Ki-67 proliferation index (PI) is essential for grading well-differentiated neuroendocrine tumors (WD-NETs). Pathologists traditionally assess Ki-67 PI by identifying hotspots and manually counting the positive cells among negative cells, which is expressed as a percentage. We developed an algorithm to objectively determine Ki-67 hotspots and calculate PI in WD-NETs, comparing its results with pathologists' selected hotspots to assess reliability. Hotspots for gastroenteropancreatic WD-NETs ( n = 20) were manually annotated on whole-slide images (WSIs) by six pathologists and compared with algorithm-selected areas. Ki-67 (DAKO, MIB-1 clone, and prediluted) scoring was performed using QuPath's custom object classification algorithm. Pathologists identified hotspots on WSI, captured images, and submitted them for PI determination using the same algorithm. Ki-67 PI was translated to grade per WHO classification (G1: 20%). A pathologist's consensus grade was determined based on majority pathologist grading (>3/6) for each case. Fleiss's Kappa was used to assess inter-pathologist agreement, Cohen's Kappa was used to evaluate the agreement between pathologists and the algorithm, and Friedman test was used for hotspot area variability analysis. Pathologists showed moderate agreement (Fleiss's Kappa = 0.42, 80% agreement), whereas pathologist–algorithm agreement was fair (Cohen's Kappa = 0.32, 58.9% agreement). Among cases with pathologist consensus grade ( n = 19), the algorithm assigned a higher grade in 8 cases (42%). In 60% of cases, hotspots overlapped between methods. There was significant hotspot area variability (Friedman statistic: 95.97, p < 0.001). Manual Ki-67 hotspot assessment is subjective, leading to grading variability. Algorithm-based assessment enhances reproducibility, though this occasionally leads to tumor upgrading, highlighting the need for standardization and further validation. • Subjectivity exists among pathologists in determining hotspots for assessing the Ki-67 proliferation index in well-differentiated neuroendocrine tumors using visual hotspot selection and manual counting. • Computer-assisted image analysis offers a more objective approach, showing moderate agreement with pathologists. • The algorithm-based analysis has a tendency to assign higher tumor grades, highlighting the need for standardized hotspot selection criteria. • Further validation is needed to assess the clinical impact of automated grading on objective outcome measures.
Oganesyan et al. (Sun,) studied this question.