Exosomal STK39 is a Predictive Biomarker for Anti-VEGF Treatment Response in Neovascular Age-Related Macular Degeneration

Background: Age-related macular degeneration (AMD) is the leading cause of blindness worldwide, predominantly affecting individuals over 60. AMD is classified into two clinical forms: dry (DAMD) and neovascular (NAMD). NAMD, characterized by choroidal neovascularization and increased vascular endothelial growth factor (VEGF) levels, is treated with anti-VEGF therapy, the gold standard for preventing vision loss. However, challenges including non-response to treatment, persistent disease activity, and unnecessary anti-VEGF administration necessitate the identification of predictive biomarkers to optimize therapy. Exosomes (exo) are extracellular vesicles known for their stability and role in cellular communication, carrying diverse molecular contents, including RNA. Although the role of exo-RNA in AMD pathogenesis is well known, their role in the response to anti-VEGF treatment has not yet been investigated. Current study aim to investigate the role of exo-RNAs as potential biomarkers for predicting anti-VEGF treatment response in NAMD patients. Methods: Plasma-derived exo-RNA were analyzed pre- and post-treatment to identify predictive biomarkers from 41 NAMD patients treated with intravitreal bevacizumab. Exo were isolated using exoRNAeasy midi kit and characterized by TEM, NTA, and Western blot. Then, RNA-seq was performed to profile exo-RNAs and key candidates were validated using RT-qPCR. Results: Following anti-VEGF therapy, 68.3% of NAMD patients (mean age: 70.8 ± 6.14), demonstrated anatomical success, while 53.7% achieved functional success. Based on RNA-seq comparative analysis between the two groups revealed that translation mechanisms were significantly regulated, particularly within the context of upregulated pathways. Genes associated with molecular functions such as insulin response, transport, angiogenesis, and oxygen transport were prominently implicated. Conversely, the evaluation of downregulated pathways indicated substantial alterations in RNA catabolic processes and the regulation of gene expression. RNA-seq revealed BMP6 and STK39 as key candidate genes, with post-treatment dysregulation in hsa-miR-4692 expression, a microRNA targeting STK39. Finally, RT-qPCR validation show that elevated pre-treatment exo-STK39 levels were significantly associated with anatomical success (*p=0.04). RNA-seq analysis was performed to evaluate the effect of anti-VEGF treatment on exo-RNA regulation in NAMD patients. For RNA-seq analysis, a cohort of 5 anti-VEGF(-) and 4 anti-VEGF(+) patients were selected to evaluate the treatment response at the exo-mRNA and exo-miRNA levels. Following treatment, one patient was excluded from the analysis due to suboptimal read quality, resulting in a final analysis of 5 anti-VEGF(-) and 4 anti-VEGF(+) patients. For the miRNA analysis, 3 anti-VEGF (-) and 3 anti-VEGF(+) patients were selected, based on the quality of small RNA sequencing data, ensuring comparable read quality across both groups. The most optimal exo- RNA candidate at the mRNA level was identified by analyzing the pathways, interactions, and external regulations associated with the obtained miRNA data. This approach allowed for the determination of key biological processes and regulatory networks in which the miRNA- derived exo-RNAs were involved, thereby facilitating the identification of the most relevant candidate for further investigation. Total RNA isolation was performed using the QIAGEN exoRNeasy Midi kit (Qiagen, Hilden, Germany). RNA quality and concentrations were analyzed using Agilent Technologies 4200 TapeStation (Agilent Technologies, Santa Clara, California, USA). To perform RNA sequencing, mRNA and small RNA libraries were generated using the Illumina NextSeq 500 platform with the SMARTer smRNA-Seq Kit (Illumina, Takara Bio, Shiga, Japan) based on the manufacturer’s instructions. The libraries were gel-purified using the D1000 ScreenTape System (Agilent Technologies, Waldbronn, Germany). Then, RPKM-normalized count were used for analysis to identify differentially expressed genes and mature miRNAs between anti- VEGF (-) vs anti-VEGF (+). A p-value correction was performed using the false discovery rate (FDR) correction determined for each gene or miRNA. The cut- off value for the adjusted p- value was accepted at 0.05, Dysregulated genes/miRNAs were determined by fold changes.

