Clinicopathological and prognostic significance of programmed cell death ligand 1 expression in patients diagnosed with breast cancer: meta-analysis.

2021 
BACKGROUND Uncertainty exists regarding the clinical relevance of programmed cell death ligand 1 (PD-L1) expression in breast cancer. This systematic review and meta-analysis evaluated the clinicopathological and prognostic role of PD-L1 in breast cancer. METHODS A systematic review was performed in accordance with PRISMA guidelines. Observational studies that compared high versus low expression of PD-L1 on breast cancer cells were identified. Log hazard ratios (HRs) for disease-free and overall survival and their standard errors were calculated from Kaplan-Meier curves or Cox regression analyses, and pooled using the inverse-variance method. Dichotomous variables were pooled as odds ratios (ORs) using the Mantel-Haenszel method. RESULTS Sixty-five studies with 19 870 patients were included; 14 404 patients were classified as having low and 4975 high PD-L1 expression. High PD-L1 was associated with achieving a pathological complete response following neoadjuvant chemotherapy (OR 3.30, 95 per cent confidence interval 1.19 to 9.16; P < 0.01; I2 = 85 per cent). Low PD-L1 expression was associated with human epidermal growth factor receptor 2 (OR 3.98, 1.81 to 8.75; P < 0.001; I2 = 96 per cent) and luminal (OR 14.93, 6.46 to 34.51; P < 0.001; I2 = 99 per cent) breast cancer subtypes. Those with low PD-L1 had favourable overall survival rates (HR 1.30, 1.05 to 1.61; P = 0.02; I2 = 85 per cent). CONCLUSION Breast cancers with high PD-L1 expression are associated with aggressive clinicopathological and immunohistochemical characteristics. While they more likely to achieve a pathological complete response after neoadjuvant chemotherapy, they are, however, associated with worse overall survival outcomes. Immunomodulation may hold promise as a therapeutic strategy in treatment resistant breast cancer subtypes.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    103
    References
    5
    Citations
    NaN
    KQI
    []