Spread through air spaces is a poor prognostic indicator in patients undergoing lobectomy for lung cancer: A systematic review and meta-analysis


 Background: Previous studies have confirmed the poor prognostic value of spread through air spaces (STAS) in patients undergoing limited resection for lung cancer. Nevertheless, its prognostic value remains controversial in patients undergoing lobectomy. Consequently, we aim to systematically and comprehensively evaluate the prognosis of patients with STAS undergoing lobectomy for lung cancer. Methods: An extensive search of literature databases was conducted. Recurrence-free survival (RFS) and overall survival (OS) were compared between patients with or without STAS undergoing lobectomy. In addition, results of the limited resection were also evaluated and presented. Results: 5 studies with 1531 patients reported the outcomes of lobectomy and 4 studies including 505 patients evaluated the survival of limited resection. In patients undergoing lobectomy, STAS was associated with significantly worse survival than non-STAS, including both RFS (HR=1.700; 95% CI: 1.265–2.283; P＜0.001; P for heterogeneity=0.637; I2=0.0%) and OS (HR=2.620; 95% CI: 1.138–6.031; P=0.024; P for heterogeneity=0.128; I2=51.4%). STAS was also correlated with shorter RFS (HR=3.434, 95%CI 2.173 to 5.428; P＜0.001; P for heterogeneity=0.828, I2=0.0%) and OS (HR=3.494, 95%CI 2.128 to 5.736; P＜0.001; P for heterogeneity=0.501, I2=0.0%) in limited resection. Conclusions: STAS is a poor prognostic indicator in patients undergoing lobectomy for lung cancer.


Background
In the past, lung cancer invasion was established as (1) a histological subtype other than a lepidic pattern, (2) myofibroblastic stroma associated with invasive tumor cells, and (3) vascular or pleural invasion. In 2015, the concept of spread through air spaces (STAS) was proposed by World Health Organization (WHO) as a new pattern of invasion in adenocarcinoma on the basis of 2 validated large independent studies from the US 1 and Germany 2 . It was defined as "micropapillary clusters, solid nests, or single cells spreading within air spaces beyond the edge of the main tumor" and regarded as the fourth category of lung cancer invasion. Though the concept of STAS was initially proposed for adenocarcinoma, current studies have confirmed that STAS was also observed in squamous cell carcinoma 3 − 5 , pleomorphic carcinoma 6 and small-cell lung cancer 7  Since then, a variety of studies have been updated and the prognostic value of STAS in limited resection for lung cancer has been confirmed 8 − 10 . According to a previous metaanalysis 11 calculating the hazard ratios (HRs) from multivariate analyses, the presence of STAS predicted both reduced RFS and OS in patients with non-small cell lung cancer (NSCLC), further validating its prognostic value in clinical practice. However, it is worth noting that most included studies in this meta-analysis consisted of multiple types of surgery and its subgroup analyses showed STAS in lobectomy was not related to reduced RFS while STAS in limited resection predicted significantly shorter RFS. Consequently, the overall conclusion of this meta-analysis could be influenced by the effect of the limited resection. In addition, in accordance to a recent study conducted by Yokoyama S 6 , STAS in patients undergoing lobectomy for lung cancer predicted both shorter RFS and OS. Thus, in patients with lung cancer treated with lobectomy, it remains controversial whether STAS could result in worse prognosis and the prognostic value of STAS in lobectomy 4 should be further evaluated with more current data.
Overall, the prognostic value of STAS in lobectomy remains uncertain. With more current data updated on lobectomy, it is the time to update a systematic review for a better recognition for its prognostic role in lobectomy for lung cancer.

Methods
Academic retrieval strategies: An extensive literature retrieval of network databases including PubMed, Embase, Web of Science, Cochrane Library and Google Scholar was conducted to affirm all possibly relevant studies before April 2019. We amalgamated "lung cancer", "spread through air spaces", "STAS", "lobectomy", "limited resection", "sublobar resection", "segmentectomy", "wedge resection", "prognosis", "recurrence-free survival" and "overall survival" as well as their Medical Subject Headings (MeSH) terms. An additional manual search was carried out, containing the reference lists originating from retrieved review articles, primary studies, and abstracts from meetings.
All retrieved results were evaluated in accordance to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The original studies were screened gradually with the investigation on titles, abstracts, and full texts. The eligibility criteria were described as follows: (i) status of lung cancer and STAS should be confirmed year of publication, country, median age of patients, number of patients, incidence of STAS, histological type, pathological stage, surgical type, quality score. RFS or OS was obtained or estimated from each study to assess the prognostic value of STAS in lobectomy and limited resection.
Quality assessment of observational studies was carried out with the use of the Newcastle-Ottawa Scale, including 3 metrics: patient selection, comparability of groups, and the ascertainment of either the exposure or outcome of interest for case-control or cohort studies, respectively. The high-quality observational study was considered as a study with a quality score of ≥ 7 stars with the total score of 0-9 (allocated as stars).

Statistical analysis:
HR and 95%CI was obtained directly or estimated from the original article and they were used to evaluate for RFS and OS. HR > 1 demonstrated a significantly worse prognosis for RFS and OS. Heterogeneity across studies was examined by the Cochran Q chi-square test and the I² statistic and statistical heterogeneity among studies was defined as I 2 statistic greater than 50%. A random-effects model was preferred if high heterogeneity of the 6 studies (P < 0.1 or I 2 > 50%) was observed, and vice versa when there was no statistically significant heterogeneity. Publication bias was assessed using Funnel plots test, Begg's rank correlation test and Egger's linear regression method. Sensitivity analysis was performed by sequential removal of each study. The statistical analysis was performed in Stata software (version 12, StataCorp, TX). All the P values were two tailed, and statistical significance was set as P < 0.05.

Results
The Characteristics of the Studies and Quality Assessment: A total of 176 studies were initially scanned from the previously mentioned 5 network databases before April, 2019. By inspecting the reference lists following retrieved articles and relevant reviewed articles, 10 studies were identified additionally. Of these, 144 studies were excluded through the removal of duplicated ones and identification in accordance to their titles and abstracts. After further confirmation in terms of full texts of the remaining 42 studies, 6 studies with a total of 2036 patients were finally included.
Among them, 5 studies with 1531 patients reported the prognosis with regards to lobectomy and 4 studies including 505 patients evaluated the survival of limited resection.
The retrieval process is elucidated in a PRISMA flow chart in Fig. 1.
All included studies were observational studies. As for the studies reporting the prognosis in lobectomy, they were performed in Japan (n = 3) and China (n = 2). Others reporting the results of limited resection were performed in Japan (n = 3) and China (n = 1). All studies were conducted after 2015. In addition, as far as the quality assessment is concerned, since all included studies were observation studies, the Newcastle-Ottawa Scale was applied and it manifested all included studies achieved high quality with a quality score of ≥ 7 stars. The detailed characteristics and the quality assessment of each study were 7 summarized in Table 1.

Conclusion
The presence of STAS is a poor prognostic indicator in patients undergoing lobectomy for lung cancer. Future clinical practice should attach great importance to its prognostic value, independent of the surgical procedure.

Declarations
Ethics approval and consent to participate: 13 Not applicable

Consent for publication:
Not applicable

Availability of data and materials:
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request

Competing interests:
The authors declare that they have no competing interests