The effect of lobar versus sublobar video-assisted thoracoscopic surgery lung resection on patient quality of life

Introduction

Lung cancer is one of the most common forms of cancer and the leading cause of cancer mortality (1). The World Health Organization Global Cancer Observatory estimates that lung cancer incidence is 2.26 million cases resulting in 1.8 million lung cancer deaths worldwide annually (2). Surgical resection is a central component of cancer treatment for eligible patients. Two key factors are generally considered when planning a pulmonary resection: (I) surgical access, meaning an open thoracotomy versus minimal access techniques like video-assisted or robot-assisted thoracic surgery (VATS or RATS), and (II) extent of pulmonary resection which can include but is not limited to wedge resection, segmentectomy, lobectomy, bilobectomy or pneumonectomy. Minimal access techniques have grown in popularity since being introduced 30 years ago and are considered standard for the removal of a wide array of pulmonary malignancies (3-5).

A Japanese phase 3 non-inferiority trial (JCOG0802) of 1,106 patients compared lobectomy to segmentectomy and suggests that the overall survival of the patients with small-peripheral non-small cell lung cancer (NSCLC) is improved when they receive a segmentectomy (6). These data suggest that segmentectomy may be associated with longer overall survival when compared with lobectomy for early-stage lung cancer and are accompanied by equivalent survival safety profiles (7,8). This was reinforced by a recent study from Altorki et al. (2023) that demonstrated non-inferiority of sublobar (segmentectomy or wedge) resection as compared to lobar resection (9). Notably, both studies demonstrate marginal differences in residual lung function between lobar and sublobar resections. If survival and lung function are comparable, what other outcome variables could be considered to assist surgeons with decision-making between the available surgical options?

Patient quality of life (QoL) is emerging as a patient-centric outcome measure of great importance in the assessment of surgical therapeutic efficacy. The phase 3 VIOLET trial from Lim et al. (2022) compared open procedures to VATS by using QoL as a primary outcome. They found that VATS was associated with significantly better QoL as compared to open lobectomy as a composite measure during the first year after surgery (10). However, at 3 months and beyond post-operation, patients who had VATS or open lobectomy demonstrated a similar return to near baseline levels of QoL. This suggests that the short-term postoperative period is when patient QoL is most likely to be observed with respect to the impact of a surgical intervention.

We lack data on how QoL varies across the spectrum of extent of lung resection performed by minimal access surgery. Such data could provide important guidance for surgeons and patients when discussing the procedure of choice. While studies have evaluated pneumonectomy versus lobectomy and open versus minimally invasive procedures, there exists minimal investigation regarding the QoL impact of wedge versus segmentectomy versus lobectomy (11,12). Therefore, this retrospective study aims to better understand the potential impact of the extent of lung resection on QoL after surgery amongst patients undergoing VATS lung resection in a center with a standardized and uniformly applied enhanced recovery protocol. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-260/rc).

Methods

Study population

This retrospective study was conducted at a tertiary referral center affiliated with McGill University in Montréal, Québec, Canada. The data is sourced from the McGill University Thoracic Oncology Clinical Database and Biobank (TOCDB), covering the period from November 14, 2017, to March 17, 2021. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the McGill University Health Centre Research Ethics Board (No. 2014-1119). Informed consent was taken from all participants. To be included, patients must have undergone a segmentectomy, wedge resection, or lobectomy using a VATS approach and have completed a baseline and postoperative QoL questionnaire during their follow-up visit to the clinic. Patients who underwent complex procedures requiring an atypical approach or not clearly defined approach, received neoadjuvant treatment, individuals without a baseline or postoperative questionnaire, and patients who underwent an open or converted procedure were excluded. At our center, surgeons systematically perform VATS procedures with a two- or three-port approach and postoperative pain management is standardized with intercostal and/or erector spinae nerve bloc(s) during the procedure and utilize opioid sparing protocols.

