Comparative diagnostic and prognostic performances of the hematoxylin-eosin and phospho-histone H3 mitotic count and Ki-67 index in adrenocortical carcinoma.

Mitotic count on hematoxylin and eosin slides is a fundamental morphological criterion in the diagnosis and grading of adrenocortical carcinoma in any scoring system employed. Moreover, it is the unique term strongly associated with patient's prognosis. Phospho-histone H3 is a mitosis-specific antibody, which was already proven to facilitate mitotic count in melanoma and other tumors. Therefore, a study was designed to assess the diagnostic and prognostic role of phospho-histone H3 in 52 adrenocortical carcinomas, comparing manual and computerized count to standard manual hematoxylin- and eosin-based method and Ki-67 index. Manual hematoxylin and eosin and phospho-histone H3 mitotic counts were highly correlated (r=0.9077, P<0.0001), better than computer-assisted phospho-histone H3 evaluations, and had an excellent inter-observer reproducibility at Bland-Altman analysis. Three of 15 cases having <5 mitotic figures per 50 high-power fields by standard count on hematoxylin and eosin gained the mitotic figure point of Weiss Score after a manual count on phospho-histone H3 slides. Traditional mitotic count confirmed to be a strong predictor of overall survival (P=0.0043), better than phospho-histone H3-based evaluation (P=0.051), but not as strong as the Ki-67 index (P<0.0001). The latter further segregated adrenocortical carcinomas into three prognostic groups, stratifying cases by low (<20%), intermediate (20-50%), and high (>50%) Ki-67 values. We conclude that (a) phospho-histone H3 staining is a useful diagnostic complementary tool to standard hematoxylin and eosin mitotic count, enabling optimal mitotic figure evaluation (including atypical mitotic figures) even in adrenocortical carcinomas with a low mitotic index and with a very high reproducibility; (b) Ki-67 proved to be the best prognostic indicator of overall survival, being superior to the mitotic index, irrespective of the method (standard on hematoxylin and eosin or phospho-histone H3-based) used to count mitotic figures.