PO.CL01.11 · 临床研究

血浆表观基因组图谱分析捕获B细胞淋巴瘤生物学特征

Plasma epigenomic profiling captures B-cell lymphoma biology

海报缩略图:血浆表观基因组图谱分析捕获B细胞淋巴瘤生物学特征
编号 7832 展板 13 时间 4/22 09:00–12:00 区域 Section 45 主讲 Aleem Aamir, BS
分会场 Liquid Biopsies: Circulating Nucleic Acids 5
查看 PDF 下载 PDF 🔒 查看 / 下载完整 PDF 需登录并开通下载套餐 · 查看套餐 / 开通 AACR 官方页面

作者与单位 Authors & Affiliations

Aleem Aamir1, Steven De Michino1, Michael Hong2, Ting Liu2, Victoria Shelton2, Benjamin Haibe-Kains1, Michael M. Hoffman1, Mathieu Lupien1, Robert Kridel2, Benjamin Lok1, Scott V. Bratman1

1Medical Biophysics, University of Toronto, Toronto, ON, Canada,2Princess Margaret Cancer Center, Toronto, ON, Canada

摘要 Abstract

中文摘要
背景:B细胞淋巴瘤(BCLs),包括弥漫大B细胞淋巴瘤(DLBCL)和滤泡性淋巴瘤(FL),表现出广泛的表观遗传失调。组蛋白甲基转移酶EZH2催化抑制性标记H3K27me3,是B细胞发育的重要表观遗传调控因子。尽管既往工作已在BCL肿瘤样本中研究了H3K27me3,但BCL中H3K27me3修饰的无细胞染色质图景在很大程度上仍未被探索。 方法:我们对来自Princess Margaret LIBERATE研究的n=55例DLBCL、n=19例FL和n=21例非癌对照进行了靶向H3K27me3的cfChIP-seq。对于全局分析,在10kb基因组窗口中计数读长,使用截断泊松MLE减除背景读长,并进行文库大小归一化。BCL评分使用针对健康参考panel和高负荷淋巴瘤原型的留一法NNLS计算。对于启动子分析,计数TSS ±2kb内的读长并进行文库大小归一化。 结果:晚期BCL的全局血浆H3K27me3谱与局限期BCL和非淋巴瘤样本分离。晚期BCL(p<0.0001)和局限期BCL(p<0.05)的平均BCL评分较非淋巴瘤对照更高。此外,DLBCL(p<0.0001)和FL(p<0.001)的平均BCL评分均高于非淋巴瘤对照。BCL评分与PC1高度相关(ρ=0.902,p=2.97e-35)以及与正交肿瘤分数指标高度相关,包括靶向测序的平均VAF(ρ=0.850,p=4.79e-18)、约0.5X WGS的ichorCNA肿瘤分数(ρ=0.779,p=3.86e-12)和FDG-PET/CT的总病灶糖酵解(ρ=0.669,p=1.37e-10)。启动子处H3K27me3信号与平均VAF之间的相关性分析揭示,在PAX5(r=-0.705,p=1.29e-09)和DTX1(r=-0.622,p=3.11e-07)等B细胞谱系位点呈负相关,而在MCEMP1(r=0.746,p=4.1e-11)和IL6R(r=0.635,p=1.46e-07)等其他造血谱系位点呈正相关,表明随着肿瘤分数增加,B细胞谱系的贡献增加。 结论:这些数据表明,血浆H3K27me3谱可捕获疾病负荷并反映BCL的潜在肿瘤生物学。此外,这些谱可捕获谱系特异性的启动子抑制模式,表现为H3K27me3信号与B细胞谱系标志物和肿瘤分数之间的负相关,与在其他造血谱系标志物处观察到的正相关形成对比。总体而言,H3K27me3的cfChIP-seq提供了BCL中循环无细胞染色质的无创表观基因组读出。
查看英文原文 English abstract
Background: B-cell lymphomas (BCLs), including diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL), display widespread epigenetic dysregulation. The histone methyltransferase EZH2 catalyzes the repressive mark H3K27me3 and is an important epigenetic regulator for B-cell development. Although previous work has examined H3K27me3 in BCL tumour samples, the landscape of H3K27me3-decorated cell-free chromatin in BCL remains largely unexplored. Methods: We performed cfChIP-seq targeting H3K27me3 on n=55 DLBCL, n=19 FL, and n=21 non-cancer controls from the Princess Margaret LIBERATE study. For global analyses, reads were counted in 10kb genomic windows and background reads were subtracted using a truncated Poisson MLE and data were library size normalized. BCL scores were computed using leave-one-out NNLS against a healthy reference panel and a high-burden lymphoma archetype. For promoter analyses, reads ±2kb of the TSS were counted and library size normalized. Results: Global plasma H3K27me3 profiles in advanced-stage BCL separate from limited-stage BCL and non-lymphoma samples. Mean BCL scores were higher in advanced-stage BCL(p<0.0001) and limited-stage BCL (p<0.05) compared to non-lymphoma controls. Moreover, both DLBCL (p<0.0001) and FL (p<0.001) had higher mean BCL scores than non-lymphoma controls. BCL scores were highly correlated to PC1 (ρ=0.902, p=2.97e-35) and the orthogonal tumour fraction metrics including mean VAF from targeted sequencing (ρ=0.850, p=4.79e-18), ichorCNA tumour fraction from ~0.5X WGS (ρ=0.779, p=3.86e-12), and total lesion glycolysis from FDG-PET/CT (ρ=0.669, p=1.37e-10). Correlation analysis between H3K27me3 signal at promoters and mean VAF revealed a negative association at B-cell lineage loci such as PAX5 (r=-0.705, p=1.29e-09) and DTX1 (r=-0.622, p=3.11e-07), and a positive association at other hematopoietic lineage loci such as MCEMP1 (r=0.746, p=4.1e-11) and IL6R (r=0.635, p=1.46e-07), indicating an increased contribution of B-cell lineage contribution with increasing tumour fraction. Conclusions: These data show that plasma H3K27me3 profiles capture disease burden and reflect underlying tumour biology in BCL. Furthermore, these profiles capture lineage-specific promoter repression patterns, manifesting in negative correlations between H3K27me3 signal B-cell lineage markers and tumour fraction, contrasting positive correlations observed at markers of other hematopoietic lineages. Overall, cfChIP-seq of H3K27me3 provides a non-invasive epigenomic readout of circulating cell-free chromatin in BCL.
利益披露 Disclosure
A. Aamir, None. S. De Michino, Roche Employment. M. Hong, None.. T. Liu, None.. V. Shelton, None.. B. Haibe-Kains, None. M. M. Hoffman, Adela Patent. M. Lupien, None. R. Kridel, Roche ). Abbvie ). B. Lok, Pfizer ). Astrazeneca Other, Personal fees, non financial support. Daiichi-Sankyo Personal fees. S. V. Bratman, Roche Molecular Diagnostics Patent. Adela g., Board of Directors, non-salaried role), Patent.

← 返回 AACR 2026 检索