Study design

At each visit, patients were approached to partake in a departmental initiative to collect data on patient QoL by means of a multi-assessment paper questionnaire. Those who provided written consent to participate in the database received a questionnaire at every appointment, typically starting at the preoperative appointment and receiving a follow-up questionnaire at their 1-month postoperative follow-up appointment. Questionnaires were provided in both English and French to meet the needs of the study population. Each patient was assigned a unique anonymized identifier, and questionnaires were stored on site in a secure location. Questionnaire data were then transcribed to a digital format for statistical analysis.

Edmonton Symptom Assessment Scale (ESAS) (Figure S1)

The ESAS is a self-reported analogue pain assessment score allowing patients to respond to nine items (13). This metric has been used for an extensive period of time, has been validated by multiple groups, and has been translated into over twenty languages for official use (14). In addition to the nine items, respondents had the opportunity give a written response to additional issues they wanted addressed; the research team did not consider these in analysis and viewed it as a note for the clinical staff only. A scale of 0–10 is used with higher values correlating to higher severity of the symptom. A change of 1 point is deemed clinically significant (15).

Functional Assessment of Cancer Therapy-Lung (FACT-L) (Figure S2)

The FACT-L version 4 is a patient-reported 36-item questionnaire that measures multidimensional QoL and has been validated on numerous accounts (16). The assessment is broken down into 5 themes: physical well-being, functional well-being, emotional well-being, social/family well-being, and lung cancer specific questions. Each question is a 5-point self-reported Likert scale. A composite FACT-L score in the range of 0–136 is calculated across all sections, with higher scores correlating to higher QoL. Composite scores factor out skipped questions and are normalized.

Demographics

Demographic data was extracted from the patients’ electronic medical records and compiled for anonymized aggregate analysis. Sex, age, smoking status, pulmonary function test (PFT) results, and tumor staging were collected to characterize the study population. Tumor staging was categorized using the Eighth edition of the American Joint Commission on Cancer TNM staging system for NSCLC. PFTs were performed preoperatively and evaluated using metrics of low forced expiratory volume in one second (FEV1), forced vital capacity (FVC), and diffusing capacity of the lung for carbon monoxide (DLCO). If baseline assessment was below 70%, indicating emphysema, a bronchodilator was used prior to re-evaluation. Clinical variables such as length of stay (LOS) and operative times were collected from chart reviews of the included patients. Complication data was retrieved from our Canadian Association of Thoracic Surgeons prospectively recorded national database of thoracic surgery quality improvement metrics.

Statistical analysis

Statistics were performed with Prism 9 by GraphPad (2022, version 9.4.1; CA, USA) and P<0.05 was considered significant. Demographic and QoL data were analyzed by calculating mean scores, standard deviations, interquartile ranges and P values via univariate analyses. Surgical outcome variables such as complications, LOS, and operative times were compared across cohorts using analysis of variance (ANOVA) or Kruskal-Wallis tests. Linear regressions and Pearson correlation coefficients were used to compare variables. Figures were generated with the same software. Multiple linear regression models comparing QoL scores between types of surgery with beta coefficients, confidence intervals, and P values being calculated using Stata Software (version 14.2; College Station, USA).

Results

Overall, 134 patients from a database of 1,305 met the inclusion criteria for this study (Figure 1). The mean age of patients was 68 (range, 28–90) years and 39% (52/134) of patients were male. Of all patients, 76% (102/134) had a smoking history and 27% (36/134) were actively smoking at the time of surgical consultation. Patients receiving lobectomy accounted for 65% (87/134) of surgeries performed, with segmentectomy representing 17% (23/134) and wedge resection 18% (24/134). Stage I disease represented 73% (98/134), stage II 19% (26/134), stage III 8% (10/134) (Table 1). Notably, 69.6% (16/23) of the segmentectomies were unisegmental resections and 13.0% (3/23) were classified as left S1, S2, S3 resections by the Boyden classification (17) (Table S1).

Figure 1 Study inclusion criteria and screening. VATS, video-assisted thoracoscopic surgery.

