INTRODUCTION

The pathological evaluation of lung cancer has undergone significant refinement in recent years. The 8th edition of the TNM classification of lung tumors has introduced the invasive size of adenocarcinoma—rather than total tumor size—as the primary determinant for staging [1]. This shift enhances prognostic stratification while demanding a more nuanced approach to tumor measurement in morphologically challenging cases. Concurrently, the adoption of novel histological grading criteria, based on the predominant histologic patterns proposed by the International Association for the Study of Lung Cancer (IASLC), demonstrates superior prognostic accuracy and clinical utility compared to legacy grading systems [2]. Alongside these changes, spread through air spaces (STAS) was formally recognized as a specific invasive pattern in lung adenocarcinoma [2,3]. Morphologically, STAS is defined by the dissemination of neoplastic cells into the pulmonary alveolar spaces extending beyond the primary tumor margin (Fig. 1). This phenomenon manifests in three distinct architectural patterns: (1) micropapillary clusters lacking central fibrovascular cores; (2) solid nests or tumor islands filling the air spaces; and (3) scattered, discohesive single cells (Fig. 2) [2,3]. The diagnostic threshold for STAS is met when these cellular components are identified beyond the main tumor boundary, regardless of whether they remain localized to the immediately adjacent alveolar layer. Given its profoundly deleterious impact on patient prognosis, the presence of STAS serves as an absolute exclusion criterion for the diagnoses of adenocarcinoma in situ and minimally invasive adenocarcinoma [2,3]. Bolstered by consistent evidence from multiple meta-analyses identifying STAS as a negative prognostic factor in early-stage disease [4-7], the IASLC Staging Project has officially endorsed its integration as a histologic T descriptor in the 9th edition of the TNM classification of lung cancer [8]. This formal stance supports the clinical relevance of STAS despite ongoing concerns regarding ex vivo artifacts. Accordingly, this review aims to comprehensively synthesize the prevailing controversies and recent academic milestones regarding STAS, the current limitations in preoperative STAS risk prediction, the logistical hurdles of intraoperative diagnosis, choice of surgical extent, potential role for adjuvant systemic therapy in patients with early-stage epidermal growth factor receptor (EGFR) mutant/STAS+ disease and provide future directions for investigating STAS as a potential dynamic mechanism of tumor dissemination in lung cancer. A comprehensive literature search was conducted using PubMed and MEDLINE databases for articles published up to May 2026. Search terms included "spread through air spaces," "STAS," "lung adenocarcinoma," "TNM classification," and "tumor evolution." Peer-reviewed original research, meta-analyses, and international consensus guidelines were prioritized for inclusion to ensure a rigorous and globally representative synthesis.

THE EX VIVO ARTIFACT DEBATE VS. BIOLOGICAL REALITY

Since its formal recognition, the conceptual validity of STAS has been debated, with several investigators postulating that it may merely represent an ex vivo artifact [9-13]. Thunnissen et al. [10] delineated four putative mechanisms for such mechanical distortions in pulmonary tissue: surgical collapse, ex vivo contraction of bronchial and bronchiolar smooth muscle, clamping-induced edema, and the mechanical drag of neoplastic fragments across a scalpel blade [11]. Surgical collapse, precipitated by the exsufflation of air and exsanguination of resected lung specimens, can compress alveolar septa, thereby generating pseudopapillary architectures that mimic true biological invasion. Furthermore, the "Spread through a knife surface" hypothesis posits that shear forces exerted during macroscopic tissue sectioning can mechanically dislodge loosely adherent tumor fragments, depositing them into adjacent, otherwise uninvolved alveolar spaces. Additionally, the presence of several histologic mimics of tumor cells poses a significant challenge to interobserver reliability in the definitive diagnosis of true STAS (Fig. 3). Critics have historically contended that the topological dependency of STAS—wherein morphologically identical cellular aggregates are classified as either true invasion or artifact based solely on their spatial proximity to the primary tumor mass—introduces an inherent degree of subjectivity and diagnostic inconsistency. However, advanced 3D pathological reconstructions of paraffin-embedded tissue have compellingly challenged the artifact hypothesis. Onozato et al. demonstrated that "tumor islands"—defined as discrete, voluminous collections of neoplastic cells within alveolar spaces that lack a well-demarcated micropapillary configuration—are not freely floating, mechanically dislodged anomalies [14]. Rather, these structures form complex, continuous neoplastic networks that connect with one another and with the primary tumor across diverse anatomical planes. Crucially, these contiguous networks frequently parasitize pre-existing alveolar capillary beds, an active biophysical interaction that indicates a distinct biological survival and adaptive mechanism rather than a passive ex vivo occurrence [14,15]. Moreover, the identification of viable STAS cells in independently processed secondary resections provides empirical evidence against the single-knife-cut artifact hypothesis, further suggesting STAS as an authentic, biologically driven mechanism of pulmonary tumor dissemination [16,17]. In addition, Han et al. [18], in a prospectively collected cohort of 1,869 surgically resected lung cancer specimens, demonstrated no significant difference in STAS occurrence or STAS distribution according to surgery type, i.e., endoscopic surgery vs thoracotomy specimen. This is additional evidence against STAS representing an artifact from endoscopic surgical processing by surgeons. Together, these studies support STAS being an in vivo invasive mechanism similar to lymphovascular or pleural invasion [14-18].

