Surgical treatment of lung cancer

Abstract

Background: Surgery is a central part in the treatment of lung cancer. There have been recent advances in minimal invasive and lung sparing surgical techniques, but evidence regarding safety and implementability is not well established. There are also differences in outcomes after surgery for lung cancer related to income and patient sex that are not fully explored. Moreover, there are regional differences in rates of surgery for lung cancer in Sweden that are not entirely understood. The studies included in this thesis aimed to explore these aspects.

Methods and results: Study I Patients who underwent uniportal and multiportal video-assisted thoracic surgery (uVATS and mVATS) lobectomy at Karolinska University Hospital from 2016 to 2018 were identified in the Swedish National Quality Register for General Thoracic Surgery (ThoR) (122 uVATS and 211 mVATS lobectomies). Baseline differences were handled using Inverse Probability of Treatment Weighting (IPTW). The proportion of uVATS lobectomies increased during the study period and conversions to thoracotomy decreased. Perioperative mortality was low and 94% in both groups had no complication. Patients who underwent uVATS lobectomy were more often discharged directly to home compared to mVATS (76.2% vs. 62.1%, p=0.008).

Study II All patients in Sweden who underwent surgery for early-stage non-small cell lung cancer (NSCLC) from 2008 to 2017 were identified in ThoR (5500 patients), and record linkages were made to other national health data registers. The age- and sex-adjusted incidence rate for death per 100 person-years was 9.4 and 15 in the highest and lowest income groups, respectively. Compared to high income patients, poor patients were less likely to have preoperative positron emission tomography or minimally invasive surgery. They were more often older and had more comorbidities. After multivariable adjustment, the hazard ratio (HR) for death was 0.77 (0.62-0.96, 95% confidence interval, CI) in the highest compared to the lowest income group.

Study III All patients in Sweden who underwent surgery for lung cancer from 2008 to 2017 were identified in ThoR (6536 patients), and record linkages were made to other national health data registers. Baseline differences were handled using IPTW. Women were found to have a clear and consistent survival benefit compared to men, also in subgroups according to age, histopathology, and stage. The HR for death was 0.73 (0.67-0.79, 95% CI) for women compared to men and the absolute survival difference (95% CI) at 1, 5, and 10 years was 3.0% (2.2%-3.8%), 10% (7.0%-12%), and 12% (8.5%-15%), respectively.

Study IV Patients who underwent uVATS segmentectomy (232 patients) and lobectomy (601 patients) at Karolinska University Hospital from 2017 to 2022 were included. Baseline differences were handled using IPTW and an as treated analysis was used. Implementation of uVATS was gradual over the study period. In the segmentectomy and lobectomy groups respectively, 97% and 94% of patients had no postoperative complications, the median number of lymph node stations sampled was 4 vs. 5, and non-radical microscopic resection occurred in 1.7% vs. 1.8%. Drains were removed on postoperative day 1 in 75% vs. 72%, and 90% vs. 89% were discharged directly home after segmentectomy and lobectomy, respectively.

Study V All patients in Sweden who underwent surgery for NSCLC from 2013 to 2017 were identified in ThoR (3050 patients), and record linkages were made to other national health data registers. Data on regional lung cancer incidence was gathered from the Swedish National Lung Cancer Register. 14 451 patients were diagnosed with NSCLC in Sweden during the study period. The average yearly rate of surgery was 6.1 per 100 000 inhabitants and varied between regions from 4.4 to 7.9. The proportion of patients with lung cancer who received surgery varied between regions from 12% to 28% with most regions being close to the national average of 21%. In patients with stage I-II disease, the proportion of new cases operated varied between regions from 57% to 116%. No differences in demography, stage at diagnosis, performance status, or presentation at multidisciplinary team stood out as explanatory factors.

Conclusions: uVATS lobectomy was safe to implement and had similar postoperative outcomes compared with mVATS lobectomy at our institution. Despite a tax-funded, universal health care system, there was an inverse correlation between household disposable income and mortality after surgery for lung cancer. Women have a consistently better prognosis after surgery for lung cancer compared to men. uVATS segmentectomy was safe to implement and had similar postoperative outcomes compared with uVATS lobectomy at our center. There were regional differences in the rates of surgery for early-stage NSCLC in Sweden that could not be explained by regional differences in lung cancer incidence.