P4-09-23: Monitoring Autophosphorylation of Mammalian Target of Rapamycin (mTOR) for Histoprognostic Grading of Invasive Breast Cancer: Impact on the Reclassification of Patients with Grade 2 Tumors Using the Nottingham Grading System.

Abstract

Abstract Histological grade is one of the strongest prognostic factors in operable breast carcinomas (BC). Although Grade 1 and Grade 3 are biologically and clinically distinct tumor entities that clearly separate low-risk from high-risk BC patients, 30–60% of BC are classified as histological Grade 2, which is associated with an intermediate risk of recurrence and is not informative for clinical decision making. Previous studies using a 97-gene measure of histological tumor grade (i.e. the Genomic Grade Index, GGI) have confirmed GGI's ability to accurately reclassify Grade 2 patients into 2 BC groups with high versus low risks of recurrence. However, GGI's applicability is still expensive, technically demanding and is not readily available in a routine clinical setting in most hospital laboratories. The prognostic abilities of BC gene expression signatures such as GGI are due mostly to the detection of proliferation activity. One of the strongest, yet simple and well-reproducible proliferation-associated prognostic factors is the mitotic activity index (MAI). Here, we have tested whether immunohistochemical assessment of MAI by assessing the autophosphorylation status of mammalian Target Of Rapamycin (mTOR) at Serine 2481 could significantly impact on the histopathological classification of Grade 2 BC when using the Nottingham Grading System. 1.) We validated the sensitivity of phospho-mTORSer2481 (PP-mTOR) labeling in detecting & counting mitotic figures and also its usefulness for histoprognostic grading in a series of 144 BC biopsies; 2.) we investigated the correlation between PP-mTOR MAI and the MAI determined by using an antibody selective for the Ser10-phosphorylated Histone H3 (PP-H3), a well-recognized mitosis-specific marker. PP-mTOR-labeled mitotic figures were easily seen and permitted a quick identification of the area of highest mitotic activity, even at low-power magnification. Our study showed a statistically significant correlation between the subjective mitotic counts obtained by using PP-mTOR labeling and those assessed by either standard H&E (r = 0.662) or staining with PP-H3 (r = 0.896). PP-mTOR MAI correlated also with tumor proliferative activity as measured with the Ki-67 labeling index (r = 0.633). Average mitotic counts were significantly higher when using labeling with PP-mTOR (range = 0–164 mitotic figures, mean = 17) and PP-H3 (0-146, 19) than with the standard H&E protocol (0-60, 5). When the global histoprognostic score of the Nottingham Grading System was re-evaluated on the basis of PP-mTOR staining, there was a statistically significant shift from Grade 1 to Grade 2 in 14 cases, and from Grade 2 to Grade 3 in 23 cases. Indeed, using PP-mTOR MAI instead of standard H&E-determined mitotic counts drastically reduced by ∼50% the number of Grade 2 tumors while increasing ∼3 times the number of Grade 3 BC. Our findings reveal for the first time that immunolabeling with PP-mTORSer2481 constitutes a simple & reliable method for quantifying proliferative potential of tumors that notably improves Nottingham histoprognostic grading of invasive BC. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-09-23.