Robot-Assisted Thoracic Surgery for Lung Cancer is Associated with Reduced Conversion to Open Following Neoadjuvant Therapy

S. Sakowitz¹, S. Bakhtiyar², S. Mallick³, N. Chervu¹, H. Lee⁴, P. Benharash⁵, J. Yanagawa^6
1UCLA David Geffen School of Medicine, Los Angeles, California ²University of California Los Angeles UCLA, Los Angeles, California ³UCLA, Los Angeles, California ⁴David Geffen School of Medicine, Los Angeles, California ⁵UCLA Division of Cardiac Surgery, Los Angeles, California ⁶UCLA, David Geffen School of Medicine, Los Angeles, California

Purpose: Increasing utilization of neoadjuvant chemo-immunotherapy in the treatment of non-small cell lung cancer (NSCLC) has raised concerns regarding added surgical complexity from post-therapy hilar fibrosis. Using a national cohort, we compare the impact of neoadjuvant therapy on conversion to open surgery rates following robotic versus video-assisted thoracic surgical approaches.

Methods: All adults undergoing either video-assisted thoracoscopic (VATS) or robot-assisted thoracic surgery (RATS) for NSCLC were identified within the 2010-2020 National Cancer Database. Patients with AJCC 8th Edition stage IIA-IIIA disease were included for analysis. Those receiving neoadjuvant chemo- or immunotherapy were considered the Neoadjuvant cohort (others: N-Neoadjuvant). Patients undergoing palliative resection were excluded.

Entropy balancing was applied to adjust for innate variation in patient, disease, and hospital factors and generate balanced patient cohorts. Multivariable models were constructed to evaluate the independent association between surgical approach and likelihood of CTO. Covariates were admitted based on automated selection and ultimately included receipt of neoadjuvant therapy, stage, demographics, Charlson-Deyo comorbidity index (CDI), hospital type and volume, and year of diagnosis. The predicted risk of CTO was calculated for the Neoadjuvant and N-Neoadjuvant cohorts across each stage and approach.

Results: Of 229,310 stage IIA-IIIA NSCLC patients, 28,825(13%) comprised Neoadjuvant and 200,485(87%) N-Neoadjuvant. Considering N-Neoadjuvant, 48,276(24%) underwent RATS and 152,209(76%) VATS. Among the Neoadjuvant cohort, 12,552(44%) had RATS and 16,273(56%) VATS. Further, among Neoadjuvant, RATS utilization increased with stage (stage IIA 30% RATS vs 70% VATS; IIB 42% vs 58%; IIIA 45% vs 55%, P< 0.001). Across stage, CTO rates were lower for RATS than VATS for Neoadjuvant (stage IIA 10% RATS vs 18% VATS; IIB 6% vs 14%; IIIA 6% vs 15%, all P< 0.001) as well as for N-Neoadjuvant (IIA 8% RATS vs 14% VATS; IIB 7% vs 13%; IIIA 8% vs 15%, all P< 0.001).

Following entropy balancing and risk adjustment, RATS was linked with significantly reduced odds of CTO(AOR 0.52, CI 0.48-0.56), while Neoadjuvant remained independently associated with greater likelihood of CTO(AOR 1.13, CI 1.04-1.23). When evaluating patients undergoing VATS, receipt of neoadjuvant treatment was linked with greater odds of CTO(AOR 1.15, CI 1.08-1.23), but no significant clinical difference in risk (stage IIA 14.5% Neoadjuvant vs 14.6% N-Neoadjuvant; IIB 12.5% vs 12.9%; IIIA 14.4 vs 15.2%). However, among those undergoing RATS, receipt of neoadjuvant therapy had no independent impact on likelihood of conversion(AOR 1.06, CI 0.95-1.19).

Conclusion: MIS approaches are increasingly being used following neoadjuvant therapy with acceptable outcomes. We report greater RATS utilization following neoadjuvant treatment. Across stage, RATS was linked with lower likelihood of CTO, relative to VATS. Further, receipt of neoadjuvant therapy did not have an independent impact on likelihood of conversion following RATS.

Identify the source of the funding for this research project: None