br Material and methods br Results br Discussion Regarding t
Material and methods
Discussion Regarding the definition of STAS, the 2015 WHO classification introduced tumor STAS officially as a new tumor invasion characteristic . STAS was defined by Kadota et al. as tumor E-64-c within air spaces in the lung parenchyma beyond the edge of the main tumor and was composed of three morphological patterns: single cells, micropapillary structures, and solid nests or tumor islands . In the report, the presence of STAS was a significant risk factor for recurrence in patients with small stage I adenocarcinomas treated with limited resection but not in those who underwent lobectomy . Kadota et al. speculated that undetected STAS in the alveolar space beyond the surgical margin may cause recurrence . Before that, an early report indicated that the pathologic features of aerogenous spread with floating cancer cells (ASFC) was a significant prognostic factor in patients with metastatic lung cancer of colorectal origin [20,21]. Subsequently, tumor islands in the alveolar spaces were discovered by three-dimensional reconstruction analysis that was connected to the main tumor . In a further study, Onozato et al. reported that the prognosis of stage I-II lung adenocarcinomas with tumor islands was significantly worse than that of those without . Therefore, tumor islands could be one type of morphological pattern of STAS, although it is controversial whether tumor islands should be included in STAS or not . Free tumor clusters (FTC) have been reported as distinctive morphological features in patients with lung adenocarcinoma harboring a micropapillary component and influenced the decision regarding the pathologic T stage . FTC could be one type of STAS. Warth et al. subsequently defined STAS as a small solid cell nest including at least five tumor cells. In the cohort, STAS was reported as a significant risk factor for reducing OS and disease-free survival in patients with completely resected stage I–IV lung adenocarcinoma, and there was no significant difference in patients with extensive STAS and those with limited STAS . In our study, we used a comprehensive definition of STAS. When we evaluated the presence of STAS, which was observed in 49.5% of patients in the study cohort with stage I-III lung adenocarcinomas ≤4 cm, 54.9% of patients in the validation cohort were included. Kadota et al. reported that STAS was observed in 38% of patients with lung adenocarcinomas ≤2 cm . By contrast, Warth et al. reported that 50.6% of patients with stage I–IV lung adenocarcinoma showed STAS . Consequent research showed similar findings of the presence of STAS in patients with other NSCLCs: 47.6% of patients with stage I lung adenocarcinoma , 50.6% of patients with stage I-III lung adenocarcinoma , 58.4% of patients with stage I–IV lung adenocarcinoma , and 73% of patients with lung adenocarcinoma with lymph node metastasis . In particular, we found that there were significant correlations between STAS and several pathologic features including tumor size, pathologic stage, HS, VPI and VI. Similarly, STAS was reported to be associated with tumor size [9,29], pathologic stage [6,7,26,27,30], lymphatic or vascular invasion, [2,7,9,26,27,29,31] pleural invasion [7,9,, , ] and more invasive histological subtypes, such as micropapillary and solid patterns [2,6,8,9,25,26,31]. Therefore, it was inferred that STAS could be affected by the pathologic features of the main tumor itself and by invasive tumor behavior. Thus, the positive rate for STAS can be diverse in any cohort because of the pathologic features of the main tumor itself and its invasive pattern, even lymph node metastasis. Additionally, we focused on VPI, VI and HS. Among them, VI included VI in the lymphatics or vasculature. Univariate and multivariate Cox proportional hazards regression analysis in the study cohort showed that STAS, VPI, VI and HS were significant unfavorable risk factors for RFS and OS in patients with resected lung invasive adenocarcinomas ≤4 cm, which was consistent with studies in patients with lung adenocarcinoma [7,27,, , ]. Recently, STAS was designated as a novel invasive pattern in lung adenocarcinoma. Additionally, STAS was associated with poorer OS and RFS in squamous cell carcinomas [, , ] and lung pleomorphic carcinoma  but not small cell lung cancer . Although it should be emphasized that STAS-presence tumors occurred concomitantly with tumors exhibiting VPI and VI more frequently than STAS-absence tumors, multivariate analysis showed that STAS, VPI, VI and HS were significantly independent prognostic factors of RFS and OS. Therefore, it can be inferred that STAS, VPI and VI may be different pathologic characteristics of invasive tumor behavior. Additionally, there was no sufficient evidence to clearly prove any causality between them.