A critical analysis of segmentectomy versus lobectomy for non-small-cell lung cancer

Abstract

Bao et al. [1] should be commended for completing a thorough and timely meta-analysis to compare the survival outcomes of segmentectomy versus lobectomy for Stage I non-small-cell lung cancer (NSCLC). This analysis represents the first of its kind to distinguish segmentectomies from wedge resections, an important distinction from both an oncological and technical perspective that has been overlooked in previous meta-analyses that compared ‘sublobar resections’ with lobectomy procedures [2, 3]. We concur with the authors that the current literature lacks robust high-level evidence on this topic and a meta-analysis of observational studies may provide insightful guidance for clinicians. Results of this study found that patients who underwent segmentectomy for Stage I, IA and IA (2–3 cm) NSCLC were associated with inferior combined overall survival (OS) and cancer-specifi cs urvival (CSS) outcomes compared with those who underwent lobectomy. However, patients with Stage IA (≤2 cm) NSCLC had no statistically significant difference in survival compared with lobectomy. It is imperative to analyse these results in detail to avoid misleading conclusions. In two of the four analysed subgroups, Stage IA and Stage IA (2–3 cm), OS outcomes were significantly worse after segmentectomy, but CSS was not significant. The combined OS/CSS were significant in both subgroups, and the authors concluded that patients who underwent segmentectomy in these cohorts resulted in inferior outcomes. However, it must be emphasized that patient baseline characteristics between these two treatment groups differed significantly, and patients were often selected for segmentectomies due to significant comorbidities and limited pulmonary reserve prohibiting them from lobectomy procedures [4, 5]. In such cases, differences in OS may be misleading compared with CSS, as patients died due to causes unrelated to NSCLC and the oncological efficacy of their surgical procedures. Hence, worse OS outcomes in the segmentectomy group may be a reflection of their patient characteristics rather than their operative outcome. Similarly, a combined statistic of OS and CSS may overestimate the adverse survival outcomes of segmentectomy procedures. To address the issue of differing patient baseline characteristics, Tsutani et al. [6] recently published a propensity-score analysis involving 481 patients with clinical Stage IA adenocarcinoma who were subsequently matched into 81 pairs according to clinicopathological factors. Results of this study demonstrated no differences in regard to OS and recurrence-free survival at 3 years. In our opinion, it is no longer controversial to perform sublobar resections for patients with NSCLC who are deemed ineligible for lobectomy procedures due to comorbidities or limited pulmonary function. The area of interest lies with the cohort of patients who can tolerate either a segmentectomy or a lobectomy procedure. For this growing patient population who are diagnosed with early stage NSCLC through more aggressive screening programmes, oncological efficacy needs to be balanced with pulmonary preservation and potentially reduced perioperative complications associated with segmentectomy. Future analyses should aim to differentiate data from studies that included ‘compromised’ patients who underwent sublobar resections as a result of their inability to undergo lobectomy, or ‘intentional’ patients who could have tolerated either procedure.