PREOPERATIVE BIOPSY AND RISK OF IATROGENIC STAS GENERATION

Although percutaneous needle biopsy (PCNB) and bronchoscopic biopsy serve as fundamental preoperative diagnostic modalities in thoracic oncology, persistent concerns exist regarding the potential for iatrogenic tumor dissemination and needle-tract seeding. To address whether these procedures elevate the risk of STAS or exacerbate its associated adverse outcomes, Lee et al. [19] analyzed a cohort of 2,169 patients with resected stage I non–small cell lung cancer (NSCLC). While univariate analysis hinted at a potential association, multivariate modeling confirmed no statistically significant correlation between preoperative biopsy and the presence of STAS. Importantly, while sublobar resection emerged as a significant risk factor for recurrence and cancer-specific mortality within the STAS-positive subgroup, the performance of a preoperative biopsy did not adversely affect survival metrics. These findings are supported by Kameda et al. [20], who similarly reported no significant correlation between preoperative PCNB and the subsequent histological identification of STAS in a cohort of 808 small lung adenocarcinomas. The collective evidence robustly indicates that preoperative biopsy modalities are clinically safe and do not precipitate STAS-associated dissemination [19,20].

EVOLUTION OF GRADING OF STAS

While the recognition of tumor STAS as a pattern of invasion in lung adenocarcinoma is now widely noted, the methodological standardization for defining and grading its extent remains an area of active pathological research. Contemporary investigations have progressed beyond a simplistic binary assessment of STAS, pivoting toward semiquantitative metrics. Recent literature reveals three primary methodological philosophies in the quest to stratify STAS severity: anatomical landmarks, morphometric enumeration, and standardized spatial distance [18,21-24].

The anatomical landmark approach: Warth et al. (2015) [21]

This early classification categorized STAS into "limited" and "extensive" grades based on the number of alveolar spaces separating the tumor clusters from the main tumor mass [21]. The system defined limited STAS as a detachment of small solid cell nests (at least 5 tumor cells) fewer than three alveolar spaces away from the main tumor mass and extensive STAS as dissemination beyond this 3-alveolar threshold. While the study successfully established that the presence of STAS (both limited and extensive) was associated with reduced overall survival (OS) and disease-free survival, the grading system itself did not reveal a prognostic gradient. Patients with limited STAS and extensive STAS demonstrated virtually identical survival times [21]. Consequently, relying on a three-alveolar-space cutoff proved insufficient for meaningful clinical stratification.

The morphometric enumeration approach: Uruga et al. (2017) [22]

Uruga et al. shifted to a semiquantitative cluster-counting method. Using a 20× objective and a 10× ocular lens, they evaluated the most prominent area of spread [22]. The system stratified tumors into no STAS, low STAS (1–4 single cells or clusters), and high STAS (≥5 single cells or clusters). This approach successfully delineated a prognostic gradient; a higher STAS cluster burden was predictive of shorter recurrence-free survival (RFS) on univariate analysis and multivariate Cox proportional hazards analysis (p = .015). However, as an enumeration-based system, manual cluster counting may be inherently susceptible to inter-observer variability and limit diagnostic throughput in routine pathology practice.

The spatial distance paradigm: Han et al. (2021) [18], Lee et al. (2024) [24], and Lee et al. (2025) [23]

Overcoming the limitations of alveolar landmarks and manual counting, these studies introduced a two-tiered grading system based on absolute distance. STAS I is defined as tumor clusters exclusively within <2,500 µm (exactly one field of a 10× objective lens) from the tumor edge (Fig. 4). STAS II is defined by the presence of clusters ≥2,500 µm away [18]. In stage IA non-mucinous adenocarcinoma, STAS II emerged as an independent poor prognostic factor for RFS and lung cancer-specific survival. This grading system was validated by Lee et al. (2025) [23] in the cohort of 1,660 patients with radically resected stage I non-mucinous adenocarcinoma. According to their study, the 5-year cumulative incidence of recurrence (CIR) was 3.9% in the STAS (–) group, 8.4% in the STAS grade I group, and 17.0% in the STAS grade II group (p < .001). Multivariable competing risk regression analyses revealed that the STAS grading system was associated with an increased risk of any recurrence (STAS grade I: hazard ratio [HR], 1.08; 95% confidence interval [CI], 0.66 to 1.78; p = .76; STAS grade II: HR, 1.94; 95% CI, 1.10 to 3.42; p = .022). In subgroup analysis stratified by pathologic stage, the CIR in the pathologic stage IA STAS grade II subgroup was comparable to that of the pathologic stage IB subgroup (IA, STAS grade II: 13.0%; IB, total: 16.0%; p = .8). The 2,500 µm cutoff (corresponding to a single field of view under a 10× objective lens) may optimize workflow efficiency by providing pathologists with an immediate, standardized visual boundary without the need to count individual cells or alveoli [18,23]. Similarly, Lee et al. [24] adopted a 2-mm cutoff to stratify 208 patients with resected stage I NSCLC into low- and high-STAS groups. Interestingly, while OS rates were similar across groups, patients with more extensive spread (2 mm or more) experienced a higher chance of cancer recurrence. The evolution of these methodologies offers insights into the biological behavior of STAS. The finding that limited STAS (STAS I) may lose prognostic significance for recurrence in multivariate models supports the hypothesis that it may represent localized marginal extension or connected tumor islands in 3D, whereas extensive STAS (STAS II) may reflect a more extensive degree of aerogenous dissemination carrying independent prognostic significance [18,23]. Standardized distance-based grading may offer clinical utility by reducing inter-observer variability and identifying patients who might benefit from careful surveillance.

