 |
 |
 |
Trends and Outcomes following Video- versus Robotic-Assisted Thoracoscopic Lobectomy among Stage I-III Non-Small Cell Lung Cancer Patients: A Target Trial Emulation
Zihan Gao1, Troy Coaston1, Sara Sakowitz2, Giselle Porter1, Jeffrey Balian1, Jane Yanagawa3, Peyman Benharash4
1Cardiovascular Research Outcomes Laboratories (CORELAB), David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA, 2Department of Surgery, Massachusetts General Hospital, Boston, MA, 3Division of Thoracic Surgery, Department of Surgery, University of California Los Angeles David Geffen School of Medicine, Los Angeles, CA, 4Department of Surgery, University of California Los Angeles David Geffen School of Medicine, Los Angeles, CA
Background:
The selection of video- (VATS) versus robotic-assisted thoracoscopic surgery (RATS) for lobectomy among non-small cell lung cancer (NSCLC) largely depends on surgeon preference. Prior studies showed RATS to be associated with comparable outcomes and lower rates of conversion-to-open. However, direct comparisons between the approaches remain limited. This study aims to characterize trends in RATS versus VATS and associated mortality in stage I-III NSCLC. We further sought to use target trial emulation (TTE) to compare survival between the surgical techniques.
Methods:
All adults (≥18 years) with stage I-III NSCLC undergoing lobectomy via RATS or VATS were identified in the 2010-2022 National Cancer Database and stratified by the operative approach (Cohorts: RATS versus VATS). Outcomes included 3- and 5-year overall survival (OS), conversion to open, and timing from diagnosis to procedure. Multivariable regression, Kaplan-Meier time-to-event, and Cox proportional hazard models were used to evaluate the association of operative approach with outcomes. A TTE using advanced methodology, including clone-censor-weighting, was used to estimate the causal effect of surgical techniques on OS, addressing selection and immortal time biases inherent to retrospective studies. Adjusted odds ratios (AOR), β coefficients, and adjusted hazard ratios (aHR) with 95% confidence interval (CI) were reported. Subgroup analyses were performed within each cancer stage.
Results:
Of 113,726 included patients who underwent minimally invasive lobectomy, 43.2% underwent RATS. The use of RATS (vs VATS) increased from 14.0 to 65.4% from 2010 to 2022 (nptrend <0.001). Stages I, II, and III comprised 69.1, 19.8, and 11.1% of the cohort, respectively. Compared to VATS, RATS had a longer time between diagnosis and procedure (38 vs 31 days, p<0.001) and more frequent care at integrated network cancer programs (23.7 vs 18.1%, p<0.001; Table). Notably, RATS had higher survival at 3 (82.7 vs 81.1%, log-rank p<0.001) and 5 years (71.4 vs 70.5%, log-rank p<0.001; Figure) as well as lower rates of conversion to open (4.8 vs 11.2%, p<0.001) compared to VATS.
Following risk-adjustment, RATS was associated with reduced risk of mortality (AOR 0.95, 95% CI 0.92 - 0.98), conversion to open (AOR 0.48, 95% CI 0.46 - 0.50), and longer days from diagnosis to procedure (β 3.30, 95% CI 2.69 - 3.90). Interesting, delay to procedure was associated with higher mortality among RATS (AOR 1.15, 95% CI 1.08 - 1.21) and VATS (AOR 1.21, 95% CI 1.16 - 1.27). The TTE analysis revealed RATS to have a lower hazard of mortality at 3 years (aHR 0.93, 95% CI 0.90 - 0.96) but similar HR at 5 years (aHR 1.00, 95% CI 0.98 - 1.02).
Subgroup analyses showed RATS to be associated with reduced hazard of mortality at 3 years (aHR 0.93, 95% CI 0.89 - 0.98) but similar at 5 years (aHR 0.96, 95% CI 0.93 - 1.00) in stage I NSCLC. However, within NSCLC stage II and III, RATS was associated with similar hazard of mortality within 3 (II: aHR 0.98, 95% CI 0.93 - 1.05; III: aHR 0.96, 95% CI 0.90 - 1.03) and 5 years (II: aHR 1.00, 95% CI 0.96 - 1.06; III: aHR 0.96, 95% CI 0.91 - 1.02). Finally, all three subgroups continued to be associated with reduced conversion to open (I: AOR 0.47, 95% CI 0.44 - 0.50; II: AOR 0.52, 95% CI 0.47 - 0.58; III: AOR 0.46, 95% CI 0.40 - 0.52) and longer delays (I: β 3.00 days, 95% CI 2.29 - 3.72; II: β 4.62, 95% CI 3.30 - 5.95; III: β 3.34, 95% CI 1.29 - 5.38).
Conclusion:
The use of RATS has significantly increased and was associated with lower conversion to open and longer diagnosis-to-surgery intervals. Delays to surgery were linked with higher mortality within both approaches. Within subgroups, RATS had a lower hazard of 3-year mortality in stage I but similar aHR in stage II and III NSCLC. Future efforts should focus on identifying patients more likely to benefit from RATS while optimizing timing, access, and costs to reduce treatment delays and financial burden.
Table. Patient sociodemographic, clinical, and hospital characteristics, stratified by video- vs robotic-assisted thoracoscopic lobectomy. CDI Charlson-Deyo Comorbidity Index; IQR interquartile range; RATS robotic-assisted thoracoscopic surgery; VATS video-assisted thoracoscopic surgery.
Figure. Unadjusted overall survival at 3 and 5 years by operative approach in stage I-III non-small cell lung cancer patients (A) in the original cohort and (B) following target trial emulation methodology. RATS robotic-assisted thoracoscopic surgery; VATS video-assisted thoracoscopic surgery.
Back to 2026 Abstracts
