Abstract P3-04-04: The prognostic effect of a negative progesterone receptor (PgR) by immunohistochemistry (IHC) in luminal HER-2 negative breast cancer (BC) by age at diagnosis: 10 years follow-up study

Abstract

Abstract Background After menopause, an absent PgR predicts distant metastases (DM) in patients with ER positive HER-2 negative invasive BC. Yamamoto et al. (JCO 2013) questioned the prognostic effect of PR in premenopausal patients, as ovarian estrogens might affect PgR expression. We investigated the effect of PgR on DM free interval (DMFI) and BC specific survival (BCSS) by age at diagnosis as surrogate marker for menopause. Patients and methods Retrospective data of consecutive patients with a primary operable ER positive HER-2 negative BC (1/1/2000 - 31/12/2009) were retrieved from our prospectively managed database. Cases that received neo-adjuvant chemotherapy or were operated in another hospital, were excluded. BC with missing values for PgR, grade or lymph node status and/or lost to follow-up were excluded in some subgroup analyses. A multivariate (MV) competing risk model for DMFI and BCSS was established considering age at diagnosis (age ≤50 yrs vs >50 yrs), PgR status, tumor grade (1-3), tumor size (mm) and lymph node status (neg/pos). Steroid receptors were considered positive if ≥1% stained on IHC. Differential prognostic effects of these variables according to PR or age were tested by means of interaction effects. Only significant interactions were included. Results We included 3326 BCs (8 with missing PgR status); 2911 (87.5%) PgR positive [870 (26.2%) ≤50 yrs and 2041 (61.4%) >50 yrs] and 407 (12.5%) PgR negative [68 (2.0%) ≤50 yrs and 339 (10.2%) >50 yrs]. In absolute numbers and compared to PgR positive BCs, PgR negative cases >50yrs had more DM if grade 3 and more BCSS for all grades. In BC ≤50 yrs these differences were not found, but only 7.2% were PgR negative (Tab.1). Results from the MV models showed a significant interaction with PgR on DMFI and BCSS (respectively p=0.01 and p=0.002) resulting in higher risk of DM (HR, 1.9; 95% CI, 1.4-2.7; p=0.001) and BCSS (HR, 2.4; 95% CI, 1.6-3.7; p<0.001) when PgR was absent for BCP >50 yrs. This difference in DMFI and BCSS according to PgR status was not found in BC ≤50 yrs (Tab.2). Table 1. Number of BCP with distant metastasis and BCSS according to age (yrs), PgR status(+/-) and tumor grade.Distant Metastatic DiseaseBreast Cancer Specific Survival≤50 yrs>50 yrs≤50 yrs>50 yrsPgR+PgR-PgR+PgR-PgR+PgR-PgR+PgR-Grade 1-245/607 (7.4%)5/42 (11.9%)66/1512 (4.4%)11/226 (4.9%)28/607 (4.6%)2/42 (4.8%)33/1511 (2.1%)7/226 (3.1%)Grade 350/263 (19.0%)4/26 (15.4%)69/529 (13.0%)35/113 (31.0%)31/263 (11.8%)2/26 (7.7%)44/529 (8.3%)26/113 (23.0%) Table 2. Results of MV model including significant main effects and interaction effects between tumor characteristics, age and PgR status on DMFI and BCSS. 95% Confidence interval InteractionsHazard ratioLower limitUpper limitP-valueA. BMFIPgR Neg. vs Pos. |-50yrs0.7840.3951.5560.4869PgR Neg. vs Pos. |+50yrs1.9371.3862.7090.0001B. BCSSPgR Neg. vs Pos. |-50yrs0.5750.2081.5880.2854PgR Neg. vs Pos. |+50yrs2.4391.6203.672<.0001 Conclusions A negative PgR is only prognostic for DMFI and BCSS in women aged >50 at diagnosis. There was no difference of PgR for these endpoints in patients ≤50 yrs at diagnosis but PR negativity is rare in this age group. Citation Format: Siel JAR Olbrecht, Kathleen Van Asten, Annouschka Laenen, Chantal Remmerie, Wildiers Hans, Floris Guiseppe, Christiaens Marie-Rose, Vergote Ignace, Neven Patrick. The prognostic effect of a negative progesterone receptor (PgR) by immunohistochemistry (IHC) in luminal HER-2 negative breast cancer (BC) by age at diagnosis: 10 years follow-up study [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-04-04.