PO.TB10.04 · 肿瘤生物学
纵向肿瘤微环境分析以预测涎腺导管癌对免疫化疗的反应
Longitudinal tumor microenvironment analysis to predict immunochemotherapy response in salivary duct carcinoma
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景：
涎腺导管癌（SDC）是一种罕见但高度侵袭性的恶性肿瘤。尽管新辅助细胞毒化疗后手术是局部晚期高级别SDC的标准治疗，但近期研究提示免疫检查点抑制剂（ICI）在围手术期环境中具有潜在获益。为了表征治疗相关的肿瘤微环境（TME）特征，我们分析了来自围手术期临床试验（ONO-4538-X78）的配对免疫化疗前后样本。
方法：
使用全转录组测序（WTS）分析了来自30例SDC患者的共42份肿瘤样本。样本在诊断时（T1）和三个周期免疫化疗后手术时（T2）采集。残留存活肿瘤<10%的患者被归类为主要病理缓解者（MPR）。样本被分类为T1非MPR（n=7）、T1 MPR（n=15）、T2非MPR（n=11）和T2 MPR（n=9），并跨时间点和反应组比较TME。
结果：
为了表征TME差异，我们推断了每个反应组的细胞状态。在基线（T1），非MPR肿瘤表现出由恶性上皮细胞驱动的促血管生成TME，与EMT和血管生成通路一致。相比之下，T1 MPR肿瘤显示出具有干扰素相关信号的免疫浸润性更强的TME。
从手术样本（T2）来看，MPR和非MPR的直接比较揭示了不同的抗肿瘤免疫模式，表明MPR中存在强劲的T细胞和B细胞介导的适应性免疫，而非MPR则显示慢性炎症特征而非有效的抗肿瘤反应。使用来自T细胞和B细胞相关GOBP通路的核心富集基因进行的ssGSEA证实，T2 MPR组的免疫激活显著高于T2非MPR组（p=0.001）。
纵向比较进一步突出了不同的轨迹。在非MPR肿瘤中，推断T2非MPR肿瘤保留了恶性上皮特征且免疫细胞极少；DEG分析支持这一点，表现为T2的炎症反应和T1的EMT程序。相反，推断T2 MPR肿瘤丧失了恶性上皮特征并显示免疫激活，与T细胞介导的适应性反应通路一致，而T1 MPR则显示上皮细胞周期通路。
结论：
在本研究中，我们发现MPR肿瘤在基线时表现出预先存在的干扰素激活的肿瘤微环境，在免疫化疗后转变为适应性免疫反应。这些发现提示，干扰素预激活的TME是有效的治疗后免疫激活的前提，并可能作为免疫化疗反应性的预测生物标志物。
查看英文原文 English abstract
Background:
Salivary duct carcinoma (SDC) is a rare but highly aggressive malignancy. Although neoadjuvant cytotoxic chemotherapy followed by surgery is standard for locally advanced high-grade SDC, recent studies suggest potential benefit from immune checkpoint inhibitors (ICIs) in the perioperative setting. To characterize treatment-related tumor microenvironment (TME) features, we analyzed paired pre- and post-immunochemotherapy samples from a perioperative clinical trial (ONO-4538-X78).
Method:
A total of 42 tumor samples from 30 SDC patients were analyzed using whole-transcriptome sequencing (WTS). Samples were collected at diagnosis (T1) and at surgery (T2) after three cycles of immunochemotherapy. Patients with <10% residual viable tumor were classified as major pathologic responders (MPR). Samples were classified into T1 non-MPR (n=7), T1 MPR (n=15), T2 non-MPR (n=11), and T2 MPR (n=9), and TMEs were compared across time points and response groups.
Result:
To characterize TME differences, we inferred cell states for each response group. At baseline (T1), non-MPR tumors exhibited a pro-angiogenic TME driven by malignant epithelial cells, consistent with EMT and angiogenesis pathways. In contrast, T1 MPR tumors showed a more immune-infiltrated TME with interferon-related signaling.
From the surgical samples (T2), direct comparison of MPR and non-MPR revealed distinct antitumor immune patterns, indicating robust T- and B-cell-mediated adaptive immunity in MPR, whereas non-MPR displayed chronic inflammatory signatures rather than effective antitumor responses. ssGSEA using core enrichment genes from T- and B-cell-related GOBP pathways confirmed significantly higher immune activation in the T2 MPR group compared with T2 non-MPR (p=0.001).
Longitudinal comparison further highlighted divergent trajectories. In non-MPR tumors, T2 non-MPR tumors were inferred to retain malignant epithelial features with minimal immune cells; DEG analysis supported this with inflammatory responses at T2 and EMT programs at T1.Conversely, T2 MPR tumors were inferred to lose malignant epithelial signatures and show immune activation, consistent with T cell-mediated adaptive responses pathways, whereas T1 MPR displayed epithelial cell-cycle pathways.
Conclusion:
In this study, we found that MPR tumors exhibited a pre-existing interferon-activated tumor microenvironment at baseline, which transitioned into an adaptive immune response after immunochemotherapy. These findings suggest that an interferon-primed TME is a prerequisite for effective post-treatment immune activation and may serve as a predictive biomarker for immunochemotherapy responsiveness.
利益披露 Disclosure
B. Choi, None..
H. Kim, None..
D. Oh, None..
M. Ahn, None..
H. Jung, None..
J. Cho, None..
H. Jeong, None..
S. Lee, None..
K. Yang, None..
N. Choi, None..
E. Kim, None..
S. Park, None.