The utility of a robust STAS grading system lies in guiding surgical extent. While early literature frequently debated whether STAS only impacted sublobar resections, Han et al. [18] and Lee et al. [23] demonstrated that STAS II is an independent poor prognostic factor for recurrence regardless of whether the patient underwent limited resection or radical resection (lobectomy). This indicates that high-grade STAS represents an inherently aggressive tumor biology—potentially driven by mechanisms such as epithelial-to-mesenchymal transition, though this hypothesis requires further validation—rather than merely acting as an artifact or localized surgical margin risk [25-27].

STAS IN OTHER HISTOLOGIC SUBTYPES OF LUNG CANCER

The invasive and destructive biological capacity of tumor STAS is not restricted to specific histological subtypes. Emerging histopathological data confirm that the deleterious prognostic implications of STAS extend far beyond non-mucinous adenocarcinoma, encompassing a diverse array of histological classifications [28-30]. Lu et al. [28] reviewed 445 resected stage I to III lung squamous cell carcinomas, identifying STAS in 132 patients (30%). Utilizing competing risks analyses, they revealed that the cumulative incidences of overall recurrence, distant metastasis, locoregional failure, and cancer-specific mortality were elevated in STAS-positive patients. Multivariable modeling solidified STAS as an independent prognosticator for both overall recurrence and cancer-specific death, establishing it as an important histological variable in squamous cell carcinoma [28]. Similarly, in invasive mucinous adenocarcinoma, the presence of STAS directly correlates with higher rates of local and distant recurrence, as well as significantly shorter RFS [29]. Furthermore, across the spectrum of neuroendocrine tumors—ranging from carcinoids to high-grade neuroendocrine carcinomas—STAS has been identified as an independent predictor of lymph node metastasis and poor OS [30]. Collectively, these observations necessitate a paradigm shift: STAS might be conceptualized not as an isolated histological anomaly, but as a universal, pan-histological mechanism driving aggressive aerogenous dissemination across the diverse pathological landscape of lung cancer.

MOLECULAR CORRELATES AND CONTEXT-DEPENDENT PROGNOSIS

The frequency and implications of STAS vary depending on the tumor size and underlying genomic landscape (Fig. 5) [14,18,20,26,27]. STAS is frequently observed in ROS1-rearranged adenocarcinomas, where altered E-cadherin expression may facilitate aerogenous dissemination [25-27]. Conversely, its prevalence is generally lower in EGFR-mutated tumors; however, its prognostic interpretation in this cohort is highly context-dependent. Lee et al. analyzed a cohort of 1,117 patients with pathologic stage IA disease, demonstrating that while STAS was less frequent in EGFR-mutated variants (particularly L858R), its presence served as an independent predictor of recurrence in this specific subgroup (subdistribution HR, 5.04; p = .004) (Fig. 6). In the EGFR wild-type cohort, STAS did not reach statistical significance as an independent recurrence predictor. It is important to consider that these molecular associations may be partially mediated by underlying histologic phenotypes. For instance, anaplastic lymphoma kinase (ALK)- or ROS proto-oncogene 1 (ROS1)-rearranged tumors frequently exhibit solid or micropapillary architectures—patterns inherently associated with STAS. Conversely, EGFR-mutated adenocarcinomas often present as earlier-stage or better-differentiated tumors, which may partially explain their lower overall frequency of STAS. This indicates that pathological evaluation of STAS may benefit from being integrated with molecular profiling for optimal risk stratification [18,27].

SURGICAL EXTENT: LOBECTOMY VS. SUBLOBAR RESECTION ACCORDING TO STAS

The prognostic impact of STAS is closely associated with the selected surgical approach. To determine the optimal operation for STAS, several retrospective studies have been performed, mainly comparing clinical outcomes in sublobar resections (wedge resection or segmentectomy) to lobar resections (lobectomy) in stage I tumors [18,20,27,31-34]. Currently, investigations into outcomes of STAS+ patients following sublobar and lobar resections are limited to retrospective studies. Most of these retrospective studies showed that in STAS+ tumors, a sublobar resection was associated with higher recurrence rates and worse survival. In cohorts undergoing limited resection, the risk of recurrence is significantly elevated in patients with STAS-positive tumors compared to their STAS-negative counterparts (5-year CIR, 42.6% vs. 10.9%) [31]. Chae et al. [33] analyzed a specific subgroup of 115 patients with stage IA part-solid lung adenocarcinoma who underwent sublobar wedge resection. In their study, multivariable analysis established STAS as a robust, independent risk factor for disease recurrence (HR, 9.447). However, among patients undergoing standard lobectomy, the presence of STAS is generally not associated with an increased risk of overall recurrence, suggesting that lobectomy may mitigate the localized recurrence risk associated with STAS [31,33,34]. Therefore, if STAS is strongly predicted preoperatively or identified intraoperatively, conversion from a planned sublobar resection to a standard lobectomy is often considered to optimize local control. However, Han et al. [18] and Lee et al. [23] demonstrated that while STAS predominantly exacerbates recurrence risk following sublobar resection, extensive dissemination (STAS grade II) functions as an independent predictor of recurrence even among patients receiving standard lobectomies. Assuming the accurate pre- and intraoperative diagnosis of STAS can be reliably achieved, standard lobectomy offers adequate locoregional control for limited STAS. Conversely, the presence of extensive STAS warrants vigilant postoperative surveillance and the potential administration of adjuvant therapy. Synthesizing these findings, the complex interplay between surgical extent and STAS status can be characterized through the following clinical perspectives.

