256 Chapter 17 pneumocytes or bronchial cells with benign cytologic features and/or presence of cilia; or (4) linear strips of cells that are lifted off alveolar walls.” They state that using these criteria, separation of STAS from artifacts was highly reproducible (average kappa 0.86) in a study of selected images, based on a poster abstract by Baine et al.372. From one of the participants in that study, we learned that it was a highly selected set of easy cases. However, to be applicable in routine practice, the diagnostic criteria must be clearly defined to ensure consistency over time (intra-observer reproducibility) in a consecutive series of cases and lead to the same diagnosis between different pathologists (low inter-observer variability). Research on inter-observer reproducibility can be categorized as either local, involving a few pathologists within an institute or region, or global, involving pathologists from various parts of the world. The latter type of study is considered to be more robust. Informative in this sense are studies examining STAS in frozen sections. Three studies have reported a sensitivity range of 55-86% for STAS204 205 172. Villalba and colleagues, who are regional pathologists with a particular focus on pulmonary pathology, have clearly articulated the concerns regarding STAS analysis in frozen sections in their paper’s summary: “As current accepted definitions for STAS and artifactual clusters are variably interpreted by pathologists, more precise criteria should be established and standardized, before the assessment of STAS can be implemented globally in the intraoperative setting to aid surgical decision making”204. Therefore, this “easy distinction” of STAS as mentioned in the 2021 WHO classification371 seems far too optimistic. To better comprehend the discourse surrounding STAS, it is important to note that the WHO provides morphological criteria for artifacts, and if these are absent, a “loose tumor fragment” aligns with the definition of STAS. However, a topological component also plays a role. Quote “The morphological patterns of STAS, namely single cells, micropapillary and solid tumour cell clusters are artifacts if present without continuity with the (invasive) tumor border or are present at a too far distant from this tumor border.” Therefore, although “loose tumor fragments” may appear morphologically consistent with STAS, without proper topological annotation, they should be considered an “artifact” under the WHO criteria. This may aid in the confusion among pathologists and could partly account for the low reproducibility of STAS. In conclusion, there are currently no clear and consistent criteria for STAS. This was also the opinion of > 100 responders from a recent questionnaire distributed in the Pulmonary Pathology Society network, presented at USCAP meeting in New Orleans, March 2023. Possible causes of “STAS” After our chapter 6 publication in 2017, Metovic and colleagues attempted to validate our findings in a prospective study, comparable to ours, except that they used a fresh knife blade for every cut they made during gross handling of the resection specimen175. However, they didn’t find a correlation between the amount of STAS and the direction of the cuts made, as we did in our study. In chapter 9 we commented on the fact that they also found tumor cells or fragment in the opposite direction the cutting. We argued
RkJQdWJsaXNoZXIy MTk4NDMw