Nature Cancer

Multimodal data integration using machine learning improves risk stratification of high-grade serous ovarian cancer

Integration of genomic, histopathology, radiomics and clinical data using machine learning improves risk stratification for 444 patients with high-grade serous ovarian cancer.

Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology

Abstract Clinical decision-making in oncology is complex, requiring the integration of multimodal data and multidomain expertise. We developed and evaluated an autonomous clinical artificial intelligence (AI) agent leveraging GPT-4 with multimodal precision oncology tools to support personalized clinical decision-making. The system incorporates vision transformers for detecting microsatellite instability and KRAS and BRAF mutations from histopathology slides, MedSAM for radiological image segmentation and web-based search tools such as OncoKB, PubMed and Google. Evaluated on 20 realistic multimodal patient cases, the AI agent autonomously used appropriate tools with 87.5% accuracy, reached correct clinical conclusions in 91.0% of cases and accurately cited relevant oncology guidelines 75.5% of the time. Compared to GPT-4 alone, the integrated AI agent drastically improved decision-making accuracy from 30.3% to 87.2%. These findings demonstrate that integrating language models with precision oncology and search tools substantially enhances clinical accuracy, establishing a robust foundation for deploying AI-driven personalized oncology support systems.

A phase 1 trial of adoptive transfer of vaccine-primed autologous circulating T cells in ovarian cancer

We have previously shown that vaccination with tumor-pulsed dendritic cells amplifies neoantigen recognition in ovarian cancer. Here, in a phase 1 clinical study ( NCT01312376 /UPCC26810) including 19 patients, we show that such responses are further reinvigorated by subsequent adoptive transfer of vaccine-primed, ex vivo-expanded autologous peripheral blood T cells. The treatment is safe, and epitope spreading with novel neopeptide reactivities was observed after cell infusion in patients who experienced clinical benefit, suggesting reinvigoration of tumor-sculpting immunity.

Advances in ovarian cancer, from biology to treatment

A first-in-class polymerase theta inhibitor selectively targets homologous-recombination-deficient tumors

DNA polymerase theta (POLθ) is synthetic lethal with Homologous Recombination (HR) deficiency and thus a candidate target for HR-deficient cancers. Through high-throughput small molecule screens we identified the antibiotic Novobiocin (NVB) as a specific POLθ inhibitor that selectively kills HR-deficient tumor cells

Eliminating cervical cancer in the COVID-19 era

Clinical and translational advances in ovarian cancer therapy

Ovarian cancer is an aggressive disease that is frequently detected at advanced stages and is initially very responsive to platinum-based chemotherapy. However, the majority of patients relapse following initial surgery and chemotherapy, highlighting the urgent need to develop new therapeutic strategies. In this Review, we outline the main therapeutic principles behind the management of newly diagnosed and recurrent epithelial ovarian cancer and discuss the current landscape of targeted and immune-based approaches.

Single-cell analyses implicate ascites in remodeling the ecosystems of primary and metastatic tumors in ovarian cancer

AbstractOvarian cancer (OC) is an aggressive gynecological tumor usually diagnosed with widespread metastases and ascites. Here, we depicted a single-cell landscape of the OC ecosystem with five tumor-relevant sites, including omentum metastasis and malignant ascites. Our data reveal the potential roles of ascites-enriched memory T cells as a pool for tumor-infiltrating exhausted CD8+T cells and T helper 1-like cells. Moreover, tumor-enriched macrophages exhibited a preference for monocyte-derived ontogeny, whereas macrophages in ascites were more of embryonic origin. Furthermore, we characterized MAIT and dendritic cells in malignant ascites, as well as two endothelial subsets in primary tumors as predictive biomarkers for platinum-based chemotherapy response. Taken together, our study provides a global view of the female malignant ascites ecosystem and offers valuable insights for its connection with tumor tissues and paves the way for potential markers of efficacy evaluation and therapy resistance in OC.

Sentinel lymph node biopsy without systematic pelvic lymphadenectomy in females with early-stage cervical cancer: final outcome of the SENTIX prospective, single-arm, noninferiority, international trial

Sentinel lymph node (SLN) biopsy with ultrastaging is standard in endometrial and vulvar cancers, whereas systematic pelvic lymphadenectomy (PLND) remains recommended in cervical cancer. The SENTIX trial prospectively evaluated the safety of SLN biopsy without PLND in early-stage cervical cancer. Female patients, International Federation of Gynaecology and Obstetrics 2018 stage IA1/LVSI+ to IB2 disease, were enrolled between 2016 and 2020 across 47 sites in 18 countries. All underwent SLN biopsy followed by hysterectomy/trachelectomy. Patients with undetected, unilateral or intraoperatively metastatic SLNs were excluded from the intention-to-treat cohort. SLNs were assessed by pathological ultrastaging. Of 731 patients enrolled, 594 formed the intention-to-treat cohort. SLN metastases were identified in 82 patients (12%), 56.1% intraoperatively and 43.9% by ultrastaging. At 2 years, the recurrence rate was 6.1% (one-sided 95% CI 7.9%), confirming noninferiority to the 7% reference rate. Two-year disease-free and overall survival rates were 93.3% (95% CI 94.9-91.6) and 97.9% (95% CI 98.9-97.0), respectively. Here we show that SLN biopsy without systematic PLND did not increase the risk of recurrence in patients with early-stage cervical cancer. Pathological ultrastaging of SLNs detected about 44% of N1 cases, which would be missed by a standard lymph node assessment. Trial registration: ClinicalTrials.gov ( NCT02494063 ).

Suvemcitug plus chemotherapy in women with platinum-resistant recurrent ovarian cancer: the SCORES randomized, double-blinded, phase 3 trial

Abstract In the SCORES study ( NCT04908787 ), women with ovarian cancer that progressed within 6 months after completing platinum-based therapy were randomized (2:1) to receive suvemcitug (1.5 mg kg −1 ), an antibody to vascular endothelial growth factor or placebo every 2 weeks, with chemotherapy (paclitaxel, topotecan or PEGylated liposomal doxorubicin). The primary endpoint was progression-free survival (PFS). The key secondary endpoint was overall survival (OS). Other secondary endpoints included objective response rate, disease control rate, duration of response, quality of life, safety, pharmacokinetics and antidrug antibodies. Between June 5, 2021 and October 11, 2024, 421 participants were randomized (49.4% and 49.4% previously exposed to antiangiogenic agents and poly(ADP-ribose) polymerase inhibitors, respectively). Median PFS was 5.5 and 2.7 months in the suvemcitug and placebo arms, respectively (hazard ratio: 0.46, 95% confidence interval (CI): 0.35–0.60, P  < 0.001), meeting the primary endpoint. Median OS was 15.3 versus 14.0 months, respectively (hazard ratio: 0.77, 95% CI: 0.60–0.99, P  = 0.03). Decreased neutrophil count and decreased white blood cell count were the most common grade ≥3 treatment-emergent adverse events (TEAEs) in the suvemcitug arm. No suvemcitug-related grade 5 TEAE occurred. In conclusion, the addition of suvemcitug to chemotherapy significantly improved PFS and OS, with tolerable toxicities.