Fig. 3: irAE-related changes in T cell clonotypes.

a Boxplots showing TCR diversity in CD4⁺ T cells, measured by the Shannon indexes. In all boxplots, the center line represents the median, the bounds of the box indicate the 25th and 75th percentiles, and the whiskers extend to 1.5 × IQR. Each dot represents an individual patient. P values were determined by using the two-sided Wilcoxon rank-sum test and FDR values were calculated by the Benjamini–Hochberg method. b Overlap of the CD4⁺ T cell TCR repertoire across samples collected at different time points for patients with and without irAEs. A TCR clonotype was defined as shared when it was detected and shared in T cells from at least two samples from any patient. The heat map shows the percentage of TCR clonotypes from samples at one time point that overlapped with those from another time point. The primary samples are plotted on the x-axis, and the secondary samples are plotted on the y-axis. Statistical comparisons were performed between irAE and non-irAE patients for the proportion of overlapping TCR clonotypes identified between specific sample pairs. For clonotype overlaps between pre-imm PBMC and on-imm PBMC, P = 0.014 and the FDR = 0.23; for overlaps between on-imm PBMC and on-radio tumor, P = 0.046 and FDR = 0.23; for overlaps between pre-imm PBMC and pre-radio tumor, P = 0.034 and FDR = 0.23; and for overlaps between on-imm PBMC and pre-radio tumor, P = 0.034 and FDR = 0.23. P values were determined by using the two-sided Wilcoxon rank-sum test, and FDR values were calculated by the Benjamini–Hochberg method. Sample sizes for a-b: pre-radio tumor (irAEs absent: n = 6; irAEs present: n = 10); on-radio tumor (irAEs absent: n = 5; irAEs present: n = 10); pre-radio PBMC (irAEs absent: n = 5; irAEs present: n = 4); pre-imm and on-imm PBMC (irAE absent: n = 3; irAE present: n = 8). c Analysis of the expanded TCR clonotypes shared across clusters in patients with and without irAEs. For each sample and subcluster, the number of expanded TCR clonotypes from the CD8⁺ T cells was calculated and summed up for the indicated irAE groups. A comparison was made between blood samples collected obtained at the pre-imm and on-imm time points. P values were determined by using the two-sided Wilcoxon rank-sum test and FDR values were calculated by the Benjamini–Hochberg method. The expanded clonotypes shared with an irAE-associated cluster (CD8.Temra) are shown in red, while those shared with other CD8⁺ T cell subclusters are shown in blue. Sample sizes for c: pre-radio tumor (irAEs absent: n = 6; irAEs present: n = 10); on-radio tumor (irAEs absent: n = 5; irAEs present: n = 10); pre-radio PBMC (irAEs absent: n = 5; irAEs present: n = 3); pre-imm and on-imm PBMC (irAE absent: n = 3; irAE present: n = 8). d Spearman correlation between HPV abundance (determined by count per million reads) and the percentage of CD8⁺ T cell subcluster. The red and blue lollipops indicate positive and negative correlations, respectively. The shade of the dot represents the correlation coefficient (Rs). The size of each dot corresponds to the significance level, with the large, medium, and small dots representing FDR<0.01, 0.01–0.1, and ≥ 0.1, respectively. The sample size for the correlation analysis is 51. e, Samples with TCR clonotypes recognizing the HPV epitope. An assessment of the publicly available databases revealed 13 samples with HPV-specific TCRs. All sample sizes reported are biological replicates, with the patient serving as the independent unit of study. TCR T cell receptor, HPV human papillomavirus. Source data are provided as a Source Data file.