262 Chapter 17 the elastin stain shows otherwise. The Yambayev study provides more granularity in a classification that has proven low reproducibility14. The revised classification of chapters 14 and 15 is indebted to morphological principles. In 2012 a reproducibility study showed that one group of pathologists consistently judged a subset of adenocarcinomas to be invasive (Invasive Group, ING), while another group of pathologists consistently judged the same subset to be non-invasive (non-ING)14. This clearly points toward a necessity for calibration or re-evaluation of the diagnostic criteria. Conceptually, either the ING or the non-ING has the correct diagnosis. If the ING interpretation is correct, then the non-ING are underdiagnosing invasive lung cancer. The non-ING pathologists should have perceived since 2012 a number of patients with recurrences (and in the US heavy juridical consequences). What is not known in the field of pulmonary pathologists. If the non-ING interpretation is correct, the ING pathologists are over-diagnosing in-situ lung cancer as invasive lung cancer. Noteworthy, is that pathologists from one country had a higher tendency to belong to the ING group. Also, in other studies moderate interobserver agreement is found. Shih et al showed in an international panel of 3 pulmonary pathologists, who scored 60 cases of small adenocarcinomas in 3 rounds an interobserver agreement in small lung adenocarcinomas that fair to moderate with kappa scores of 0.44 or lower311. The results improved minimally with elastic stains. They concluded that poor agreement is primarily attributable to subjectivity in pattern recognition and that high-grade cytology increases agreement. The IALSC Pathology Committee recently published that the reproducibility of invasion assessment is still ‘suboptimal’, with a proposal for improvement, involving morphologic features of invasion in pulmonary non-mucinous adenocarcinoma with lepidic growth, and the effects of collapse on morphology317. Actually, high variation in invasion measurement is present in the group of pathologists that is frequently leading the changes in the WHO classification of pulmonary adenocarcinomas. In chapter 15 we describe the largest pathology reproducibility study to date on the recognition of non-invasive patterns in lung adenocarcinoma. The new heatmap analysis supported the recognition of larger and smaller common invasive areas in several cases as well as variation in size and location of invasion in other areas (likely non-invasive areas). This study also revealed that pathologists feel the uncertainty in assignment of invasion and together score opposite categories with associated high variation. Revised histological classification of pulmonary adenocarcinomas For the diagnosis of iatrogenic collapsed AIS in our studies, presented in the chapters 13-15 several aspects differ from the 2021 WHO classification of pulmonary adenocarcinomas.
RkJQdWJsaXNoZXIy MTk4NDMw