Fig. 1: The enrichment of CD8⁺ and CCT subsets predict the efficacy of anti-PD-1 therapy.

a The frequency of CD8⁺ T cells from responders’ (n = 11) PBMC were lower than that of non-responders (n = 14) (baseline). This box plots represent the interquartile range (IQR), with the horizontal line indicating the median. The whiskers extend to the farthest data point within a maximum of 1.5 × IQR. b Much lower frequencies of CCT subtype were observed in responders’ PBMC (n = 11) as well (baseline) compared to non-responders’ (n = 14). This box plots represent the interquartile range (IQR), with the horizontal line indicating the median. The whiskers extend to the farthest data point within a maximum of 1.5 × IQR. c By longitudinal analysis, both CD8⁺ T cells and CCT T cells remained at low levels in most responders during the 10 treatment cycles. Data are presented as mean values +/− SD. d High frequency of CD8⁺ T cells and CCT T cells was correlated with bad prognosis after the treatment of anti-PD1 therapy. e High level of total CD8⁺ T cell exhibited worse median PFS. The percentages for each subpopulation are based on overall live CD45⁺ singlet cell counts. p-value in PFS analysis was examined by the log-rank test. Other p-value was calculated using two-sided t-tests and were corrected for multiple comparisons using the Benjamini-Hochberg adjustment. R responder, NR non-responder, CR complete response, PR partial response, SD stable disease, PD progressive disease. Two baseline samples (J015A and J024A) and 4 samples after treatment (J013B, J029B, and J024 B & C) as they failed the CyTOF QC test were excluded.