Shu Tong, Yang Fan, Gao Lin, Zhou Jinhua, Zhang Chao, Chen Youguo, Zheng Hong, Li Jundong
Peking University Cancer Hospital & Institute, Department of Gynecologic Oncology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Beijing, China.
Sun Yat-sen University Cancer Center, Guangdong Provincial Clinical Research Center for Cancer, State Key Laboratory of Oncology in South China, Department of Gynecological Oncology, Guangzhou, China.
Int J Gynecol Cancer. 2025 Jun 25;35(9):101987. doi: 10.1016/j.ijgc.2025.101987.
This study aimed to develop a high-performance prognostic model to predict poly(ADP-ribose) polymerase inhibitor (PARPi) treatment outcomes in patients with ovarian cancer.
This was a retrospective cohort study. Inclusion criteria were high-grade serous or endometroid carcinoma, clear cell carcinoma with platinum-sensitive disease (>6 months without progression from the end of platinum) or platinum-responsive disease eligible for front-line PARPi therapy. All collected samples underwent OncoWES-HRD analysis, with an Homologous recombination deficiency (HRD) score threshold set at 39. We performed LASSO regression analysis to develop a predictive model for assessing the effectiveness of PARPi treatment in patients with ovarian cancer. The data were analyzed using R software.
We collected primary tumors from 221 Chinese patients with ovarian cancer, of whom 99 patients with high-grade serous ovarian carcinoma received PARPi treatment. Based on the HRD score threshold, 144 patients were classified as HRD-positive and 77 as HRD-negative. We found that the HRD-positive group had higher mutation frequencies of ANKHD1 and MUC16 compared to the HRD-negative group. Furthermore, biomarkers such as clonal mutations, BRCA mutations, high indel burden, and high loss-of-heterozygosity were associated with notably higher HRD scores and longer progression-free survival. Using HRD genomic features, we established a LASSO regression-based risk score model for predicting PARPi treatment outcomes. This model showed promising performance compared to other HRD assessments (the OncoWES-HRD score and the OncoWES-HRD and BRCA metrics), with a higher area under the curve and significantly longer progression-free survival (p< .05) in both training and test cohorts.
We developed a novel prognostic model that can predict PARPi treatment outcomes, offering a valuable tool for identifying patients who may benefit from PARPi therapy in ovarian cancer. However, the model needs further validation.
本研究旨在开发一种高性能的预后模型，以预测卵巢癌患者接受聚（ADP - 核糖）聚合酶抑制剂（PARPi）治疗的结果。
这是一项回顾性队列研究。纳入标准为高级别浆液性或子宫内膜样癌、铂敏感疾病（铂治疗结束后>6个月无进展）的透明细胞癌或符合一线PARPi治疗条件的铂反应性疾病。所有收集的样本均进行了OncoWES - HRD分析，同源重组缺陷（HRD）评分阈值设定为39。我们进行了LASSO回归分析，以建立一个预测模型，用于评估PARPi治疗卵巢癌患者的有效性。数据使用R软件进行分析。
我们收集了221例中国卵巢癌患者的原发性肿瘤，其中99例高级别浆液性卵巢癌患者接受了PARPi治疗。根据HRD评分阈值，144例患者被分类为HRD阳性，77例为HRD阴性。我们发现，与HRD阴性组相比，HRD阳性组的ANKHD1和MUC16突变频率更高。此外，克隆突变、BRCA突变、高插入缺失负担和高杂合性缺失等生物标志物与明显更高的HRD评分和更长的无进展生存期相关。利用HRD基因组特征，我们建立了一个基于LASSO回归的风险评分模型，用于预测PARPi治疗结果。与其他HRD评估（OncoWES - HRD评分以及OncoWES - HRD和BRCA指标）相比，该模型表现出良好的性能，在训练和测试队列中均具有更高的曲线下面积和显著更长的无进展生存期（p <.05）。
我们开发了一种新的预后模型，可预测PARPi治疗结果，为识别可能从PARPi治疗中获益的卵巢癌患者提供了有价值的工具。然而，该模型需要进一步验证。