Determination of surgical extent

STAS has been proven to function as a predictor of limited resection failure. While recurrence rates are significantly higher following limited resection in STAS-positive patients, performing a standard lobectomy appears to mitigate this risk. Consequently, the preoperative assessment or intraoperative pathological identification of STAS provides a compelling rationale for converting a planned sublobar resection to a standard lobectomy, thereby optimizing local control and reducing recurrence risk [31,33,34]. However, prior research rarely differentiates between the two distinct types of sublobar resection—segmentectomy and wedge resection—when comparing outcomes to lobectomy. This distinction is critical because wedge resection, unlike segmentectomy or lobectomy, is a non-anatomical procedure [35,36]. The few studies that do separate these approaches indicate divergent outcomes for STAS-positive disease. For instance, a retrospective study by Kagimoto et al. [36] of 609 patients with stage IA lung adenocarcinoma (293 STAS-positive) found no significant difference in 5-year RFS between STAS-positive patients who underwent segmentectomy versus lobectomy (81.3% vs. 68.2%, p = .225) [36]. Similarly, Ikeda et al. [37] analyzed 555 patients with stage IA NSCLC and reported comparable 5-year RFS (70.6% vs. 72.6%) and OS (92.9% vs. 83.6%) between the segmentectomy and lobectomy groups in STAS-positive patients. Conversely, wedge resection yielded significantly worse 5-year RFS (19.3% vs. 72.6%, p < .001) and OS (41.3% vs. 83.6%, p = .001) when compared to lobectomy [37]. While it is evident that wedge resection results in poorer clinical outcomes than lobectomy for STAS-positive patients, the comparative efficacy of segmentectomy versus lobectomy in stage I adenocarcinoma remains ambiguous. Therefore, prospective randomized clinical trials are necessary to clarify this risk stratification and establish definitive surgical guidelines.

Biological grade of the tumor and its surgical limitations

Moving beyond a binary assessment (presence vs. absence), a quantitative approach utilizing a high STAS (≥5 clusters) or 2,500 μm threshold reveals that extensive STAS confers an elevated recurrence risk (HR, 4.41) even within lobectomy cohorts [18]. This paradigm indicates that extensive STAS reflects an aggressive biological phenotype that transcends mere surgical margin inadequacy. Because standard lobectomy alone may be insufficient to control tumor dissemination in cases with extensive spread [18,23], meticulous postoperative surveillance and the proactive consideration of adjuvant chemotherapy might be warranted [38-40].

Integrated surgical and therapeutic decision-making (Fig. 7)

Synthesizing these complementary criteria may allow for the theoretical formulation of a precision surgical strategy. When planning a limited resection, the potential presence of STAS may be predicted via preoperative radiological findings [41] or assessed via intraoperative frozen sections [42-44]. If STAS is reliably identified intraoperatively, conversion to a standard lobectomy may be considered in an appropriate clinical setting. Following identification, quantifying the spatial extent of STAS further refines risk stratification. For limited dissemination (low STAS or STAS I, <2,500 μm), standard lobectomy generally provides adequate local control. However, if dissemination extends ≥2,500 μm or comprises ≥5 clusters, the patient would be reclassified into a high-risk subgroup, warranting multidisciplinary consideration regarding postoperative surveillance and the potential role of adjuvant systemic therapy, acknowledging that STAS currently serves primarily as a prognostic marker rather than a validated predictive biomarker (Fig. 7). Nevertheless, the clinical implementation of this decision-making framework faces logistical hurdles. The accurate preoperative prediction of STAS via computed tomography (CT) or positron emission tomography (PET) imaging remains challenging, and intraoperative detection using frozen sections is not yet fully established. Consequently, STAS is currently definitively identified only upon final pathological examination. Elucidating robust preoperative imaging predictors alongside improving the diagnostic accuracy of frozen section analyses is imperative. Thus, realizing this precision strategy is fundamentally contingent upon substantial advancements in the sensitivity and interobserver reproducibility of current diagnostic modalities, coupled with consensus within the thoracic pathology community.

INTRAOPERATIVE DIAGNOSIS OF STAS AND GROSS HANDLING CHALLENGES

Accurate intraoperative detection of STAS during sublobar resection is of clinical interest, yet it presents logistical challenges. Surgical atelectasis can complicate the differentiation of genuine tumor clusters from mechanical artifacts. Common mimickers include detached bronchial epithelial cells, alveolar macrophage aggregates, and jagged tumor edges from surgical sectioning (Fig. 3). Existing studies on frozen sections report moderate interobserver agreement (kappa values ranging from 0.34 to 0.67) and variable sensitivity (44% to 86%), highlighting a need for standardized diagnostic criteria [42-44]. Intraoperative lung inflation methods have shown potential in improving architectural preservation and STAS detection rates [43,45,46], and the future integration of artificial intelligence (AI)–assisted digital pathology tools may assist in this domain by providing automated tumor edge detection, objective recognition of STAS clusters against mimickers, and standardized distance-based grading.