背景：年龄相关性黄斑变性（age-related macular degeneration, AMD）是全球首要致盲病因，主要累及60岁以上人群。AMD分为两种临床类型：干性（dry, DAMD）与新生血管性（neovascular, NAMD）。其中新生血管性AMD以脉络膜新生血管形成及血管内皮生长因子（vascular endothelial growth factor, VEGF）水平升高为特征，治疗采用抗VEGF疗法，该方案是预防视力丧失的金标准。然而，当前仍面临治疗无应答、疾病持续活动及不必要的抗VEGF给药等挑战，亟需识别预测性生物标志物以优化治疗方案。外泌体（exosomes, exo）是一类以稳定性及细胞间通讯功能为核心特征的细胞外囊泡，携带有包括RNA在内的多种分子内容物。尽管exo-RNA在AMD发病机制中的作用已得到明确阐明，但其对抗VEGF治疗应答的相关作用尚未得到研究。本研究旨在探讨exo-RNA作为预测新生血管性AMD患者抗VEGF治疗应答的潜在生物标志物的应用价值。
方法：本研究纳入41例接受玻璃体腔内贝伐珠单抗治疗的新生血管性AMD患者，对治疗前后的血浆来源外泌体RNA进行分析，以筛选预测性生物标志物。采用exoRNAeasy midi试剂盒分离外泌体，并通过透射电子显微镜（TEM）、纳米颗粒跟踪分析（NTA）及蛋白质印迹（Western blot）对其进行表征。随后开展RNA测序（RNA-seq）以分析外泌体RNA的表达谱，并通过实时定量聚合酶链反应（RT-qPCR）对关键候选分子进行验证。
结果：接受抗VEGF治疗后，68.3%的新生血管性AMD患者（平均年龄：70.8±6.14岁）达到解剖学成功标准，53.7%的患者实现功能学成功。通过两组间的RNA测序比较分析发现，翻译机制受到显著调控，尤以通路上调最为显著。与胰岛素应答、物质转运、血管生成及氧转运等分子功能相关的基因显著富集。反之，下调通路的分析则显示RNA分解代谢过程及基因表达调控发生显著改变。RNA测序筛选出BMP6与STK39作为关键候选基因，且治疗后靶向STK39的microRNA hsa-miR-4692的表达出现异常调控。最终，RT-qPCR验证结果显示，治疗前升高的外泌体STK39水平与解剖学成功显著相关（*p=0.04）。
本研究同时开展RNA测序分析，以评估抗VEGF治疗对新生血管性AMD患者外泌体RNA调控的影响。在RNA测序分析中，选取5例抗VEGF治疗无应答组（anti-VEGF(-)）与4例抗VEGF治疗应答组（anti-VEGF(+)）患者，以在外泌体mRNA及外泌体miRNA水平评估治疗应答情况。治疗后，1例患者因测序读长质量不佳被排除出分析，最终纳入分析的样本为5例anti-VEGF(-)与4例anti-VEGF(+)患者。
针对miRNA分析，根据小RNA测序数据的质量，选取3例anti-VEGF(-)与3例anti-VEGF(+)患者，以确保两组间测序读长质量具有可比性。通过分析所获得的miRNA数据相关的通路、相互作用及外部调控机制，筛选出mRNA水平上最优的外泌体RNA候选分子。该方法可明确miRNA来源的外泌体RNA所参与的关键生物学过程及调控网络，从而助力筛选出最具研究价值的候选分子以供后续验证。
总RNA提取采用QIAGEN exoRNeasy Midi试剂盒（德国希尔德市凯杰公司）。RNA的质量与浓度使用Agilent Technologies 4200 TapeStation系统（美国加利福尼亚州圣克拉拉市安捷伦科技公司）进行检测。为开展RNA测序，基于制造商说明书，使用SMARTer smRNA-Seq试剂盒（日本滋贺县宝生物工程株式会社，Illumina）结合Illumina NextSeq 500平台构建mRNA与小RNA文库。文库通过D1000 ScreenTape系统（德国瓦尔德布龙市安捷伦科技公司）进行凝胶纯化。随后，采用每百万reads每千碱基片段的转录本数（RPKM）标准化后的计数数据，用于分析anti-VEGF(-)与anti-VEGF(+)组间差异表达的基因及成熟miRNA。通过错误发现率（false discovery rate, FDR）校正对每个基因或miRNA的p值进行校正，校正后p值的截断值设定为0.05，差异表达基因/miRNA的筛选依据为倍数变化。