An immune-related signature for prediction of risk of late recurrences beyond proliferation and ER-related genes in ER-positive breast cancer.
Abstract No:
530
Attend this session at the
2014 ASCO Annual Meeting!
Session: Breast Cancer - HER2/ER
Type: Poster Highlights Session
Time 1: Sunday June 1, 8:00 AM to 11:00 AM
Location 1: E354b
Time 2: Sunday June 1, 11:30 AM to 12:45 PM
Location 2: E Arie Crown Theater
Author(s): Giampaolo Bianchini, Gabriella Rossi, Maurizio Callari, Takayuki Iwamoto, Eugenia Galeota, Thomas Karn, Catherine Margaret Kelly, Milvia Zambetti, Libero Santarpia, Lajos Pusztai, Luca Gianni; San Raffaele Hospital, Milan, Italy; Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; Okayama University Hospital, Okayama, Japan; Department of Obstetrics and Gynecology, Goethe University, Frankfurt, Germany; Department of Medical Oncology, Mater Misericordiae University Hospital, Dublin, Ireland; Cancer Center Humanitas, Milan, Italy; Yale School of Medicine, New Haven, CT
Abstract Disclosures
Abstract:
Background: In ER+ tumors, a dentritic metagene (DM) predicts for lower risk of relapse (Bianchini G JCO 2010; Bianchini G ASCO 2012). We investigated if other immune metagenes provided additional prognostic information and their involvement in early and late (after 5 years) relapses.
Methods: We evaluated public available Affymetrix-based gene expression profiles from ER+ untreated (n=599) and adjuvant tamoxifen (TAM)-treated pts (n=683). Four previously defined immune metagenes were evaluated: CD8, MHC1, STAT1 and interferon inducible (IF.I) (Gianni L SABCS 2012). Multivariate analyses were adjusted by estrogen-related genes, proliferation (Bianchini G Breast Cancer Res 2013) and HER2 status. Median cut-off points were used to define low and high expression groups. Outcome was assessed according to distant relapse.
Results: In untreated breast cancer, adjusting for other markers only IF.I metagene retained independent prognostic value [HR 1.19 (1.00-1.41), p=0.043], with higher risk for higher expression value. This effect was driven by its prognostic value in the late period [HR 1.50 (0.99-2.28), p=0.057]. Low and high expression groups of DM and IF.I were combined. The lowDM/highIF.I group had a higher risk of recurrences in the overall period [HR 3.68 (1.96-6.91), p<0.0001], with a similar trend in the late period [HR 2.98 (0.85-10.5), p=0.08]. In tamoxifen treated patients, adjusting for other markers, IF.I was confirmed as prognostic in the overall period [HR 1.25 (1.05-1.49), p=0.01] with an even stronger time varying effect [late period: HR 1.51 (1.14-2.01), p=0.004 and early period: HR 1.12 (0.90-1.39), p=0.29]. From 5 to 10 years, the lowDM/highIF.I group had the highest risk [HR 7.26 (1.78-33.9), p=0.006] with a corresponding relapse rate of 22.8% compared to 3% in the highDM/lowIF group.
Conclusions: In ER+ breast cancer, high expression of interferon inducible-related genes predicts for higher risk of late recurrences in untreated and tamoxifen-treated patients. Immune related-functions contribute to tumor dormancy and late relapses. Assessment of immune markers might contribute in tailoring extended adjuvant endocrine treatment.