Interobserver variability in the application of the novel IASLC/ATS/ERS classification for pulmonary adenocarcinomas

Abstract

Recently, a novel classification for pulmonary adenocarcinomas (ADCs) was published, the cornerstone of which is the quantification of growth patterns. Data on reproducibility in the routine diagnostic setting are lacking. 100 constitutive cases of lung ADC resection specimens from our archives were classified independently by five pulmonary pathologists and two residents according to the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society classification. The most frequent predominant pattern in our cohort was solid (37%), followed by acinar (35%), lepidic (20%), papillary (5%) and micropapillary (3%). κ-values for the denomination of the predominant pattern revealed substantial agreement for pulmonary pathologists (κ=0.44-0.72) and fair agreement for residents (κ=0.38-0.47). Interobserver variability was significantly higher in cases with higher slide numbers (p=0.028) and was considerably reduced after training. Intraobserver variability was low (κ=0.79-0.87). Papillary and micropapillary patterns were the most complicated patterns to evaluate, while evaluation of lepidic and solid tumour growth was straightforward. Our data imply that the novel classification of pulmonary ADC is applicable with acceptable interobserver variability if performed by specifically trained pathologists. Additional efforts are needed to harmonise the application of this novel and clinically important classification scheme of pulmonary ADC.