Abstract P1-07-10: Prediction of 10yr distant recurrence (DR) using the Prosigna® (PAM50) assay in histological subgroups of a Danish breast cancer group (DBCG) cohort of postmenopausal Danish women with hormone receptor-positive (HR+) early breast cancer (EBC) allocated to 5yr of endocrine therapy (ET) alone

Abstract

Abstract Background: The Prosigna (PAM50) risk of recurrence (ROR) score predicts 10yr DR in early breast cancer patients treated with ET alone (level 1 evidence). Invasive lobular breast cancer (ILBC) accounts for 10-15% of all breast cancer histological subtypes. Sporadic ILBC is characterized by somatic CDH1 mutations with loss of E-cadherin and a majority of low proliferative ER positive/HER2 negative tumors consistent with Luminal A subtype. Here we examine the ability of PAM50/ROR to predict 10yr DR in postmenopausal women who following a diagnosis of HR+ early ILBC were allocated to 5yr of ET alone. Methods: Using the population based clinical DBCG database FFPE primary tumor blocks, treatment, and follow-up data were collected from all patients diagnosed from 2000-2003 with HR+, postmenopausal EBC (N0-N1) who by nationwide guidelines were allocated to 5yr of ET alone, N=2,722 (1256 N0, 1466 N1). PAM50 intrinsic subtype classification (Luminal A, Luminal B, HER2-enriched, Basal-like) was conducted using the NanoString nCounter®Analysis System. Univariate and multivariate analyses tested the ability of PAM50 to predict DR. Patients were categorized as Low or High Risk based upon pre-specified ROR cutoff value of 40. HER2 positive tumors by immunohistochemistry were excluded from analysis. Results: Median follow-up was 9.25 years. Risk of 10yr DR by ROR and PAM50 (Luminal A and Luminal B) is shown in the table by ILBC as compared to invasive ductal breast cancer (IDBC) type. Cumulative incidence for 10 yr distant recurrence (DR)Histological subtypeRisk Group % [95% CI](N row%)Intrinsic Subtype % [95% CI](N row%)Histological subtype% [95% CI](N) LowHighLuminal ALuminal BAllDuctal3.8 [2.5-5.6] (738 35%)16.6 [14.4-18.8] (1388 65%)6.6 [5.1-8.4] (1174 59%)18.1 [15.2-21.3] (832 41%)12.1% [10.6-13.7] (2126)Lobular9.7 [5.5-15.4] (181 53%)23.9 [16.8-31.6] (159 47%)12.7 [8.6-17.8] (256 77%)24.1 [14.5-35.1] (77 23%)16.4 [12.2-21.1] (340) In this specific cohort the ILBC subgroup had a worse 10yr DR of 16.4% [12.2-21.1] as compared to IDBC of 12.1% [10.6-13.7] (Gray´s test, p = 0.046). A significant difference regarding the distribution of ILBC into both High and Low Risk Group and intrinsic subtypes as compared to IDBC was identified (p < 0.0001). Molecular intrinsic subtype analysis for ILBC by Prosigna (PAM50) showed DR 12.7 [8.6-17.8] for Luminal A vs 24.1 [14.5-35.1] for Luminal B. The difference between ILBC subtypes was not significant (p = 0.09). Adding ROR to a Fine and Gray's proportional sub-hazards model containing clinical and pathological variables significantly improved the model for ILBC, (likelihood ratio: p = 0.0001); HR for a 20-point change in ROR = 1.84 [1.37-2.48] notable with DR 9.7 [5.5-15.4] for the Low Risk Group. Conclusions: Following 5-yr of endocrine treatment alone patients with ILBC in this large population-based study had an inferior DR as compared to patients with IDBC and the apparent difference between ILBC and IDBC must be interpreted with caution. Citation Format: Laenkholm A-V, Jensen M-B, Buckingham W, Schaper C, Knoop A, Eriksen JO, Rasmussen BB, Ferree S, Haffner T, Kiboel T, Ejlertsen B. Prediction of 10yr distant recurrence (DR) using the Prosigna® (PAM50) assay in histological subgroups of a Danish breast cancer group (DBCG) cohort of postmenopausal Danish women with hormone receptor-positive (HR+) early breast cancer (EBC) allocated to 5yr of endocrine therapy (ET) alone [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 P1-07-10.