PREOPERATIVE PREDICTION OF STAS

The preoperative prediction of STAS via imaging modalities such as CT or PET has evolved significantly, progressing from the visual assessment of traditional imaging features to the application of advanced AI algorithms [41,47-49]. While visual evaluation of conventional CT features establishes a foundational predictive baseline, it remains inherently subjective. Quantitatively, the consolidation-to-tumor ratio or percentage of solid component serves as the most robust independent predictor, with STAS predominantly observed in pure solid tumors rather than ground-glass nodules. Furthermore, larger tumor diameters (≥2 cm) and distinct morphological markers (lobulation, spiculation, pleural indentation, central low attenuation, and ill-defined margins) and a maximum standardized uptake value ≥ 2.5 are significantly correlated with STAS positivity [41,47-49]. It is important to note that many of these radiologic predictors (e.g., solid appearance, spiculation) are also indicative of broader aggressive tumor biology, such as solid or poorly differentiated histologic patterns [41]. Consequently, these imaging models may reflect general tumor aggressiveness that correlates with STAS, rather than features entirely specific to STAS itself. To mitigate subjective bias, radiomics translates CT images into objective, quantitative data [47]. Predictive accuracy is substantially improved when models integrate peritumoral features (extending 5 to 15 mm beyond the primary tumor margin) with clinical covariates (e.g., age, smoking history, and carcinoembryonic antigen levels), yielding area under the curve (AUC) values between 0.75 and 0.99. Currently, deep learning architectures—specifically three-dimensional convolutional neural networks employing attention mechanisms—represent the state-of-the-art methodology [48]. By automating complex feature extraction, these advanced models achieve superior predictive performance of STAS, with AUCs exceeding 0.93. While current imaging modality-based models show strong predictive efficacy, further standardization of the pathological definition of STAS and validation through multicenter prospective trials are necessary for widespread clinical implementation.

STAS AS AN EVOLUTIONARY BOTTLENECK: GENERAL MECHANISMS AND THE EGFR-MUTANT PARADIGM

Although STAS manifests less frequently in EGFR-mutated tumors compared to ALK or ROS1 cohorts, its presence within the EGFR subgroup appears to correlate with a notable prognostic disadvantage [27]. This observation suggests that in EGFR-mutant lung adenocarcinoma, STAS might represent a meaningful evolutionary transition, signifying the acquisition of an invasive aerogenous phenotype. When this switch engages—morphologically manifesting as STAS—it signifies that the tumor has successfully bypassed native biological barriers restricting dissemination, thereby elucidating the profound 5.04-fold elevation in recurrence risk observed within the stage IA EGFR-mutant cohort [27]. For early-stage disease, STAS-negative stage IA EGFR-mutant patients generally exhibit low recurrence rates, which escalate upon STAS detection. STAS may function as a histological proxy for occult micrometastatic disease. However, biological intent must be matched by physical capability. To substantiate this evolutionary bottleneck hypothesis, it is imperative to examine the general biophysical survival mechanisms of STAS across all subtypes. Upon detachment from the primary mass, these neoplastic cells must navigate the unique biomechanical constraints of the alveolar space, which lacks direct vascular scaffolding. Proposed biophysical models suggest that, from a fluid dynamics perspective, the survival and dissemination of STAS cells can be elucidated through the interplay of surface tension and a gradient effect. By exploiting surfactant layers, adapting to the periodic dynamic capillary pressure of alveolar deformation, and utilizing capillary bridges for strategic adhesion and detachment, we hypothesize that STAS cells establish a resilient metastatic niche. This advection-diffusion-reaction integration of cellular metabolism and fluidic mechanics provides the critical biological rationale for how STAS represents a genuine evolutionary adaptation rather than a passive artifact.

CLINICAL IMPLICATIONS: THE PROGNOSTIC ROLE OF STAS AND CONSIDERATIONS FOR ADJUVANT THERAPY

The primary clinical corollary of STAS is its potential utility in risk stratification for early-stage disease [38]. In stage IA EGFR-mutant lung adenocarcinoma, the presence of STAS escalates the 5-year recurrence rate from a baseline of 1.7% to 16.9% [27]. This substantial differential indicates that STAS serves as a visible histological proxy for occult micrometastasis present at the time of resection. While adjuvant osimertinib is the established standard of care for stage IB–IIIA disease [39,40], STAS-positive stage IA patients represent a clinically concealed, high-risk demographic that warrants careful multidisciplinary consideration. Consequently, STAS functions powerfully as a prognostic factor; for this specific cohort, biological risk may dictate a stage IB clinical approach despite stage IA anatomical dimensions [18,27]. Furthermore, significant intra-group heterogeneity exists within the EGFR-mutant cohort, with STAS manifesting less frequently in L858R mutations (30.7%) compared to exon 19 deletions (41.2%) [27]. This discrepancy implies that distinct primary EGFR-mutant subtypes may utilize divergent molecular mechanisms to facilitate aerogenous spread. Spatial transcriptomic profiling of these invasive cellular clusters is warranted to isolate subtype-specific gene signatures, potentially guiding the development of targeted anti-invasion therapies. These molecular observations align with emerging longitudinal trajectory models of lung cancer, which challenge the traditional staging paradigm that assumes a linear correlation between tumor volume and biological aggression. Instead, tumor evolution exhibits nonlinear, discrete transitions in aggressiveness [50]. We hypothesize that the manifestation of STAS or grade 3 (G3) histology indicates that a tumor has breached a critical evolutionary bottleneck, adapting to the biomechanical constraints of the alveolar microenvironment to secure a metastatic niche. At this threshold, the malignancy undergoes a functional biological upstaging, irrespective of its anatomical size, reinforcing the imperative for intercepting the disease prior to phenotypic expansion or initiating adjuvant therapy once this threshold is crossed [38]. However, it is crucial to acknowledge that current evidence advocating for adjuvant systemic therapy in STAS-positive stage IA disease is predominantly derived from retrospective observational cohorts. While this functional biological upstaging provides a compelling rationale for intervention, the lack of prospective, randomized controlled data limits its immediate integration into standard therapeutic guidelines as a predictive biomarker. Therefore, designing prospective clinical trials specifically tailored to this STAS-positive, early-stage demographic is an urgent prerequisite for validating its role as a definitive predictive biomarker for systemic therapy.

