Abstract 483: Automated Ki67-LI assessment in prostate cancer using artificial intelligence in multiplex fluorescence immunohistochemistry

Abstract The Ki67 labeling index (Ki67-LI) is a strong prognostic marker in prostate cancer. Its analysis requires cumbersome manual quantification of Ki67 immunostaining in at least 200 tumor cells. To enable automated Ki67-LI assessment in routine clinical practice, we have developed and validated a framework for automated Ki67-LI quantification, which comprises three different artificial intelligence analysis steps and an algorithm for cell-distance analysis of multiplex fluorescence immunohistochemistry staining. The prognostic impact of the Ki67-LI was tested on a tissue microarray (TMA) containing one sample each of 12475 prostate cancers. A “heterogeneity TMA” containing 3 to 6 samples from different tumor areas was used to model Ki67 analysis of multiple different biopsies. The Ki67-LI provided strong and independent prognostic information in 11845 successfully analyzed prostate cancers (p<0.001 each). The analysis of the heterogeneity TMA revealed that the Ki67-LI of the sample with the highest Gleason score (AUC:0.68) was similarly prognostic as the mean Ki67-LI of all six foci (AUC:0.71 [p=0.24]). The combined analysis of the Ki67-LI and Gleason grades obtained on identical tissue spots showed that the Ki67-LI added significant additional prognostic information in case of classical ISUP grades (AUC:0.82 [p=0.002]) and quantitative Gleason grades (AUC:0.83 [p=0.018]). The Ki67-LI is a powerful prognostic parameter in prostate cancer, which can be efficiently analyzed in multiplex fluorescence IHC. In case of multiple cancer positive biopsies, the sole analysis of the worst biopsy can be sufficient. Citation Format: Niclas C. Blessin, Tim Mandelkow, Elena Bady, Ronald Simon, Claudia Hube-Magg, Maximilian Lennartz, Guido Sauter, Markus Graefen, Stefan Steurer. Automated Ki67-LI assessment in prostate cancer using artificial intelligence in multiplex fluorescence immunohistochemistry [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 483.