Patients undergoing lobectomy had a significantly longer operative time with a mean of 109 (range, 42–226) minutes when compared to the mean of 60 (range, 20–138) minutes for patients receiving a wedge and 83 (range, 40–98) minutes for patients receiving a segmentectomy (P<0.0001; Table 2). There was no difference in complications between the cohorts (P=0.52). However, LOS was significantly different between sublobar and lobar approaches with patients receiving a lobectomy having a median LOS of 3 days [interquartile range (IQR), 2–4 days] vs. 2 days (IQR, 1–2 days) for segmentectomy procedures and 1 day (IQR, 1–2 days) for patients receiving a wedge resection (P=0.01; Table 2). Despite these differences between cohorts, univariable linear regression analysis and multiple linear regression models did not reveal any correlation between operative time or LOS with changes in QoL (Figure 2; Figure S3).

Figure 2 Correlation of FACT-L changes to LOS & operative time. Linear regressions assess the correlation of change in FACT-L scores with (A) LOS and (B) operative times. Pearson correlations coefficients are also reported. FACT-L, Functional Assessment of Cancer Therapy-Lung; LOS, length of stay.

Patients completed their questionnaires a median of 34 days (IQR, 23–52 days) before their surgery, and a median of 27 days (IQR, 22–34 days) after their surgery. Patients on average completed 95% of questions, with 63% of questionnaires being completed in their entirety.

For those who underwent a wedge resection, the average preoperative FACT-L score was 102.8 [standard deviation (SD): 20.65], with the average postoperative score being 104.9 (SD: 21.28; P=0.64; Table 3, Figure S4). The only statistical difference found in the ESAS variables was a decrease in anxiety from 2.42 (SD: 2.9) pre-op to 1.33 (SD: 2.2) post-op (P=0.02; Table 4, Figure S5).

Patients who underwent a segmentectomy demonstrated no significance in FACT-L QoL when comparing pre- and postoperative scores (P=0.42). Scores for ESAS pain was significantly different (0.9 vs. 3.2; P<0.001; Table 4) as well as fatigue (2.3 vs. 4.2; P=0.01) and shortness of breath (2.3 vs. 3.9; P=0.01).

Patients who underwent a lobectomy had a statistically significant decrease in FACT-L scores (104.9 vs. 93.8; P<0.001). When combining both wedge and segmentectomy FACT-L scores as sublobar procedures, FACT-L scores demonstrate no significance when compared to the lobar group (Figure S4). As shown in Table 4, ESAS scores for patients who underwent lobar section had statistically significant increases in pain (1.5 vs. 3.1; P<0.001), fatigue (2.7 vs. 3.8; P=0.01), depression (0.6 vs. 1.1; P=0.03), loss of appetite (2.5 vs. 3.6; P=0.045), and shortness of breath (2.1 vs. 3.9; P<0.001).

We also ran an analysis of only the Stage I patients who received a lobectomy to assess FACT-L scores, but this analysis similarly found the same differences in QoL (P<0.0001; Table S2). Stage I patients for patients undergoing a wedge or segmentectomy remained non-significant (P=0.97 and P=0.89, respectively; Table S2).

Multiple linear regression models comparing the extent of lung resection by FACT-L scores found that there was no significance between lobectomy vs. segmentectomy (P=0.85; Table 5) and wedge resections vs. segmentectomy (P=0.07). Lobectomy vs. wedge was significant (P=0.01). These models additionally assessed the impact that differences in LOS, complications, pathological stage, and date of completing the survey between groups influenced QoL results. This analysis revealed no major differences in results, further reinforcing that these possibly confounding variables do not affect the trends we observed.