Ultimately, transitioning toward a multidimensional conceptual model—one that integrates oncogenic drivers, nonlinear phenotypic trajectories, and the complex biomechanics of STAS—will transform risk stratification in early-stage NSCLC. Future global pathology standards must transcend the simplistic, binary reporting of STAS, strictly contextualizing its presence within specific molecular profiles to optimize surgical and adjuvant management accurately.

CONCLUSIONS AND FUTURE DIRECTIONS

Accumulating data suggest that STAS is not merely a static morphological feature, but a potential indicator of divergent tumor biology. Its clinical interpretation appears most accurate when contextualized alongside specific molecular drivers. Despite these advancements, questions remain regarding the precise molecular delineation of high-risk patient cohorts and the potential role of adjuvant systemic regimens in this specific context. Addressing these uncertainties will require longitudinal data and prospective trials. Transitioning toward a multidimensional framework that integrates oncogenic drivers, growth trajectories, and the biomechanics of STAS may ultimately inform more nuanced risk stratification and therapeutic strategies in early-stage NSCLC.

Fig. 1.

Histologic features of spread through air spaces (STAS). STAS is defined by the dissemination of neoplastic cells into the pulmonary alveolar spaces extending beyond the primary tumor margin. Arrows indicate tumor cells in the alveolar spaces. Inset: higher magnification of STAS.

Fig. 2.

Representative images of spread through air spaces. Micropapillary (A), ring-shaped tumor cell clusters (B), solid tumor nests (C), and singly scattered floating tumor cells (D).

Fig. 3.

Common mimickers of spread through air spaces. (A) Detached bronchial epithelium. (B) Aggregates of alveolar macrophages. (C) Jagged edge of tumor cells. (D) Linear strips of tumor cells.

Fig. 4.

Distance-based grading of spread through air spaces (STAS). STAS I is defined as tumor clusters exclusively within <2,500 µm (exactly one field of a 10× objective lens) from the tumor edge. STAS II is defined by the presence of clusters ≥2,500 µm away.

Fig. 5.

Observed probability of spread through air spaces (STAS) in resected non-small cell lung cancer. STAS is more frequently identified in non–small cell lung cancer with adenocarcinoma histology, older patients, larger size, advanced stage, wild-type epidermal growth factor receptor (EGFR) status, or anaplastic lymphoma kinase (ALK)/ROS proto-oncogene 1 (ROS1) gene rearrangements.

Fig. 6.

Recurrence-free survival of stage IA adenocarcinoma (n=1,117) according to spread through air spaces (STAS) and epidermal growth factor receptor (EGFR) mutation status. (A) STAS (+) EGFR mutant subgroup showed worse clinical outcome. (B) In the EGFR-wild type cohort, STAS did not reach statistical significance as an independent recurrence predictor. Based on data reported in Lee et al. J Thorac Cardiovasc Surg 2026; 171: 1320-31 [27].

Fig. 7.

Hypothetical clinical management of patients based on spread through air spaces (STAS). Provided that pre- and intraoperative diagnostic modalities achieve sufficient accuracy, lobectomy can effectively secure the resection margins against limited STAS. For cases exhibiting extensive STAS, rigorous postoperative monitoring and consideration of adjuvant therapy are recommended. Crucially, the clinical integration of this treatment algorithm relies on significant improvements in diagnostic sensitivity and interobserver agreement, coupled with broader consensus from ongoing radiological and pathological research.

