RESUMEN
Immune checkpoint inhibitors (ICI) have been applied in treating advanced hepatocellular carcinoma (aHCC) patients, but few patients exhibit stable and lasting responses. Moreover, identifying aHCC patients suitable for ICI treatment is still challenged. This study aimed to evaluate whether dissecting peripheral immune cell subsets by Mann-Whitney U test and artificial intelligence (AI) algorithms could serve as predictive biomarkers of nivolumab treatment for aHCC. Disease control group carried significantly increased percentages of PD-L1+ monocytes, PD-L1+ CD8 T cells, PD-L1+ CD8 NKT cells, and decreased percentages of PD-L1+ CD8 NKT cells via Mann-Whitney U test. By recursive feature elimination method, five featured subsets (CD4 NKTreg, PD-1+ CD8 T cells, PD-1+ CD8 NKT cells, PD-L1+ CD8 T cells and PD-L1+ monocytes) were selected for AI training. The featured subsets were highly overlapping with ones identified via Mann-Whitney U test. Trained AI algorithms committed valuable AUC from 0.8417 to 0.875 to significantly separate disease control group from disease progression group, and SHAP value ranking also revealed PD-L1+ monocytes and PD-L1+ CD8 T cells exclusively and significantly contributed to this discrimination. In summary, the current study demonstrated that integrally analyzing immune cell profiling with AI algorithms could serve as predictive biomarkers of ICI treatment.
RESUMEN
OBJECTIVE: Diagnosis of SLE is based on clinical manifestations but is heterogeneous in early onset. Hence, we aimed to evaluate the feature of the immunoprofiling in patients with SLE and apply it to develop an immune signature algorithm for supporting SLE diagnosis. METHODS: We enrolled 13 newly diagnosed patients with SLE and 9 healthy controls (HCs) followed by analysing their immunoprofilings within their peripheral blood mononuclear cells (PBMCs) through flow cytometry. The immunoprofiling from the patients with SLE and HCs were ranked and formed an immune signature score. Besides, we enrolled four patients with SLE and monitored the changes in their immunoprofilings after immunosuppressant treatment. RESULTS: Among 93 immune cell subsets, 29 differed significantly between patients with SLE and HCs, and lower dendritic and natural killer cell percentages and a higher CD8+ T-cell percentage were identified in patients with SLE. In an investigation of immune-tolerant-related cell subsets, higher concentrations of CD8+ regulatory natural killer T cells, programmed cell death 1 (PD-1)+ T cells, and lower concentrations of programmed cell death ligand 1 (PD-L1)+ PBMCs were observed in the SLE group. The immune signature score from patients with SLE was significantly different from that from the HCs. After treatment, the disease activity of the four patients were tended to stable and percentages of PD-L1+ monocytes, PD-1+ CD4 T and CD8 T cells in patients with SLE exhibited positively and negatively correlation with the SLEDAI-2K (Systemic Lupus Erythematosus Disease Activity Index 2000) score, which might associate with the remission of SLE. CONCLUSIONS: The comparison of immunprofiling between patients with SLE and HCs exhibited a distinct pattern. This difference and its application to immune signature algorithm shed light on the studies of SLE pathogenesis and immune-based diagnostic tool development in the future.
