Changes in T lymphocyte subsets predict the efficacy of atezolizumab in advanced non-small cell lung cancer: a retrospective study

BACKGROUND: It has remained unclear how programmed cell death ligand 1 (PD-L1) inhibitors affect peripheral blood lymphocyte (PBL) subsets in patients with advanced non-small cell lung cancer (NSCLC). This study assessed the predictive and prognostic value of PBL subsets in patients with advanced NSCLC who were treated with atezolizumab. METHODS: A total of 30 patients with advanced NSCLC treated with atezolizumab were selected as the observation group, and 30 healthy individuals were chosen as the control group during same period. Flow cytometry was used to detect lymphocyte subsets before and after treatment. The relationship between the changes of lymphocyte subsets and atezolizumab in the treatment of NSCLC was analyzed and calculated. RESULTS: Before treatment, compared with the control group, the number of CD3(+), CD4(+) T, and CD4(+)/CD8(+) indexes in the observation group were significantly decreased, whereas the level of CD8(+) was significantly increased. The number of CD3(+), CD4(+) T, and CD4(+)/CD8(+) indexes gradually increased with the process of atezolizumab treatment, whereas the number of CD8(+ )T gradually decreased. After the 4 cycles, the number of CD3(+), CD4(+) T, and CD4(+)/CD8(+) indexes were significantly increased, and the number of CD8(+) was significantly decreased. In the observation group, 22 patients achieved partial response (PR)/stable disease (SD) and 8 patients achieved progressive disease (PD) after 4 cycles of atezolizumab treatment. Before treatment, there were no significant differences in the level of lymphocyte subsets between those who achieved PR/SD or PD. However, a significant difference in the level of lymphocyte subsets appeared after 4 cycles of atezolizumab treatment. Among the 22 patients who achieved PR/SD, the number of CD3(+), CD4(+) T, and CD4(+)/CD8(+) indexes were significantly increased, whereas the number of CD8(+) T lymphocytes was significantly decreased. Meanwhile, the 8 patients who achieved PD displayed different results. In addition, ROC curve combined detection of CD3(+), CD4(+), and CD8(+ )T [area under the curve (AUC) =0.9018, P<0.0001] showed good predictive ability for the efficacy of atezolizumab in advanced NSCLC. CONCLUSIONS: Atezolizumab may alter the level of lymphocyte subsets in patients with advanced NSCLC, and the changes in lymphocyte subsets may predict the efficacy of atezolizumab for these patients.