29P - A negative progesterone receptor in luminal HER-2 negative breast cancer by age at diagnosis: 10 years follow-up

Date 08 May 2014
Event IMPAKT 2014
Session Welcome reception and Poster Walk
Topics Biomarkers
Breast Cancer
Presenter Siel Olbrecht
Citation Annals of Oncology (2014) 25 (suppl_1): i8-i16. 10.1093/annonc/mdu066
Authors S. Olbrecht1, K. Van Asten2, A. Laenen3, C. Remmerie4, H. Wildiers5, G. Floris6, M.R. Christiaens7, I. Vergote8, P. Neven8
  • 1Student Master Of Medicine, KU Leuven, 3000 - Leuven/BE
  • 2Oncology, KU Leuven, 3000 - Leuven/BE
  • 3Interuniversity Centre For Biostatistics And Statistical Bioinformatics, KU Leuven, 3000 - Leuven/BE
  • 4Datamanager, University Hospitals Leuven, - - Leuven/BE
  • 5Departement Of General Medical Oncology, University Hospitals Leuven, - - Leuven/BE
  • 6Departement Of Imaging And Pathology, KULeuven, 3000 - Leuven/BE
  • 7Departement Of Surgery, University Hospitals Leuven, - - Leuven/BE
  • 8Departement Of Gynecology And Obstetrics, University Hospitals Leuven, 3000 - Leuven/BE

Abstract

Background:

An absent progesterone receptor (PR) predicts distant metastases (DM) in ER-pos HER2 neg invasive breast cancer patients (BCP). It is unknown if PR remains prognostic in younger BCP (Yamamoto et al. JCO 2013). We investigated the prognostic effect of PR on DM free interval (DMFI) and BC specific survival (BCSS) by age at diagnosis.

Patients and methods:

Retrospective data of consecutive primary operable ER pos HER2 neg BCP (1/1/2000 - 1/6/2005 ) retrieved from our prospectively managed database. BCP with missing values for PR and/or lost to follow-up (FU) were excluded. Adjuvant radio- and systemic therapy were as indicated by our in-house protocol. A multivariable (MV) competing risk model for DMFI and BCSS was established considering age at diagnosis (age ≤ 50 yrs vs > 50 yrs), PR status (pos if >1% of cells stained with IHC), tumor grade (1–3), tumor size (mm) and lymph node status (neg/pos). Differential prognostic effects of these variables according to PR or age were tested by means of interaction effects. The MV model included significant interactions.

Results:

We included 1703 BCP (median FU 10.9 yrs); 1468 (86.20%) PR-pos [486(28.54%) ≤ 50 yrs and 982 (57.65%) > 50 yrs] and 235 PR-neg [49(2.88%) ≤ 50 yrs and 186 (10.9%) > 50 yrs]. In absolute numbers and compared to PR-pos BCP, PR-neg BCP > 50 yrs (15.9% of cases) had more DM (not if grade 1–2) and BCSS. This differed in BCP ≤ 50 yrs but only 9.1% were PR-neg . Results from the MV models showed no effect of PR on DMFI. For BCSS, the effect of PR significantly interacts with age (p = 0.02) resulting in a trend towards decreased risk for PR negative BCP ≤ 50 yrs (HR 0.88; 95% CI 0.45-1.70) whereas a trend for increased risk is observed for PR negative BCP >50 yrs (HR 1.67; 95% CI 0.77-3.59). Table 29P:

Percentage of patients with DM or BCSS according to age, PR, grade and lymph node status (pN).

Endpoint DM BCSS
Age ≤ 50 > 50 ≤ 50 >50
PR PR + PR- PR + PR- PR + PR- PR + PR-
Grade 1-2 13.1 16.7 8.1 8.4 9.0 8.3 3.9 6.72
Grade 3 25.2 20.0 23.2 38.8 17.2 8.0 14.8 28.3
pN neg 10.8 9.5 6.4 12.3 7.8 4.8 3.3 9.0
pN pos 26.3 25.0 21.6 32.0 17.4 10.7 12.6 25.0

Conclusions:

A differential prognostic effect of a negative PR on DM and BCSS by age at diagnosis is clinically observed but not maintained after correction for other tumor characteristics. Validation in a larger cohort would be needed to confirm our results.

Disclosure:

All authors have declared no conflicts of interest.