Abstract P2-05-16: Establishment of molecular profiling for individual treatment decisions in early breast cancer – Clinical impact of PAM50 and PAM50 risk of recurrence score after more than 16 years follow up

Abstract

Abstract Background Molecular profiling has recently been included in recommendations for decisions on adjuvant treatment in breast cancer (BrCa). However, the use of molecular profiling has not yet been widely established in all countries. Additional studies may give important information about the clinical relevance of the tests. Aims The study aims to discover the long term prognostic impact of PAM50 and PAM50 ROR score on survival for early BrCa pts according to treatment, with comparison to the routine clinical and histopathological parameters. Patients and methods Unselected early BrCa pts (n=651) from the Oslo Micromet project (n=920) having available FFPE primary tumor tissue were included in the current study. The pts were enrolled from 1995-1998. Follow up status is available for distant disease (median FU 7 years) and BrCa death (16.0-19.7 years after study inclusion). Clinical and histopathological parameters have been collected from the hospital records. FFPE tissue sections were macrodissected, RNA isolated from the dissected tumor tissue, followed by analysis of the PAM50 gene list on the Nanostring Platform. The samples were run in research mode and the raw data was sent to Nanostring (Seattle) for determination of the PAM50 subtype and ROR score. Results Of the 651 included pts, 323 did not receive any adjuvant systemic treatment (pT1pN0 patients), 161 tamoxifen only, the rest chemotherapy+/-tamoxifen. Twelve preoperatively treated pts were excluded from the analyses. Of the 639 remaining pts, PAM50 molecular profiling defined 52.3% as LumA, 26.8% LumB, 10.6% HER2enriched and 10.3% Basal. Multivariate analysis showed that the PAM50 intrinsic subtypes yielded prognostic information in addition to the established clinicopathological variables (pT, Grade, pN, age HR/HER2 subgroups, systemic treatment)(BCSS: HazardR vs LumA: 2.7 (95% CI 1.7-4.1) for LumB, 3.5 (1.8-6.8) for HER2enriched, 1.8 (0.8-4.2) for Basal). For the HR+HER2- pts, the risk classification by ROR score was an independent prognostic factor (BCSS: HazardR vs low risk: 3.1 (1.2-8.1) for intermediate, 6.6 (2.5-17.1) for high risk). In univariate analysis, the PAM50 intrinsic subtype classification separated clinical outcome both for all pts, for no adjuvant treated pts (both p<0.001, log rank), for the HR+HER2- (p<0.001), HR+HER2+ (p=0.061) and HR-HER2- (p=0.015) subgroups. Among the pT1-2pN0 HR+HER2- pts with no adjuvant treatment (n=222), risk classification by ROR score categorized 52.7% of the pts as low risk with excellent prognosis (BrCa death 4.2%), 29.7% as intermediate risk (BrCa death 16.7%) and 17.6% as high risk (BrCa death 35.9%)(p<001, log rank). For the pT1-2pN0-1 HR+HER2- pts who received adjuvant tamoxifen only (n=102), a low and similar risk of BrCa death was observed among the low and intermediate ROR risk groups. The high risk group had poor prognosis (BrCa death 32.7%)(p<0.001). Similar results were obtained for patients classified as LumA. Conclusions PAM50 subtype classification and ROR score improves classification of BrCa pts into prognostic groups, allowing more precise identification of future recurrence risk and improved basis for adjuvant treatment decisions. Citation Format: Naume B, Borgen E, Falk RS, Ohnstad HO, Lien TG, Aaserud M, Sveli MAT, Kyte JA, Kristensen V, Geitvik G, Schlichting E, Wist E, Sørlie T, Russnes H. Establishment of molecular profiling for individual treatment decisions in early breast cancer – Clinical impact of PAM50 and PAM50 risk of recurrence score after more than 16 years follow up [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P2-05-16.