Rabe K, et al. - A total of 100 consecutive cases of invasive stage II carcinomas were individually graded twice, with a 4-week intervening wash-out period, via six breast pathologists using secured Nottingham grading criteria in order to evaluate the degree of interobserver agreement for invasive carcinoma grade among pathologists and determine its impact on pathologic prognostic stage. With even better intra-observer variability, interobserver variability was good to very good. The most reproducible factor was the tubule score. In rounds 1 and 2 respectively, complete concordance was struck in 54 cases and 58 cases. In round 1, in only 25 of discordant cases, this led to the various pathologic prognostic stages, 18 of which were stage IA vs IB. Thus, grading agreement between pathologists was good to very good and discordant grades led to minute variations to the pathologic prognostic stage.