Abstract
BACKGROUND: Programmed cell death protein-1/programmed cell death ligand-1 (PD-1/PD-L1) inhibitors works by reactivating immune cells. Considering the accessibility of noninvasive liquid biopsies, it is advisable to employ peripheral blood lymphocyte subsets to predict immunotherapy outcomes. METHODS: We retrospectively enrolled 87 patients with available baseline circulating lymphocyte subset data who received first-line PD-1/PD-L1 inhibitors at Peking Union Medical College Hospital between May 2018 and April 2022. Immune cell counts were determined by flow cytometry. RESULTS: Patients who responded to PD-1/PD-L1 inhibitors had significantly higher circulating CD8+CD28+ T-cell counts (median [range] count: 236 [30-536] versus 138 [36-460]/μL, p < 0.001). Using 190/μL as the cutoff value, the sensitivity and specificity of CD8+CD28+ T cells for predicting immunotherapy response were 0.689 and 0.714, respectively. Furthermore, the median progression-free survival (PFS, not reached versus 8.7 months, p < 0.001) and overall survival (OS, not reached versus 16.2 months, p < 0.001) were significantly longer in the patients with higher CD8+CD28+ T-cell counts. However, the CD8+CD28+ T-cell level was also associated with the incidence of grade 3-4 immune-related adverse events (irAEs). The sensitivity and specificity of CD8+CD28+ T cells for predicting irAEs of grade 3-4 were 0.846 and 0.667, respectively, at the threshold of CD8+CD28+ T cells ≥ 309/μL. CONCLUSIONS: High circulating CD8+CD28+ T-cell levels is a potential biomarker for immunotherapy response and better prognosis, while excessive CD8+CD28+ T cells (≥ 309/μL) may also indicate the emergence of severe irAEs.