Veta M, et al. - This was the first comparison study examining the evaluation of tumor proliferation from whole-slide images (WSIs). The challenge dataset included 500 of training and 321 testing breast cancer histopathology of WSIs. The challenge's first task was to predict mitotic scores, that is, to reproduce a pathologist's manual method of evaluating tumor proliferation. The second task was to predict PAM50 proliferation scores based on gene expression from the WSI. For the first task, the best performing automatic method achieved a quadratic-weighted Cohen kappa score of κ = 0.567, 95% CI between the predicted scores and ground truth. For the second task, the top method predictions had a Spearman's correlation coefficient of r = 0.617, 95% CI [0.581 0.651] with the ground truth. Given the difficulty of the tasks and the weakly labeled nature of ground truth, the results achieved are promising.