Accounting for Operability Status in Retrospective Comparisons of Surgery vs. SBRT for Early-Stage NSCLC

Abstract

Listen

Translate article

<h3>Purpose/Objective(s)</h3> Patients receiving stereotactic body radiotherapy (SBRT) for early-stage non-small cell lung cancer (NSCLC) are typically inoperable, in concordance with consensus guidelines that advocate surgical resection as standard of care for operable patients. This differential treatment allocation complicates retrospective comparisons of surgery with SBRT by introducing the potential for confounding by operability. Accounting for operability status in these comparisons offers one solution to mitigate this source of bias; however, it is unknown what proportion of published analyses do so. We conducted a meta-research study to characterize whether published retrospective series comparing surgery with SBRT for ES- NSCLC account for operability status. <h3>Materials/Methods</h3> We queried PubMed for manuscripts reporting primary data from retrospective comparisons of overall survival (OS) between patients undergoing surgery versus SBRT for early-stage NSCLC. There were two outcomes of interest: (1) whether the manuscript reported whether allocation to SBRT was based on a determination of patient operability; and (2) whether each manuscript reported a direct OS comparison between operable SBRT patients and surgically-treated patients. χ² analysis was used to measure associations between outcomes and categorical variables of interest, including temporal (year of publication), journal-specific (specialty focus, impact factor), cohort-specific (nationality, source), and author-specific (specialty) factors, with a pre-specified <i>p</i><0.10 level of statistical significance. <h3>Results</h3> Our search yielded 3,072 initial results, of which 61 analyses met our screening criteria. Twenty-one of 61 analyses (34.4%) reported operability status influencing treatment allocation. These analyses were significantly more likely to be published in journals with a surgical focus (52.4 vs 20%) and impact factor <5 (81.0 vs 57.5%), to contain patient cohorts from Asia (47.6 vs 27.5%) and from institutional datasets (81.0 vs 55.0%), and to have a radiation oncologist as first (42.9 vs 25.0%) or senior (42.9 vs 27.5%) author. Seven (13.7%) of 61 analyses directly compared OS between operable SBRT patients and patients undergoing surgery. These analyses were significantly more likely to be published in surgery-focused journals (71.4 vs 25.9%) and to contain cohorts from the National Cancer Database (57.1 vs 11.1%). <h3>Conclusion</h3> Nearly two-thirds of peer-reviewed retrospective studies that have compared OS between surgery and SBRT for early-stage NSCLC lack information on patient operability status, and nearly 90% lack a direct comparison between operable SBRT patients and those receiving surgery. Randomized comparisons of surgery versus SBRT in operable cohorts are ongoing, and the results are awaited.