Introduction: Only a subset of patients have a clinical response to cancer immunotherapy. The molecular mechanisms that mediate the immunological response or tolerance to treatment are just beginning to be understood. Several tumor signaling pathways are involved in those mechanisms: Wnt/β-catenin, STAT3, NFkB, PI3K/PTEN/AKT, and TP53. To characterize the tumor, its immune infiltrate, and its microenvironment, we examined expression of several biomarkers and determined the relationships among them in various breast cancer subtypes.
Methods: 148 breast cancer (BCa) cases from the University of Chicago Breast Cancer SPORE tissue bank under IRB approved protocols were classified into four biological subtypes based on immunohistochemical (IHC) staining: luminal A, luminal B, Her2 and basal-like. IHC staining for PD-L1, CD8, FoxP3, CD68, CD163, and β-catenin was performed on tissue microarrays. CD8+ and FoxP3+ cells were counted manually and recalculated for 1 mm2. Macrophage phenotype was determined using single and double staining with CD68 for total population and CD68/CD163 for M2. CD68+ and CD163+ cells were counted manually, confirmed by Image Analysis, and recalculated for 1 mm2. PD-L1 and β-catenin (membrane-associated, cytosolic and nuclear) were scored as negative, weak positive, moderate, and strong.
Results: The study cohort subtypes were 20.9% basal-like, 10.4% Her2, 58.3% luminal A, and 10.4% luminal B. The ratio of M2 to M1 cells increased with disease progression (p<0.001). Her2 and basal-like subtypes had a significantly higher percentage of M2 cells as compared to the luminal A subtype (p<0.001). The cases with M2 macrophage phenotype had a higher ratio of FoxP3 regulatory T cells (Tregs) to cytotoxic CD8+ cells. PD-L1 expression was present in 23% of specimens. However, the intensity of expression was markedly different across the BCa subtypes. PD-L1 high expressing tumors had, on average, greater FoxP3+ cell infiltrate and low or absent CD8+ T cells. The ratio of FoxP3 Tregs to CD8+ cells in basal-like tumors were 1.7, 2.5, and 4 -fold higher than in Her2, luminal A, and luminal B tumors, respectively (p<0.05 for all correlations). There was a statistically significant decrease in membrane-associated β-catenin in tumor compared to normal tissue (p<0.0001). The proportion of tumors positive for cytosolic or nuclear β-catenin in basal-like tumors was statistically higher than all other tumor subtypes. High β-catenin nuclear protein expression was predominantly observed in tumors that lacked CD8+ T cells. Ongoing analysis will evaluate the association of these biomarkers with clinical outcomes.
Conclusions and Future Directions: Our data suggests that IHC using a panel of antibodies may be a robust and suitable method for evaluating level of immune infiltration. Future studies will evaluate multiplex immunofluorescence of multiple biomarkers in breast tissue.
Citation Format: April Swoboda, Galina Khramtsova, Lise Sveen, Andrey Khramtsov, Rita Nanda, Olufunmilayo Olopade. Biomarkers of immune infiltration for multiplex immunofluorescence in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2950. doi:10.1158/1538-7445.AM2017-2950