.
One potential concern about the VATS approach is the possibility of insufficient LN dissection owing to the narrow visual field and limited range of instrumental movement. Previous studies have reported that the number of harvested LNs in the VATS group was fewer than that in the thoracotomy group 16,17 . However, recently, the number and quality of LNs harvested via VATS lobectomy were found to be similar to those harvested via open lobectomy 18,19 . In our study, there were no significant differences in the mean number of harvested LNs (31.9 vs. 29.4, p = 0.100) or positive LNs (4.6 vs. 4.7, p = 0.980) in patients with cN2 disease ( Table 2).
The other concern is increased morbidity and mortality among patients who undergo thoracotomy conversion from VATS. According to previous reports, unexpected thoracotomy conversion, with rates ranging from 5 to 23%, has been associated with greater perioperative morbidity compared with successful VATS completion 20,21 . In contrast, similar postoperative complications were observed between converted and planned thoracotomy patients 20,21 . A recent study based on the National Cancer Database also reported similar complication rates and long-term survival between patients with planned and converted thoracotomy for clinical N1 disease 9 . In our study, the rate of thoracotomy conversion was 19% (23/121), and the rate of emergency cases due to vascular injury was 13.0% (3/23). Also, the VATS, thoracotomy, and conversion groups had similar perioperative morbidity, early mortality (Table 3), and long-term mortality rates (Fig. 2).
Although there were no significant differences in OS and RFS between the two groups, it is worth mentioning that the differences became larger after IPTW adjustment (Figs. 2 and 3). The 5-year RFS rates of the VATS and thoracotomy groups were 52.5% and 50.9% (p = 0.762), respectively, whereas they were 60.5% and 44.6% (p = 0.069) on IPTW-adjusted analysis. Additionally, the VATS approach was associated with better RFS on multivariable Cox analysis (HR (95% CI) = 0.63 (0.42-0.95), p = 0.026). This phenomenon can be explained by comparing the differences in the pathologic findings before and after IPTW adjustment. As described in Table 2, the rate of pathologic N2 status was slightly higher in the VATS group than in the thoracotomy group (68.6% vs. 60.5%) but became similar after IPTW adjustment (67.8% vs. 66.0%). Also, the rate of adjuvant chemotherapy was similar between the two groups (53.7% vs. 52.4%) but became higher in the VATS group (54.5% vs. 46.9%) after IPTW adjustment. Therefore, according to our IPTW-adjusted analysis, the VATS approach was associated with better RFS than the thoracotomy approach among identical patients with cN2 disease, owing to the enhanced compliance of adjuvant chemotherapy.
This study had notable limitations. First, selection bias is inherent in a retrospective study from a single institution, although the data in this study were gathered prospectively. Second, as in most studies comparing two surgical approaches, there might be another selection bias stemming from the surgeons' preferences. We minimized the bias as much as possible by using a PS-based IPTW method. Third, the proportion of patients who received adjuvant chemotherapy, and the frequency of recurrence events were relatively low. Because our hospital is a tertiary referral center, some patients were followed-up at local hospitals, and we could not assess www.nature.com/scientificreports/ their postoperative information. However, given that similar clinical characteristics, rates of recurrence, and patterns of recurrence were shown between the two groups after IPTW, these unknown data cannot change our conclusions.
In conclusion, VATS may be a safe and feasible approach for patients with resectable cN2 disease and a peripherally located tumors, associated with a shorter duration of hospitalization and time to adjuvant chemotherapy initiation, as well as comparable perioperative and long-term survival outcomes, relative to open thoracotomy. When it is performed by surgeons who have adequate experience and who carry out careful surgical planning, VATS lobectomy can be an appropriate option for selected patients with clinical N2 disease without compromising oncologic efficacy.

patients.
All clinical records of patients who underwent surgery for NSCLC between January 2007 and December 2016 were retrospectively reviewed from a prospectively gathered lung cancer database at Asan Medical Center in Seoul, South Korea. During the study period, 6,145 patients with NSCLC underwent curativeintent surgery for primary lung cancer, and 681 (11.2%) of them were diagnosed with stage IIIA cN2 disease.