REFERENCES

• 1. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung Cancer Staging Project: proposals for revision of the TNM stage groupings in the forthcoming (eighth) edition of the TNM classification for lung cancer. J Thorac Oncol 2016; 11: 39-51. ArticlePubMed
• 2. Travis WD, Brambilla E, Nicholson AG, et al. The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol 2015; 10: 1243-60. ArticlePubMed
• 3. WHO Classification of Tumours Editorial Board. WHO Classification of Tumours. 5th ed. Vol. 5. Thoracic tumours. Lyon: International Agency for Research on Cancer, 2021.
• 4. Chen D, Mao Y, Wen J, et al. Tumor spread through air spaces in non-small cell lung cancer: a systematic review and meta-analysis. Ann Thorac Surg 2019; 108: 945-54. ArticlePubMed
• 5. Li J, Wang Y, Li J, Cao S, Che G. Meta-analysis of lobectomy and sublobar resection for stage I non-small cell lung cancer with spread through air spaces. Clin Lung Cancer 2022; 23: 208-13. ArticlePubMed
• 6. Liu H, Yin Q, Yang G, Qie P. Prognostic impact of tumor spread through air spaces in non-small cell lung cancers: a meta-analysis including 3564 patients. Pathol Oncol Res 2019; 25: 1303-10. ArticlePubMedPDF
• 7. Pyo JS, Kim NY. Clinicopathological impact of the spread through air space in non-small cell lung cancer: a meta-analysis. Diagnostics (Basel) 2022; 12: 1112.ArticlePubMedPMC
• 8. Travis WD, Eisele M, Nishimura KK, et al. The International Association for the Study of Lung Cancer (IASLC) Staging Project for Lung Cancer: recommendation to introduce spread through air spaces as a histologic descriptor in the ninth edition of the TNM classification of lung cancer. Analysis of 4061 pathologic stage I NSCLC. J Thorac Oncol 2024; 19: 1028-51. ArticlePubMed
• 9. Blaauwgeers H, Russell PA, Jones KD, Radonic T, Thunnissen E. Pulmonary loose tumor tissue fragments and spread through air spaces (STAS): invasive pattern or artifact? A critical review. Lung Cancer 2018; 123: 107-11. ArticlePubMed
• 10. Thunnissen E, Blaauwgeers HJ, de Cuba EM, Yick CY, Flieder DB. Ex vivo artifacts and histopathologic pitfalls in the lung. Arch Pathol Lab Med 2016; 140: 212-20. ArticlePubMedPDF
• 11. Pelosi G, Nesa F, Taietti D, et al. Spread of hyperplastic pulmonary neuroendocrine cells into air spaces (S.H.I.P.M.E.N.T.S): a proof for artifact. Lung Cancer 2019; 137: 43-7. ArticlePubMed
• 12. Blaauwgeers H, Flieder D, Warth A, et al. A prospective study of loose tissue fragments in non-small cell lung cancer resection specimens: an alternative view to "spread through air spaces". Am J Surg Pathol 2017; 41: 1226-30. ArticlePubMed
• 13. Blaauwgeers H, Dickhoff C, Pelosi G, et al. Con: "Is spread through air spaces an in vivo phenomenon or an inducible artifact?". J Thorac Oncol 2024; 19: 671-6. ArticlePubMed
• 14. Onozato ML, Kovach AE, Yeap BY, et al. Tumor islands in resected early-stage lung adenocarcinomas are associated with unique clinicopathologic and molecular characteristics and worse prognosis. Am J Surg Pathol 2013; 37: 287-94. ArticlePubMedPMC
• 15. Yagi Y, Aly RG, Tabata K, et al. Three-dimensional histologic, immunohistochemical, and multiplex immunofluorescence analyses of dynamic vessel co-option of spread through air spaces in lung adenocarcinoma. J Thorac Oncol 2020; 15: 589-600. ArticlePubMedPMC
• 16. Gross DJ, Hsieh MS, Li Y, et al. Spread through air spaces (STAS) in non-small cell lung carcinoma: evidence supportive of an in vivo phenomenon. Am J Surg Pathol 2021; 45: 1509-15. ArticlePubMedPMC
• 17. Li Y, Adusumilli PS, Chou TY, et al. Pro: "Is spread through air spaces an in vivo phenomenon or an inducible artifact?". J Thorac Oncol 2024; 19: 677-97. ArticlePubMed
• 18. Han YB, Kim H, Mino-Kenudson M, et al. Tumor spread through air spaces (STAS): prognostic significance of grading in non-small cell lung cancer. Mod Pathol 2021; 34: 549-61. ArticlePubMedPDF
• 19. Lee GY, Chung JH, Cho S, et al. Impact of preoperative diagnostic biopsy procedure on spread through airspaces and related outcomes in resected stage I non-small cell lung cancer. Chest 2022; 162: 1199-212. ArticlePubMed
• 20. Kameda K, Eguchi T, Lu S, et al. Preoperative needle biopsy and tumor spread through alveolar spaces (STAS) in resected lung adenocarcinomas. J Thorac Oncol 2017; 12(1 Suppl): S1458.Article
• 21. Warth A, Muley T, Kossakowski CA, et al. Prognostic impact of intra-alveolar tumor spread in pulmonary adenocarcinoma. Am J Surg Pathol 2015; 39: 793-801. ArticlePubMed
• 22. Uruga H, Fujii T, Fujimori S, Kohno T, Kishi K. Semiquantitative assessment of tumor spread through air spaces (STAS) in early-stage lung adenocarcinomas. J Thorac Oncol 2017; 12: 1046-51. ArticlePubMed
• 23. Lee J, Cho S, Chung JH, et al. Grading system of spread through air spaces is an independent predictor of recurrence in stage I invasive non-mucinous adenocarcinoma. J Thorac Oncol 2025; 20(3 Suppl 1): S99.
• 24. Lee JH, Kang Y, Kim S, et al. Clinical importance of grading tumor spread through air spaces in early-stage small-lung adenocarcinoma. Cancers (Basel) 2024; 16: 2218.ArticlePubMedPMC
• 25. Jin Y, Sun PL, Park SY, et al. Frequent aerogenous spread with decreased E-cadherin expression of ROS1-rearranged lung cancer predicts poor disease-free survival. Lung Cancer 2015; 89: 343-9. ArticlePubMed
• 26. Lee JS, Kim EK, Kim M, Shim HS. Genetic and clinicopathologic characteristics of lung adenocarcinoma with tumor spread through air spaces. Lung Cancer 2018; 123: 121-6. ArticlePubMed
