Braga Letícia da Conceição, Amaral Laurence Rodrigues do, Villar Delfino Pedro Henrique, Andrade Nara Rosana, de Oliveira Salles Paulo Guilherme, da Silva Filho Agnaldo Lopes, Bertarini Pedro Luiz Lima, da Silva Ramos Ana Paula Álvares, Gomes Matheus de Souza, Silva Lopes Luciana Maria
Oncotag Desenvolvimento de Produtos e Servicos Para Saúde HumanaRua Tiradentes, 2689, Sala 102, Industrial, Contagem 32230-020, MG, Brazil.
Laboratório de Pesquisa Translacional em Oncologia, Núcleo de Ensino, Pesquisa e Inovação, Instituto Mário Penna, Belo Horizonte 30380-420, MG, Brazil.
Int J Mol Sci. 2025 Aug 29;26(17):8393. doi: 10.3390/ijms26178393.
Epithelial ovarian cancer (EOC) is the most lethal gynecologic malignancy, often diagnosed at an advanced stage due to its asymptomatic progression. The high recurrence rate and development of platinum-based chemotherapy resistance contribute to its poor prognosis. Despite advancements in molecular profiling, predictive biomarkers for chemotherapy response and recurrence risk remain limited. In this study, we developed OvarianTag™, a biomarker panel integrating apoptosis and necroptosis pathways, to predict chemotherapy benefit and disease progression in EOC patients. This observational study was conducted in two phases. In the first phase, 45 patients were recruited, and RNA was extracted from fresh ovarian tissues (normal, benign, and malignant). qRT-PCR was performed to assess the relative expression of genes involved in apoptosis and necroptosis-regulated cell death pathways. Machine learning algorithms were applied to identify the relevant prognostic markers, leading to the development of OvarianTag™. In the second phase, 55 additional EOC patients were included, and their formalin-fixed, paraffin-embedded (FFPE) tumor samples were analyzed using qRT-PCR. The classifier algorithm incorporated hierarchical clustering to stratify patients based on gene expression profiles. Significant differences in , , and expression levels were observed between patient groups. downregulation was strongly correlated with platinum resistance and a poor prognosis. Decision tree models achieved 83.3% accuracy in predicting platinum response and 79.2% accuracy in recurrence risk stratification. The OvarianTag™ classifier demonstrated high sensitivity and specificity in identifying high-risk patients, supporting its potential as a prognostic tool. The OvarianTag™ panel provides a novel approach for risk stratification in EOC, integrating apoptosis and necroptosis pathways to refine chemotherapy response prediction and recurrence risk assessment. This molecular assay has the potential to guide personalized treatment strategies, enhancing clinical decision-making and improving patient outcomes. Further validation in independent cohorts is warranted to establish its clinical utility.
上皮性卵巢癌（EOC）是最致命的妇科恶性肿瘤，由于其无症状进展，通常在晚期才被诊断出来。铂类化疗耐药的高复发率和发展导致其预后较差。尽管分子谱分析取得了进展，但用于化疗反应和复发风险的预测生物标志物仍然有限。在本研究中，我们开发了OvarianTag™，这是一种整合凋亡和坏死性凋亡途径的生物标志物面板，用于预测EOC患者的化疗获益和疾病进展。这项观察性研究分两个阶段进行。在第一阶段，招募了45名患者，并从新鲜卵巢组织（正常、良性和恶性）中提取RNA。进行qRT-PCR以评估参与凋亡和坏死性凋亡调节的细胞死亡途径的基因的相对表达。应用机器学习算法来识别相关的预后标志物，从而开发出OvarianTag™。在第二阶段，又纳入了55名EOC患者，并使用qRT-PCR分析他们的福尔马林固定、石蜡包埋（FFPE）肿瘤样本。分类算法结合层次聚类，根据基因表达谱对患者进行分层。在患者组之间观察到 、 和 表达水平存在显著差异。 下调与铂耐药和不良预后密切相关。决策树模型在预测铂反应方面的准确率达到83.3%，在复发风险分层方面的准确率达到79.2%。OvarianTag™分类器在识别高危患者方面表现出高敏感性和特异性，支持其作为一种预后工具的潜力。OvarianTag™面板为EOC的风险分层提供了一种新方法，整合凋亡和坏死性凋亡途径以优化化疗反应预测和复发风险评估。这种分子检测方法有可能指导个性化治疗策略，加强临床决策并改善患者预后。有必要在独立队列中进行进一步验证以确定其临床效用。
Zotero 插件
