Poly (ADP-ribose) polymerase inhibitors (PARPis) benefit homologous recombination-deficient (HRD) malignancies, yet resistance remains a major challenge. Leveraging specimens from a prospective neoadjuvant niraparib monotherapy trial in treatment-naive, high-grade serous ovarian cancer, we integrated PhenoCycler-Fusion spatial profiling, scRNA-Seq, and multiplex immunohistochemistry to identify 2 therapeutic-modulated cellular neighborhoods: an IFN+ tumor cell–enriched niche that expands in resistant lesions and a niche enriched in tumor-associated macrophage (TAM) that persists but acquires enhanced immunosuppressive features. Mechanistically, sustained tumor cell–derived IFN induced osteopontin (SPP1) expression in TAMs via STAT signaling, creating immunosuppressive niches enriched in Tregs and myofibroblastic cancer–associated fibroblasts with intensified cell-cell interactions. SPP1 directly suppressed T cell signaling and effector function. High baseline SPP1+ cells predicted lower response rate (30.0% vs. 76.2%; P = 0.021) and shorter progression-free survival (median 13.5 vs. 28.3 months; P = 0.0006). In HRD mouse models, SPP1 blockade restored PARPi sensitivity, reversed acquired resistance, and enhanced T cell cytotoxicity—effects abrogated in immunodeficient mice, confirming immune dependence. These data establish a spatial IFN-SPP1 axis whereby persistent tumor cell IFN reprograms TAMs to promote PARPi resistance, position SPP1 as a key therapeutic target and prognostic biomarker for this therapy, and underscore therapeutic potential of microenvironment-targeted strategies to overcome PARPi resistance.
Dan Liu, Kangjia Tao, Cheng Xu, Wen Yang, Chujun Cai, Cui Feng, Kairong Xiong, Sisi Wu, Yaying Lin, Zikun Peng, Jianhua Chi, Wen Pan, Qing Zhong, Jiahao Liu, Xiong Li, Xingzhe Liu, Dongchen Zhou, Ding Ma, Guang-Nian Zhao, Yu Xia, Yong Fang, Qinglei Gao
Usage data is cumulative from March 2026 through August 2026.
| Usage | JCI | PMC |
|---|---|---|
| Text version | 3,317 | 88 |
| 1,418 | 32 | |
| Figure | 790 | 0 |
| Supplemental data | 642 | 5 |
| Citation downloads | 271 | 0 |
| Totals | 6,438 | 125 |
| Total Views | 6,563 | |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.