Intrinsic subtypes and genomic signatures of primary breast cancer and prognosis after systemic relapse. Issue 4 (18th November 2015)

• Record Type: Journal Article
• Authors: Falato, Claudette; Tobin, Nicholas P.; Lorent, Julie; Lindström, Linda S.; Bergh, Jonas; Foukakis, Theodoros
• Is Part Of: Molecular oncology. Volume 10:Issue 4(2016:Apr.)
• Page Start: 517
• Page End: 525
• Publication Date: 2015-11-18
• Subjects: Intrinsic subtypes -- Risk of recurrence score -- Proliferation score -- Gene expression signature -- Metastatic breast cancer; Cancer -- Molecular aspects -- Periodicals; 616.994005
• DOI: 10.1016/j.molonc.2015.11.004
• Languages: English
• ISSNs: 1574-7891

Abstract

Abstract : Molecular subtypes and gene expression signatures are widely used in early breast cancer but their role in metastatic disease is less explored. Two hundred‐twenty patients diagnosed with primary breast cancer and subsequent relapse in Stockholm, Sweden between 1997 and 2006 were identified and their primary tumor was assessed for immunohistochemistry (IHC)‐ and PAM50‐based subtypes, risk of recurrence (ROR‐S) score, 21‐gene and 70‐gene signatures using research‐based microarray expression profiles. Clinical and pathological data were retrospectively collected. Post‐relapse survival within intrinsic subtypes and genomic signatures was investigated by Kaplan–Meier and Cox regression methods. ROR weighted for proliferation index (ROR‐P) was explored and the prognostic contribution provided when combined to a clinical model estimated as change in LR‐ χ2. IHC classified 27%, 24%, 36% and 13% of the tumors as luminal A, luminal B, HER2+ and triple negative, respectively. PAM50 categorized 22%, 24%, 26%, 22%, 6% of the tumors as luminal A, luminal B, HER2‐enriched, basal‐like and normal‐like. Triple negative and basal tumors had a significantly shorter median post‐relapse survival in comparison with luminal. Overall, neither IHC nor PAM50 subtypes, 21‐ and 70‐ gene profiles were prognostic in multivariable models. Low and medium ROR‐S had a longer survival compared with the high‐risk group (23 vs 10 months; p = 0.04). ROR‐P independently correlated with post‐relapse survival (p = 0.002) and provided the most significant prognostic information when added to a clinical model. ROR score from primary tumor represents an independent prognostic factor of post‐relapse survival beyond classical clinical and pathological variables. Highlights: Patients diagnosed with primary breast cancer and subsequent relapse between 1997 and 2006 were retrospectively identified. The primary tumor was assessed for immunohistochemistry‐based intrinsic subtypes and genomic profiles. Risk of Recurrence score represents an independent prognostic factor of post‐ systemic relapse survival. Risk of Recurrence score was prognostic of better survival in patients treated with endocrine therapy for metastatic disease.