• 27. Lee J, Han YB, Kim S, et al. Differential prevalence and prognostic significance of spread through air spaces according to oncogenic driver mutations in lung adenocarcinoma. J Thorac Cardiovasc Surg 2026; 171: 1320-31. ArticlePubMed
• 28. Lu S, Tan KS, Kadota K, et al. Spread through air spaces (STAS) is an independent predictor of recurrence and lung cancer-specific death in squamous cell carcinoma. J Thorac Oncol 2017; 12: 223-34. ArticlePubMedPMC
• 29. Lee J, Cho S, Chung JH, et al. Prognosis of spread through air spaces in invasive mucinous lung adenocarcinoma after curative surgery. Eur J Surg Oncol 2024; 50: 108053.ArticlePubMed
• 30. Chae M, Cho S, Chung JH, Yum S, Kim K, Jheon S. Poor prognosis of grade 2 spread through air spaces in neuroendocrine tumors. J Chest Surg 2022; 55: 101-7. ArticlePubMedPMC
• 31. Kadota K, Nitadori JI, Sima CS, et al. Tumor spread through air spaces is an important pattern of invasion and impacts the frequency and location of recurrences after limited resection for small stage I lung adenocarcinomas. J Thorac Oncol 2015; 10: 806-14. ArticlePubMedPMC
• 32. Masai K, Sakurai H, Sukeda A, et al. Prognostic impact of margin distance and tumor spread through air spaces in limited resection for primary lung cancer. J Thorac Oncol 2017; 12: 1788-97. ArticlePubMed
• 33. Chae M, Jeon JH, Chung JH, et al. Prognostic significance of tumor spread through air spaces in patients with stage IA part-solid lung adenocarcinoma after sublobar resection. Lung Cancer 2021; 152: 21-6. ArticlePubMed
• 34. Eguchi T, Kameda K, Lu S, et al. Lobectomy is associated with better outcomes than sublobar resection in spread through air spaces (STAS)-positive T1 lung adenocarcinoma: a propensity score-matched analysis. J Thorac Oncol 2019; 14: 87-98. ArticlePubMedPMC
• 35. Yang Z, Mino-Kenudson M, Lanuti M, Jeffrey Yang CF, Furlow PW. Extent of resection for spread-through air spaces-positive stage IA1-2 non-small cell lung cancer. J Thorac Cardiovasc Surg 2026; 171: 1310-9. ArticlePubMed
• 36. Kagimoto A, Tsutani Y, Kushitani K, et al. Segmentectomy vs lobectomy for clinical stage IA lung adenocarcinoma with spread through air spaces. Ann Thorac Surg 2021; 112: 935-43. ArticlePubMed
• 37. Ikeda T, Kadota K, Go T, Misaki N, Haba R, Yokomise H. Segmentectomy provides comparable outcomes to lobectomy for stage IA non-small cell lung cancer with spread through air spaces. Semin Thorac Cardiovasc Surg 2023; 35: 156-63. ArticlePubMed
• 38. Chen D, Wang X, Zhang F, et al. Could tumor spread through air spaces benefit from adjuvant chemotherapy in stage I lung adenocarcinoma? A multi-institutional study. Ther Adv Med Oncol 2020; 12: 1758835920978147.ArticlePubMedPMCPDF
• 39. Passaro A, Mok TS, Attili I, et al. Adjuvant treatments for surgically resected non-small cell lung cancer harboring EGFR mutations: a review. JAMA Oncol 2023; 9: 1124-31. ArticlePubMed
• 40. Tsuboi M, Herbst RS, John T, et al. Overall survival with osimertinib in resected EGFR-mutated NSCLC. N Engl J Med 2023; 389: 137-47. ArticlePubMed
• 41. Kim SK, Kim TJ, Chung MJ, et al. Lung adenocarcinoma: CT features associated with spread through air spaces. Radiology 2018; 289: 831-40. ArticlePubMed
• 42. Villalba JA, Shih AR, Sayo TM, et al. Accuracy and reproducibility of intraoperative assessment on tumor spread through air spaces in stage 1 lung adenocarcinomas. J Thorac Oncol 2021; 16: 619-29. ArticlePubMedPMC
• 43. Eguchi T, Matsuoka S, Iwaya M, et al. Improving intraoperative diagnosis of spread through air spaces: a cryo-embedding-medium inflation method for frozen section analysis. JTCVS Tech 2024; 25: 170-6. ArticlePubMedPMC
• 44. Cao H, Zheng Q, Deng C, et al. Prediction of spread through air spaces (STAS) by intraoperative frozen section for patients with cT1N0M0 invasive lung adenocarcinoma: a multi-center observational study (ECTOP-1016). Ann Surg 2025; 281: 187-92.
• 45. Myung JK, Choe G, Chung DH, et al. A simple inflation method for frozen section diagnosis of minute precancerous lesions of the lung. Lung Cancer 2008; 59: 198-202. ArticlePubMed
• 46. Xu X, Chung JH, Jheon S, et al. The accuracy of frozen section diagnosis of pulmonary nodules: evaluation of inflation method during intraoperative pathology consultation with cryosection. J Thorac Oncol 2010; 5: 39-44. ArticlePubMed
• 47. Jiang Q, Wang X, Jiang X, et al. A narrative review of preoperative CT for predicting spread through air spaces of lung cancer. Transl Lung Cancer Res 2025; 14: 1471-81. ArticlePubMedPMC
• 48. Cao J, Chen N, Zhou L, et al. A Bayesian deep learning model with consolidation-to-tumor ratio (CTR) prior revolutionizes the prediction of spread through air spaces (STAS) in stage IA lung adenocarcinoma: a large-scale diagnostic study. Transl Lung Cancer Res 2025; 14: 1516-30. ArticlePubMedPMC
• 49. Tasnim S, Raja S, Mukhopadhyay S, et al. Preoperative predictors of spread through air spaces in lung cancer. J Thorac Cardiovasc Surg 2024; 168: 660-9. ArticlePubMed
• 50. Asamura H, Nishimura KK, Giroux DJ, et al. IASLC Lung Cancer Staging Project: the new database to inform revisions in the ninth edition of the TNM classification of lung cancer. J Thorac Oncol 2023; 18: 564-75. ArticlePubMed

Figure & Data

References

Citations

Citations to this article as recorded by