The programmed death protein 1 (PD-1) and its ligand (PD-L1) represent a well-characterized immune checkpoint in cancer, effectively targeted by monoclonal antibodies that are approved for routine clinical use. The regulation of PD-L1 expression is complex, varies between different tumor types and occurs at the genetic, transcriptional and post-transcriptional levels. Copy number alterations of PD-L1 locus have been reported with varying frequency in several tumor types. At the transcriptional level, a number of transcriptional factors seem to regulate PD-L1 expression including HIF-1, STAT3, NF-κΒ, and AP-1. Activation of common oncogenic pathways such as JAK/STAT, RAS/ERK, or PI3K/AKT/MTOR, as well as treatment with cytotoxic agents have also been shown to affect tumoral PD-L1 expression. Correlative studies of clinical trials with PD-1/PD-L1 inhibitors have so far shown markedly discordant results regarding the value of PD-L1 expression as a marker of response to treatment. As the indications for immune checkpoint inhibition broaden, understanding the regulation of PD-L1 in cancer will be of utmost importance for defining its role as predictive marker but also for optimizing strategies for cancer immunotherapy. Here, we review the current knowledge of PD-L1 regulation, and its use as biomarker and as therapeutic target in cancer.

BackgroundCTCs expressing variable levels of epithelial and mesenchymal markers in breast cancer have previously been reported. However, no information exists for keratin expression levels of CTCs in association with disease status, whereas assays for the characterization of transitional EMT phenotypes of CTCs in breast cancer are rather lacking. We investigated the correlation between keratin expression of CTCs and patients’ outcome and characterized the EMT status of CTCs via the establishment of a numerical “ratio” value of keratin and vimentin expression levels on a single cell basis.MethodsKeratin expression was evaluated in 1262 CTCs from 61 CTC-positive patients with metastatic breast cancer, using analysis of images obtained through the CellSearch System. For the determination of vimentin/keratin (vim/K) ratios, expression levels of keratin and vimentin were measured in cytospin preparations of luminal (MCF-7 and T47D) and basal (MDA.MB231 and Hs578T) breast cancer cell lines and 110 CTCs from 5 CTC-positive patients using triple immunofluorescence laser scanning microscopy and image analysis.ResultsMCF-7 and T47D displayed lower vim/K ratios compared to MDA.MB231 and Hs578T cells, while MCF-7 cells that had experimentally undergone EMT were characterized by varying intermediate vim/K ratios. CTCs were consisted of an heterogeneous population presenting variable vim/K values with 46% of them being in the range of luminal breast cancer cell lines. Keratin expression levels of CTCs detected by the CellSearch System correlated with triple negative (p = 0.039) and ER-negative (p = 0.025) breast cancer, and overall survival (p = 0.038).ConclusionsKeratin expression levels of CTCs correlate with tumor characteristics and clinical outcome. Moreover, CTCs display significant heterogeneity in terms of the degree of EMT phenotype that probably reflects differential invasive potential. The assessment of the vim/K ratios as a surrogate marker for the EMT status of CTCs merits further investigation as a prognostic tool in breast cancer.Electronic supplementary materialThe online version of this article (doi:10.1186/s12885-015-1386-7) contains supplementary material, which is available to authorized users.

Purpose: Conflicting data have been reported on the prognostic value of PD-L1 protein and gene expression in breast cancer. Experimental Design: Medline, Embase, Cochrane Library, and Web of Science Core Collection were searched, and data were extracted independently by two researchers. Outcomes included pooled PD-L1 protein positivity in tumor cells, immune cells, or both, per subtype and per antibody used, and its prognostic value for disease-free and overall survival. A pooled gene expression analysis of 39 publicly available transcriptomic datasets was also performed. Results: Of the initial 4,184 entries, 38 retrospective studies fulfilled the predefined inclusion criteria. The overall pooled PD-L1 protein positivity rate was 24% (95% CI, 15%-33%) in tumor cells and 33% (95% CI, 14%-56%) in immune cells. PD-L1 protein expression in tumor cells was prognostic for shorter overall survival (HR, 1.63; 95% CI, 1.07-2.46; P ¼ 0.02); there was significant heterogeneity (I 2 ¼ 80%, P heterogeneity < 0.001). In addition, higher PD-L1 gene expression predicted better survival in multivariate analysis in the entire population (HR, 0.82; 95% CI, 0.74-0.90; P < 0.001 for OS) and in basal-like tumors (HR, 0.64; 95% CI, 0.52-0.80; P < 0.001 for OS; P interaction 0.005). Conclusions: The largest to our knowledge meta-analysis on the subject informs on PD-L1 protein positivity rates and its prognostic value in breast cancer. Standardization is needed prior to routine implementation. PD-L1 gene expression is a promising prognostic factor, especially in basal-like breast cancer. Discrepant prognostic information might be related to PD-L1 gene expression in the stroma.