Mifepristone-mediated induction of PD-L1 expression in hormone receptor positive breast tumors

Luminal breast cancer has been associated with a poor response to immunotherapy with immune checkpoint blockade (ICB). The selection of patients with high chances of response and the combination of ICB with other therapies, with the ability to turn a ‘cold' into a ‘hot' tumour, may have a broad impact on the management of this highly frequent type of breast cancer.We previously demonstrated that an endocrine therapy with Mifepristone (MIFE), an antagonist of the classical progesterone receptor, can foster a complete remodeling of the immune landscape in hormone receptor-positive tumours, promoting immune cell infiltration, immunogenicity and activating transcriptional programs associated with response to ICB both in mouse models and patients.Here, we aim to evaluate the effect of MIFE on the expression of the ICB response predictor biomarker PD-L1 in luminal breast human tumour samples derived from the MIPRA clinical trial (NCT02651844). The MIPRA trial enrolled 20 breast cancer patients harbouring HR+ tumours that were treated with MIFE for 14 days before surgery (orally administration of 200mg/day of MIFE). Response to MIFE (reduction in percentage of Ki67 proliferation marker > 30%) was documented in 14 of the 20 patients; however, all patients presented remodeling in the tumour microenvironment. We observed that treatment with MIFE in HR+ luminal breast tumours reorganizes the ECM and the stroma, including the enrichment in lymphocyte infiltration. Immunohistochemistry (IHC) assessment of PD-L1 expression in tumour slides from patients was analyzed in the Ventana BenchMark ULTRA using the anti-PD-L1 (SP263 VENTANA).Twenty untreated luminal tumours with similar clinical characteristics were included as controls. IHC analysis of tumour slides from 20 patients enrolled in the MIPRA trial showed that after MIFE treatment, 47% presented any staining for PD-L1, as compared to newly diagnosed untreated luminal tumours, where only 20% showed PD-L1 staining. Notably, the PD-L1 staining observed was primarily located in the stroma and surrounding the infiltration area (images taken with an Aperio SC2 scanner). We scored the staining on a scale of 0-5, with 5 being the highest area stained and 0 indicating no staining detected. MIPRA trial-participating patients exhibit a significant increase in the score of PD-L1 detection after MIFE treatment (p<0.05).These results suggest that endocrine therapy with MIFE can reprogram the immune landscape of luminal breast tumours in patients, promoting a “hot” TME, increasing TILs and PD-L1 expression. This MIFE-mediated immune fitness may sensitize luminal tumours to ICI therapy and open new therapeutic avenues for this prevalent breast cancer subtype.