Abstract P1-07-11: Characterization of progesterone receptor biomarker for predicting antiprogestin activity in human cancers.

Abstract

Abstract BACKGROUND: Progesterone receptors (PR) are important prognostic biomarkers of estrogen receptor (ER) action in breast cancer (BC). More recently, PRs have emerged as independent (from ER) mediators of early BC progression. In advanced BC phase II trials, the antiprogestin (antiPg) onapristone demonstrated a 10% and 56% OR in second line and first line therapies respectively, while mifepristone demonstrated 10% and 12%. Upon ligand binding, transcriptionally active PR in normal tissues form a discrete focal subnuclear distribution pattern (FDP) that can be visualized by immunofluorescence (Arnett-Mansfield, 2004, 2007). Importantly, PR nuclear foci are indicative of transcriptional activity. FDP is also observed in breast and endometrial cancers independently of hormonal status. The goal of this study was to devise a method to identify patients that are more likely to benefit from antiPg treatment. METHOD: 12 PR [+] BC paraffin-embedded blocks were analyzed. Standard IHC was used with 4 Abs: anti-PR-A (PRA), PR-B (PRB), anti-PR-A+B, anti-PR A and B (PRAB). Samples were analyzed for each Ab, and a control for the primary Ab, with and without background staining. After standard histologic evaluation subnuclear structures were analyzed at 100X. IHC was performed using Abs directed to the following PR phosphorylated sites (pAb) pSer 162, 190, 294, 400 and 554. Six samples were selected for IHF using a secondary fluorescent Ab. RESULTS: All cases were PR [+]. At high magnification (100X), 2 PR nuclear distribution patterns were observed: a diffuse finely granular staining (D) or an aggregated pattern (A). This resulted in 3 tumor phenotypes (PR B Ab): A cells only in 2 cases, D cells only in 6, and a mixture of both A and D cells (AD) in 4. PR [−] malignant cells were present in various proportions. The IHF results were consistent with IHC studies. Only pAb to pSER190 was specific and sensitive, with a smaller number of [+] malignant cells relative to the standard PRB. CONCLUSION: We have developed an IHC method that utilizes formalin-fixed paraffin-embedded tissue allowing the technique to be applied on a routine basis. Three classes of PR nuclear distribution have been defined: D, A and AD. D is consistent with the expression of PR that is weakly transcriptionally active or inactive and thus potentially predictive of poor patient response to antiPg. The A or AD pattern of PR nuclear distribution indicates active PR; patients with this pattern may exhibit a increased probability of benefiting from an antiPg. Preclinical studies have demonstrated that activated PR [+] cells can in turn stimulate PR [−] (stem or progenitor) mammary epithelial cells via secreted factors. In light of these findings, we are expanding the cohort to refine and confirm this biomarker technique. Additional preclinical work is ongoing to expand the biological rationale and to evaluate this technique alongside other technologies such as PR gene expression analysis. We anticipate that approximately 50% of PR+ BC cases contain the A/AD phenotye and therefore may benefit from antiPg. The method of IHC described herein is a clinically applicable tool that may allow for selection of patients most likely to respond to antiPg treatment. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P1-07-11.