Scholarlink

<p>Spread through air spaces (STAS) is a recently recognized pattern of invasion in lung cancer that is strongly linked to postoperative recurrence and poor survival. For patients with early-stage disease eligible for sublobar resection, intraoperative identification of STAS on frozen section (FS) could theoretically inform surgical decision-making. However, accumulating evidence, including recent analyses from the JCOG0802/WJOG4607L trial, suggests that STAS positivity portends a poor prognosis regardless of the extent of resection, indicating it is a marker of aggressive systemic biology rather than a purely surgically modifiable risk factor. Therefore, its primary clinical value may lie in postoperative risk stratification, guiding adjuvant strategies, and intensified surveillance rather than reflexive conversion to lobectomy. Yet FS assessment is difficult because of sampling limitations, morphologic heterogeneity, tissue-handling artifacts, and interobserver variability. This narrative review summarizes current knowledge of STAS, including its definition, histologic patterns, biological correlates, and prognostic impact across non-small cell lung cancer subtypes. We then critically appraise data on the feasibility, accuracy, and reproducibility of intraoperative FS for STAS, emphasizing diagnostic pitfalls and recent technical refinements such as optimized tumor-lung interface sampling and lung-inflation techniques. We also discuss emerging roles of artificial intelligence and digital pathology for automated STAS detection, along with radiologic, radiomic, and deep learning approaches for preoperative prediction. Finally, we address controversies surrounding biological versus artifactual STAS and reframe how STAS should influence a multidisciplinary management strategy that extends beyond the immediate choice of surgical extent. We propose practical recommendations for incorporating STAS assessment into multidisciplinary care and highlight future priorities, including prospective STAS-stratified trials, standardized FS workflows, and multimodal prediction models integrating imaging, molecular data, and AI-assisted pathology to support STAS-informed precision risk management.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Current progress and future directions of intraoperative frozen section for spread through air spaces in lung cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月9日

查看全文
  • Rirong Qu,
  • Shenghui Qin,
  • Jing Xiong,
  • Xiangning Fu,
  • Yixin Cai

摘要

Spread through air spaces (STAS) is a recently recognized pattern of invasion in lung cancer that is strongly linked to postoperative recurrence and poor survival. For patients with early-stage disease eligible for sublobar resection, intraoperative identification of STAS on frozen section (FS) could theoretically inform surgical decision-making. However, accumulating evidence, including recent analyses from the JCOG0802/WJOG4607L trial, suggests that STAS positivity portends a poor prognosis regardless of the extent of resection, indicating it is a marker of aggressive systemic biology rather than a purely surgically modifiable risk factor. Therefore, its primary clinical value may lie in postoperative risk stratification, guiding adjuvant strategies, and intensified surveillance rather than reflexive conversion to lobectomy. Yet FS assessment is difficult because of sampling limitations, morphologic heterogeneity, tissue-handling artifacts, and interobserver variability. This narrative review summarizes current knowledge of STAS, including its definition, histologic patterns, biological correlates, and prognostic impact across non-small cell lung cancer subtypes. We then critically appraise data on the feasibility, accuracy, and reproducibility of intraoperative FS for STAS, emphasizing diagnostic pitfalls and recent technical refinements such as optimized tumor-lung interface sampling and lung-inflation techniques. We also discuss emerging roles of artificial intelligence and digital pathology for automated STAS detection, along with radiologic, radiomic, and deep learning approaches for preoperative prediction. Finally, we address controversies surrounding biological versus artifactual STAS and reframe how STAS should influence a multidisciplinary management strategy that extends beyond the immediate choice of surgical extent. We propose practical recommendations for incorporating STAS assessment into multidisciplinary care and highlight future priorities, including prospective STAS-stratified trials, standardized FS workflows, and multimodal prediction models integrating imaging, molecular data, and AI-assisted pathology to support STAS-informed precision risk management.