Hendrikse Cynthia S E, Theelen Pauline M M, Verhaegh Wim, Lambrechts Sandrina, Bekkers Ruud L M, van de Stolpe Anja, Piek Jurgen M J
Department of Gynecology and Obstetrics and Catharina Cancer Institute, Catharina Hospital, 5623 EJ Eindhoven, The Netherlands.
GROW School for Oncology and Reproduction, Maastricht University, 6229 ER Maastricht, The Netherlands.
Cancers (Basel). 2024 Dec 2;16(23):4041. doi: 10.3390/cancers16234041.
: Advances in treatment options have barely improved the prognosis of ovarian carcinoma (OC) in recent decades. The inherent heterogeneity of OC underlies challenges in treatment (development) and patient stratification. One hurdle for effective drug development is the lack of patient-representative disease models available for preclinical drug research. Based on quantitative measurement of signal transduction pathway (STP) activity in cell lines, we aimed to identify cell line models that better mirror the different clinical subtypes of OC. : The activity of seven oncogenic STPs (signal transduction pathways) was determined by previously described STP technology using transcriptome data from untreated OC cell lines available in the GEO database. Hierarchal clustering of cell lines was performed based on STP profiles. Associations between cell line histology (original tumor), cluster, and STP profiles were analyzed. Subsequently, STP profiles of clinical OC tissue samples were matched with OC cell lines. : Cell line search resulted in 80 cell line transcriptome data from 23 GEO datasets, with 51 unique cell lines. These cell lines were derived from eight different histological OC subtypes (as determined for the primary tumor). Clustering revealed seven clusters with unique STP profiles. When borderline tumors (n = 6), high-grade serous (n = 51) and low-grade (n = 31) OC were matched with cell lines, twelve different cell lines were identified as potentially patient-representative OC cell line models. : Based on STP activity, we identified twelve different cell lines that were the most representative of the common subtypes of OC. These findings are important to improve drug development for OC.
近几十年来，治疗方案的进展几乎未能改善卵巢癌（OC）的预后。OC固有的异质性是治疗（研发）和患者分层面临挑战的根本原因。有效药物研发的一个障碍是缺乏可用于临床前药物研究的具有患者代表性的疾病模型。基于对细胞系中信号转导通路（STP）活性的定量测量，我们旨在识别能更好反映OC不同临床亚型的细胞系模型。：使用GEO数据库中未处理的OC细胞系的转录组数据，通过先前描述的STP技术确定了七种致癌STP（信号转导通路）的活性。基于STP谱对细胞系进行层次聚类。分析了细胞系组织学（原发肿瘤）、聚类和STP谱之间的关联。随后，将临床OC组织样本的STP谱与OC细胞系进行匹配。：细胞系搜索从23个GEO数据集中获得了80个细胞系转录组数据，有51个独特的细胞系。这些细胞系源自八种不同组织学类型的OC亚型（根据原发肿瘤确定）。聚类揭示了七个具有独特STP谱的簇。当将交界性肿瘤（n = 6）、高级别浆液性（n = 51）和低级别（n = 31）OC与细胞系进行匹配时，确定了12个不同的细胞系为潜在的具有患者代表性的OC细胞系模型。：基于STP活性，我们确定了12个不同的细胞系，它们是OC常见亚型中最具代表性的。这些发现对于改进OC的药物研发很重要。
