Background: Despite considerabletreatment advances,resistance to hormone therapy remains a major clinical problem. We correlated the protein levels of multiple tyrosine kinase receptors, signal transduction pathway proteins, and PD-L1 with relapse in HR-positive, HER2-negative breast cancer. Methods: A case-control discovery Reverse-Phase Protein Array (RPPA) analysis of primary tumors from 40 HR-positive, HER2-negative breast cancer patients: 18 patients who had relapsed; and 22 patients who had not relapsed (with a minimum of 2.5 years follow-up) was conducted to evaluate the potential of a 24 analyte panel of total and phosphoproteins to predict for probability of relapse. Analytes evaluated include: PD-L1, total and phosphoproteins for EGFR, HER2, HER3, and AR; and phosphoproteins in the major downstream signaling pathways: Ras/MAPK, AKT/mTOR, and JAK/STAT. Rank sum, ROC, and multianalyte statistical analysis was conducted to correlate the FFPE analyte data with clinical outcome (relapse or not). Results: Rank sum analysis of the 24 individual analytes revealed that only PD-L1 was significantly predictive of relapse. Relapsed patients had a significantly higher PD-L1 level (median 679, range 434 to 2702 NFU-normalized flourescence intensity per unit protein) compared to the patients who did not relapse (median of 511, range 112 to 1877 NFU) (p = 0.0098). ROC curve analysis yielded an AUC of 0.74 at a cut point of 577.53 NFU for PD-L1, and revealed a sensitivity of 72.2% and specificity of 81.82% for prediction of relapse. Further analysis of doublet or triplet analyte combinations (including PD-L1) did not add any additional ability to predict for relapse. Conclusions: Quantitative analysis of PD-L1 (of 24 analytes tested) was significantly predictive of relapse; higher PD-L1 predicted for a higher relapse rate. These results suggest that PD-L1 quantitation in larger hormone receptor-positive primary breast cancer patient cohorts may show promise to identify and select patients more likely to benefit from immune checkpoint inhibitors.