Discussion

This study aimed to provide thoracic surgeons with some measure of the impact on QoL by extent of pulmonary resection amongst patients undergoing minimal access surgery and treatment using a standard recovery pathway for early-stage lung cancer. Our data suggest that patients undergoing VATS lobectomy have a greater initial postoperative decrease in QoL and that QoL may be better preserved for patients undergoing sublobar resection. Wedge resection had the least impact on QoL of the three operative techniques evaluated. Segmentectomy patients experienced an increase in pain, fatigue, and shortness of breath between timepoints. Both operative time and LOS for segmentectomies did not correlate to these symptoms (Figure S3). Furthermore, when analyzed together, no difference in QoL was observed between sublobar and lobar resections. Those who underwent a lobectomy not only had significant increases in five measured symptoms but also demonstrated an overall decrease in well-being as per the FACT-L. Validation studies of the FACT-L have shown that a clinically meaningful change is 7–9 points (18). We only observed a clinically meaningful change in our lobectomy cohort (11-point change).

Understanding this difference may help surgeons guide patients through the early postoperative period. Although we do not have data on QoL going beyond the first postoperative visit, we suspect that recovery would follow a similar trajectory as described in the VIOLET trial, where QoL scores returned to baseline by 3 months after surgery (10). Another recent study from Jiang et al. (2023) found similarly that patients undergoing both lobar and sublobar resections were comparable by the 3-month timepoint for surgical site pain and chest pain, but these data are not presented with a baseline-preoperative time-point, making changes in patient QoL difficult to assess (19).

Performing a sublobar or lobar resection is a recurrent debate in the case of peripheral lesion of less than 2 cm, without lymph node involvement. One argument in support of sublobar resections is the preservation of the lung parenchyma. However, the expected segmentectomy FEV1 benefit at 1 year is low (4–7%), without being able to know if it clinically improves the QoL of our patients (20). The JCOG0802 study found that between segmentectomies and lobectomies, the median FEV1 reduction was 2.7% at 6 months and 3.5% at 12 months (6). Similarly, the CALGB140503 study found a reduction, after 6 months following lobar resection, of FEV1 to be 6% and 4% in sublobar resections (9). The differences in PFTs between patients in their lobectomy and sublobar/segmentectomy cohorts were surprisingly low. Many surgeons have commented that with equivalent survival, it may be preferable to perform a lobectomy given the minimal difference in PFT reduction. These studies did not capture all aspects of cardiopulmonary function and it is possible that the existing data under-represent the impact of these procedures on the patient. Recent data presented by JCOG0802 suggests that non-lung cancer related patient mortality increases with each additional segment resected, further supporting that QoL and the life of the patient may be better preserved with a sublobar resection (21).

In our study, 87.5% (n=21) of patients who underwent wedge resection had Stage I disease (n=21), reflecting the peripheral and smaller nature of these tumors that allow for safe margins. However, as shown in Table 3, comparisons within the wedge cohort revealed no significant changes in QoL using the FACT-L. Similarly, in Table 5, controlling for pathological stage and other variables did not alter results. Because our dataset contains few patients with advanced disease, we do not believe intergroup comparisons by stage and operative approach would be interpretable, given the inherent differences between patients eligible for wedge resection versus those requiring more invasive surgery or neoadjuvant therapy—factors we accounted for when defining our study population.

Although limited by their retrospective, non-randomized nature, our data suggest that the extent of lung resection does have an impact on short term QoL metrics. These findings may justify the generation of new randomized data focusing on the QoL impact of pulmonary extent of surgery. These data may inform the practice of thoracic surgeons to consider less extensive resections when feasible as a means of minimizing the QoL impact of surgery when expected survival is deemed to be equivalent for the available procedures.

While there exists little data about impact of extent of resection on QoL, researchers have investigated other aspects of QoL in thoracic surgery. One group comparing thoracotomies versus VATS lobectomies found a decrease in QoL around post-op week 2 but following week 4, there was a steady rise in QoL (22). It is possible that our cohort will experience this same delayed rise in QoL. The short term postoperative period we characterized is crucial as this period is typically where operative complications arise (23). One systematic review found that patients who underwent lobar resection had a higher complication rate providing possible evidence as to why QoL is better preserved in sublobar patients (24). Our data did not find this same significance between the procedures, possibly due to sample size. Additionally, our analysis shows no correlation between LOS or operative times on changes in QoL scores. LOS, as used in our study, is a surrogate of the multifactorial nature of the recovery process; it can account for many aspects of the patients’ postoperative care pathway, including but not limited to postoperative complications, the psychological state of the patient, and hospital discharge protocols. This leads us to believe that the decrease in QoL for lobar procedures is a result of the procedure rather than postoperative complications, pathological staging, LOS, or operative time.

