COMPARATIVE SURVIVAL OUTCOMES OF SURGICAL VERSUS NON‑SURGICAL TREATMENT STRATEGIES IN EARLY‑STAGE CANCER

Abstract

Which of the two modalities of therapy, namely surgical resection and stereotactic body radiotherapy, is the most effective modality of treatment in the treatment of patients with non-small cell lung cancer at an early stage is a controversial issue, with the two options of surgical resection and stereotactic body radiotherapy being the standard ones but lacking definite comparative evidence that is presented by sufficiently powered randomized controlled trials. This study used 824 patients (412 on each arm) with T1-T2a N0 M0 non-small cell lung cancer using the help of the National Cancer Database, Surveillance, Epidemiology, and End Results registry and combined individual patient data of the ROSEL and STARS trials to compare overall survival, recurrence-free survival, and competing risk of recurrence between surgical resection and stereotactic body radiotherapy. Propensity scores were successfully matched with the balance of covariates at baseline (age, Charlson comorbidity index, tumor size, performance status, and histology). With a 5-year overall survival of 68.4% in the case of surgery and 64.7% in the case of stereotactic body radiotherapy at the median follow-up of 48.3 months. Recurrence-free survival at 5 years of time was 71.2% compared to 66.9%. Nevertheless, it was found that regional recurrence was much more frequent following stereotactic body radiotherapy, whereas there was similar local control and distant metastasis. Subgroup analysis showed that surgery had a survival advantage in patients with a tumor size greater than 2 cm but not greater than 4 cm, but showed no greater survival advantage compared to no risk stratification in patients with tumor size less than 2 cm or less than 4 cm. Machine learning models, including random survival forest and DeepSurv, demonstrated a greater predictive performance compared to traditional Cox regression (AUC 0.698), with decision curve analysis showing a positive net benefit in the model-guided risk stratification across threshold probabilities of 0.15-0.45. No meaningful differences were found in the mortality rates related to treatments at 90 days. These findings indicate that there is no overall survival difference between surgery and stereotactic body radiotherapy in propensity-matched early-stage non-small cell lung cancer cohorts, but surgery has a regional recurrence advantage and a survival advantage in larger tumors. The selection of the patients must then be based on the size of the tumor, the burden of comorbidity, and the operability but not on the belief that either one or the other modality is universally superior.