Supplementary Material for: Dynamic peripheral T-cell analysis identifies on-treatment prognostic biomarkers of atezolizumab plus bevacizumab in hepatocellular carcinoma

Introduction: Variability in response to atezolizumab plus bevacizumab (AB) treatment of hepatocellular carcinoma (HCC) underscores the critical need for the development of effective biomarkers. We sought to identify peripheral blood biomarkers reflecting response to AB treatment. Methods: We analyzed dynamic changes in peripheral blood mononuclear cells from a prospective, multicenter cohort of 65 patients with HCC, using flow cytometry to evaluate the T-cell population before and 3 weeks after the first AB treatment. Results: We found a unique response of the CD8+ T cells in terms of both frequency and phenotype, in contrast to CD4+ T cells and regulatory T cells. Notably, CD8+ T cells showed significant changes in expression of Ki-67 and T-cell immunoreceptors with Ig and ITIM domains (TIGIT). These distinct responses were observed particularly in the programmed cell death receptor-1 (PD-1)+ subpopulation of CD8+ T cells. Interestingly, the baseline differentiation status of PD-1+CD8+ T cells, particularly the central memory T-cell subset, correlated positively with greater proliferation (higher Ki-67 expression) of PD-1+CD8+ T cells after treatment. Moreover, effector memory cells expressing CD45RA correlated negatively with the increase in TIGIT+/PD-1+CD8+ T cells. The increase in TIGIT+/CD8+ T cells was associated with the development of immune-related adverse events, whereas increase in Ki-67+/PD-1+CD8+ T cells was associated with the better objective response rate. Importantly, dynamic shifts of Ki-67+/PD-1+CD8+ T cells and TIGIT+/CD8+ T cells significantly predicted progression-free survival and overall survival, as confirmed by multivariate analysis. Conclusion: These findings highlight the potential of dynamic changes in CD8+ T cells as an on-treatment prognostic biomarker. Our study underscores the value of peripheral blood profiling as a noninvasive and practical method for predicting the clinical outcomes of AB treatment in patients with HCC.

引言：阿替利珠单抗联合贝伐珠单抗（atezolizumab plus bevacizumab, AB）治疗肝细胞癌（hepatocellular carcinoma, HCC）的应答存在异质性，这凸显了开发有效生物标志物的迫切需求。本研究旨在筛选可反映AB治疗应答的外周血生物标志物。
方法：本研究针对65例肝细胞癌患者组成的前瞻性多中心队列，采集其外周血单个核细胞，采用流式细胞术检测首次AB治疗前及治疗3周后的T细胞群体动态变化。
结果：相较于CD4阳性T细胞与调节性T细胞，CD8阳性T细胞在细胞频率与表型层面均呈现出独特的应答特征。值得注意的是，CD8阳性T细胞的Ki-67及T细胞免疫球蛋白和ITIM结构域蛋白（T-cell immunoreceptors with Ig and ITIM domains, TIGIT）的表达水平发生了显著改变。上述特异性应答尤其体现在程序性死亡受体1（programmed cell death receptor-1, PD-1）阳性的CD8阳性T细胞亚群中。
有趣的是，PD-1阳性CD8阳性T细胞的基线分化状态（尤其是中枢记忆T细胞亚群）与治疗后PD-1阳性CD8阳性T细胞的增殖活性（Ki-67高表达）呈正相关。此外，表达CD45RA的效应记忆T细胞与TIGIT阳性PD-1阳性CD8阳性T细胞的扩增呈负相关。TIGIT阳性CD8阳性T细胞的扩增与免疫相关不良事件的发生相关，而Ki-67阳性PD-1阳性CD8阳性T细胞的扩增则与更高的客观缓解率相关。
尤为重要的是，经多因素分析证实，Ki-67阳性PD-1阳性CD8阳性T细胞与TIGIT阳性CD8阳性T细胞的动态变化可显著预测患者的无进展生存期与总生存期。
结论：本研究结果表明，CD8阳性T细胞的动态变化有望成为治疗过程中的预后生物标志物。本研究同时凸显了外周血谱分析作为一种无创且实用的方法，可用于预测肝细胞癌患者AB治疗的临床结局。