We acknowledge that there are important limitations to our study. As previously mentioned, Hui et al. (2015) found for the ESAS that a change of 1 point is deemed clinically significant and 7–9 points are clinically relevant on the FACT-L (15,18). We believe that the clinical significance is challenging to determine on an individual patient level, but that group population analysis, such as ours, is sound in determining trends in patient QoL. Our sublobar group’s sample size could present a risk of type 2 error, as we instituted strict variability control and inclusion criteria. Our selected patients did not receive any neoadjuvant treatments nor undergo complex segmentectomies and needed a high survey completion rate to ensure the research team had the proper data to assess QoL changes as a result of the surgical approach being examined. We recognize that our study has only two timepoints. However, we believe that these changes in the short-term postoperative time frame are clinically relevant and important to report, given the paucity of data on this important subject, and they may provide the foundation for future studies with longer follow-up and additional timepoints to enhance our understanding of how thoracic procedures alter patient QoL.

Conclusions

Our data suggests that patients who underwent a lobectomy suffered a clinically and statistically significant deterioration in QoL at their first postoperative visit, while patients who underwent wedge and segmental resections did not, or did so to a much lesser extent. Future longitudinal assessments and trials should be carried out to prospectively validate our findings. As our understanding of the oncologic impact of these surgical options evolves, it is increasingly important to address how they impact the QoL of our patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-260/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-260/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-260/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-260/coif). J.D.S. reports a relationship with AstraZeneca Canada Inc that includes: consulting or advisory, funding grants, speaking and lecture fees, and travel reimbursement; a relationship with Bristol Myers Squibb Co that includes: consulting or advisory, funding grants, speaking and lecture fees, and travel reimbursement; a relationship with Merck & Co Inc that includes: consulting or advisory, funding grants, speaking and lecture fees, and travel reimbursement; a relationship with Regeneron Pharmaceuticals Inc that includes: consulting or advisory and travel reimbursement; a relationship with Amgen Inc that includes: consulting or advisory; a relationship with Eisai Inc that includes: consulting or advisory; a relationship with DAIICHI SANKYO COMPANY, LIMITED that includes: consulting or advisory; a relationship with F Hoffmann-La Roche Ltd that includes: consulting or advisory; a relationship with Pfizer that includes: speaking and lecture fees. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the McGill University Health Centre Research Ethics Board (No. 2014-1119). Informed consent was taken from all participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424. [Crossref] [PubMed]
2. World Health Organization International Agency for Research on Cancer. Estimated number of incident cases and deaths World. 2024. Available online: https://gco.iarc.fr/today/online
3. Servais EL, Blasberg JD, Brown LM, et al. The Society of Thoracic Surgeons General Thoracic Surgery Database: 2022 Update on Outcomes and Research. Ann Thorac Surg 2023;115:43-9. [Crossref] [PubMed]
4. Zirafa CC, Romano G, Key TH, et al. The evolution of robotic thoracic surgery. Ann Cardiothorac Surg 2019;8:210-7. [Crossref] [PubMed]
5. Begum S, Hansen HJ, Papagiannopoulos K. VATS anatomic lung resections-the European experience. J Thorac Dis 2014;6:S203-10. [Crossref] [PubMed]