The exclusion criteria were: (1) concomitant malignancies (n = 23), (2) administration of neoadjuvant therapy (n = 109), (3) pneumonectomy or sublobar resection (wedge resection and segmentectomy) (n = 50), (4) distant metastasis (n = 95), and (5) history of lung cancer treatment in the past 2 years (n = 13). Because the VATS approach is not feasible for patients with advanced T stage, patients with clinical T3 and T4 disease were also excluded (n = 123) (Fig. 1). According to intent-to-treat analysis 22 , patients converted to thoracotomy were included in the VATS group. This study was performed in accordance with relevant guidelines and was approved by the Asan Medical Center Institutional Review Board. Informed consent was obtained from all participants in studies that involve human subjects.
preoperative and postoperative management. The preoperative staging workup consisted of medical history, physical examination, complete blood counts, blood chemistry with electrolytes, chest X-ray, computed tomography (CT) of the chest and upper abdomen, biopsy, bronchoscopy, pulmonary function testing, radionuclide bone scanning, and 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET). Brain CT or magnetic resonance imaging with contrast was conducted routinely as part of the staging workup 23 .
The following definition of cN2 disease was used: enlarged LNs (defined as ≥ 10 mm on the largest short axis) on chest CT or PET-positive LNs in an N2 position, according to the Mountain-Dresler modification of the American Thoracic Society (MD-ATS) LN map 24 . LNs were considered positive on PET if FDG uptake was higher than background uptake in the mediastinal blood pool. In cases of borderline tumor size or metabolic uptake of mediastinal LNs on CT or PET, biopsies of the involved mediastinal LNs were conducted, with mediastinoscopy, endobronchial ultrasound, or endoscopic ultrasound. Treatment approaches for cN2 disease were determined by a multidisciplinary team, including medical oncologists, radiologists, and thoracic surgeons. Upfront surgery for cN2 disease was generally considered when (1) the primary tumor was resectable without pneumonectomy, (2) mediastinal LN metastasis was confined within a single zone (upper zone for LNs at stations 2-4, aortopulmonary zone for stations 5 and 6, subcarinal zone for station 7, and lower zone for stations 8 and 9) 25 , (3) the involved LN was distinct from surrounding tissues and less than 3.0 cm in diameter, and (4) there was no sign of extra-nodal tumor invasion (the presence of full-thickness LN capsular invasion or extension of tumor cells beyond the LN capsule) on CT or PET.
Follow-up information on all patients was obtained through clinic follow-up notes every 3 months for the first 2 years, every 6 months for the next 3 years, and annually thereafter. Chest CT scans were performed at the time of clinical visits or at any time when disease recurrence was suspected. Treatment modalities and chemotherapeutic regimens in relapsed cases were determined at the discretion of the attending physician 23 .
Operative technique. The decision to perform VATS or open thoracotomy for cN2 disease was made by each of the four surgeons individually. However, the VATS approach was primarily considered for resectable cN2 disease, except when centrally located tumors or calcified LNs were very close to hilar structures on preoperative images. Under general anesthesia using one-lung ventilation by double-lumen endotracheal intubation, VATS was performed in the full lateral position with a 30° angled thoracoscope and the usual three-port technique. The incision sites varied slightly by surgeon but can be broadly classified into two methods. The first method involved making an approximately 4 cm utility incision in the mid-axillary line through the fourth or fifth intercostal space (ICS), a 1.2 cm camera port in the anterior axillary line through the seventh ICS, and a 1.5 cm incision in the posterior axillary line at the same level of the camera port. The second method consisted of a camera port at the mid-axillary line at the eighth ICS, an approximately 4 cm working port at the anterior axillary line at the fifth ICS, and another instrument port at the posterior axillary line at the sixth ICS. To decrease operative bleeding and avoid thermal injury along the nerve trajectory, bipolar or ultrasonic energy devices were routinely used for the VATS approach.
Operative procedures for cN2 disease included resection of the primary lung cancer and systemic LN dissection of the ipsilateral hilum and the mediastinum. Mediastinal LN dissection consisted of en bloc resection of all nodes at stations 2R, 4R, 7, 8, 9, 10R, and 11R for right-sided tumors and nodes at stations 5, 6, 7, 9, 10L, and 11L (selectively with 4L for the upper lobe, 8 and 9 for the lower lobe) for left-sided tumors, regardless of surgical approach. LN dissection with the adipose connective tissue of the related anatomic regions was performed, as determined intraoperatively by the surgeon. All LNs, either dissected intraoperatively or harvested later from specimens, were examined pathologically and classified based on anatomic location by the numbering system Scientific RepoRtS | (2020) 10:15217 | https://doi.org/10.1038/s41598-020-72272-4 www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.