P2-12-04: RACGAP1 mRNA Assay Outperforms Ki-67 as a Proliferation Marker in the FinHer Study Cohort.

Abstract

Abstract Background: Molecular subtyping of breast cancer has become an integral part of standard evaluation of breast cancer patients. Their assessment requires combining data from analyses on ER, PR, HER2 and cell proliferation markers. However, their immunohistochemical (IHC) testing carries an up to 20% risk of erroneous results. Similarly, assessment of cell proliferation by Ki-67 staining is hampered by lack of standardization of laboratory methods and agreement on cut-offs. Here we tested the prognostic value of objective quantitation of ESR1, PgR, HER2 and the proliferation markers RACGAP1 using RT-qPCR and compared the results with local and central IHC assessments. Methods: RNA was extracted from FFPE tumor tissue of 917 patients who participated in the FinHer trial. ESR1, PgR, HER2 and RACGAP1 mRNA expression were measured using RT-qPCR. The molecular subtypes (luminal, HER2−enriched and triple-negative) were determined. Prognostic significance of proliferation markers was assessed using univariate and multivariate analyses. The RT-qPCR results were compared with local and central IHC results. Results: HER2 mRNA showed a bimodal distribution with 197 (21.4%) out of the 917 tumors being above the predefined cut-off. HER2 mRNA expression increased in parallel with HER protein expression. Overall concordance of HER2 mRNA testing with central IHC and CISH was good, while local IHC testing suffered higher false positive rates. RACGAP1 mRNA expression was the greater the higher the histological grade. ESR1 and PgR mRNA correlated negatively with the histological grade (r=-0.38 and r=-0.33; p<0.0001), whereas HER2 and RACGAP1 mRNA were correlated positively (r=0.10 and r=0.49; p=0.002 and p<0.0001, respectively). RACGAP1 mRNA was negatively associated with ESR1 and PgR mRNA (r=-0.17 and r=-0.26, respectively; p<0.0001 for each). Molecular subtypes determined by RT-qPCR using predefined cut-off values were highly prognostic for overall survival (OS) (p<0.001). The 5-year OS rate for patients with luminal cancer was 94% and 86% for HER2−enriched cancer and 84% for triple-negative cancer. In the subset of luminal tumors, high expression of RACGAP1 identified a population of patients who were at a high risk of death (5-year OS 82% versus 95%; p<0.0001). In a multivariate analysis RACGAP1 mRNA expression, nodal status and chemotherapy type were independent prognostic factors, whereas IHC of ER, PgR, Ki-67 and histological grade were not significant. Conclusions: Molecular subtyping of breast cancer by RT-qPCR using RNA isolated from FFPE tissue proved successfully in this large patient cohort. RACGAP1 mRNA expression distinguished high and low risk luminal breast cancers. In a multivariate analysis mRNA-based molecular markers outperformed the immunohistochemical markers ER, PgR and Ki-67. Of note, quantitative assessment of the proliferation marker RACGAP1 was superior to semiquantitative assessment of Ki-67 from routine FFPE tissues using IHC. We conclude that quantitative assessment of ESR1, PgR, HER2 and RACGAP1 mRNA by RT-qPCR is a robust and reproducible method to assess these key tumor biological factors from archival FFPE tumor tissue. RACGAP1 is novel cell proliferation marker in breast cancer that warrants further validation. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-12-04.