6. Saji H, Okada M, Tsuboi M, et al. Segmentectomy versus lobectomy in small-sized peripheral non-small-cell lung cancer (JCOG0802/WJOG4607L): a multicentre, open-label, phase 3, randomised, controlled, non-inferiority trial. Lancet 2022;399:1607-17. [Crossref] [PubMed]
7. Suzuki K, Saji H, Aokage K, et al. Comparison of pulmonary segmentectomy and lobectomy: Safety results of a randomized trial. J Thorac Cardiovasc Surg 2019;158:895-907. [Crossref] [PubMed]
8. Altorki NK, Wang X, Wigle D, et al. Perioperative mortality and morbidity after sublobar versus lobar resection for early-stage non-small-cell lung cancer: post-hoc analysis of an international, randomised, phase 3 trial (CALGB/Alliance 140503). Lancet Respir Med 2018;6:915-24. [Crossref] [PubMed]
9. Altorki N, Wang X, Kozono D, et al. Lobar or Sublobar Resection for Peripheral Stage IA Non-Small-Cell Lung Cancer. N Engl J Med 2023;388:489-98. [Crossref] [PubMed]
10. Lim E, Batchelor TJP, Dunning J, et al. Video-Assisted Thoracoscopic or Open Lobectomy in Early-Stage Lung Cancer. NEJM Evid 2022;1:EVIDoa2100016.
11. Brunelli A, Pompili C, Koller M. Changes in quality of life after pulmonary resection. Thorac Surg Clin 2012;22:471-85. [Crossref] [PubMed]
12. Singer ES, Kneuertz PJ, Nishimura J, et al. Effect of operative approach on quality of life following anatomic lung cancer resection. J Thorac Dis 2020;12:6913-9. [Crossref] [PubMed]
13. Bruera E, Kuehn N, Miller MJ, et al. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care 1991;7:6-9.
14. Hui D, Bruera E. The Edmonton Symptom Assessment System 25 Years Later: Past, Present, and Future Developments. J Pain Symptom Manage 2017;53:630-43. [Crossref] [PubMed]
15. Hui D, Shamieh O, Paiva CE, et al. Minimal clinically important differences in the Edmonton Symptom Assessment Scale in cancer patients: A prospective, multicenter study. Cancer 2015;121:3027-35. [Crossref] [PubMed]
16. Butt Z, Webster K, Eisenstein AR, et al. Quality of life in lung cancer: the validity and cross-cultural applicability of the Functional Assessment Of Cancer Therapy-Lung scale. Hematol Oncol Clin North Am 2005;19:389-420. viii. [Crossref] [PubMed]
17. Boyden EA. Segmental Anatomy of the Lungs. McGraw-Hill; 1955.
18. Cella D, Eton DT, Fairclough DL, et al. What is a clinically meaningful change on the Functional Assessment of Cancer Therapy-Lung (FACT-L) Questionnaire? Results from Eastern Cooperative Oncology Group (ECOG) Study 5592. J Clin Epidemiol 2002;55:285-95. [Crossref] [PubMed]
19. Jiang S, Wang B, Zhang M, et al. Quality of life after lung cancer surgery: sublobar resection versus lobectomy. BMC Surg 2023;23:353. [Crossref] [PubMed]
20. Charloux A, Quoix E. Lung segmentectomy: does it offer a real functional benefit over lobectomy? Eur Respir Rev 2017;26:170079. [Crossref] [PubMed]
21. Tsutani Y, Okada M, Wakabayashi M, et al. The Association Between the Incidence of Non-Lung Cancer Death and the Number of Resected Segments and Postoperative Pulmonary Function from the Randomized Phase 3 Trial - JCOG0802/WJOG4607L. AATS 104th Annual Meeting. 2024.
22. Bendixen M, Jørgensen OD, Kronborg C, et al. Postoperative pain and quality of life after lobectomy via video-assisted thoracoscopic surgery or anterolateral thoracotomy for early stage lung cancer: a randomised controlled trial. Lancet Oncol 2016;17:836-44. [Crossref] [PubMed]
23. Bédat B, Abdelnour-Berchtold E, Perneger T, et al. Comparison of postoperative complications between segmentectomy and lobectomy by video-assisted thoracic surgery: a multicenter study. J Cardiothorac Surg 2019;14:189. [Crossref] [PubMed]
24. Divisi D, De Vico A, Zaccagna G, et al. Lobectomy versus sublobar resection in patients with non-small cell lung cancer: a systematic review. J Thorac Dis 2020;12:3357-62. [Crossref] [PubMed]