Yang Sun

Purpose Radiation proctitis (RP) is a frequent and clinically relevant complication of radiotherapy (RT) in cervical cancer (CC). However, effective tools for individualized RP risk prediction remain limited. This study aimed to develop and validate an MRI-based radiomic risk score for early RP prediction in CC patients. Methods This retrospective study included 375 CC patients who underwent radical RT. Patients were randomly assigned to training (80%) and validation (20%) cohorts. Radiomics features were extracted from pre-treatment MRI and screened using Mann–Whitney U test, intraclass correlation coefficient, and least absolute shrinkage and selection operator regression. Logistic regression analyses identified independent predictors of RP, and a clinical-radiomics nomogram was constructed. Model performance was assessed by the area under the curve (AUC), calibration, and decision curve analysis (DCA). Results Of 1,906 extracted features, 9 were selected to construct the radiomics risk score. Independent predictors of RP included lymphocyte counts ≥ 2.6 × 10⁹/L, platelet–lymphocyte ratio ≥ 196, rectal mean dose ≥ 47.22 Gy, and radiomics risk score≥-0.4. The nomogram incorporating these factors achieved AUCs of 0.810 (training) and 0.730 (validation), with good calibration and clinical utility demonstrated by DCA. Conclusion The proposed MRI-based radiomic risk score enables accurate, noninvasive prediction of RP in CC patients and provides a practical tool to optimize individualized RT planning and toxicity management.

Background: Immunotherapy has become a powerful clinical strategy for treating recurrent or metastatic cervical cancer (R/M CC). Cadonilimab, a novel anti-PD-1/CTLA-4 bispecific antibody, has shown substantial clinical benefits in cancer treatment. However, there is no real-world evidence of cadonilimab with a considerable sample size in R/M CC. Hence, we aim to assess the efficacy and safety of cadonilimab in R/M CC patients and explore its potential mechanism.Methods: This retrospective real-world study examined a sample of R/M CC patients treated with cadonilimab at 13 large academic medical centers in China from July 6, 2022, to October 1, 2023. The outcomes were objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), overall survival (OS), as well as safety profiles. Additionally, the programmed cell death 1 ligand 1 (PD-L1) was detected by immunohistochemistry to confirm its predictive values. Whole exome sequencing (WES) was also performed to investigate its potential antitumor mechanisms.Results: Among the 129 patients with measurable disease, the ORR was 38.8%, consisting of complete and partial responses in 8.5% and 30.2% of patients, respectively. The DCR was 72.1%. The median PFS was 12.4 months, while the median OS has not yet been reached. Subgroup analysis showed a numerical trend toward longer median PFS in patients with PD-L1 CPS ≥ 1 compared with CPS < 1 (14.0 vs. 12.8 months; P = 0.235). Moreover, combined therapy of cadonilimab and radiotherapy was identified as an independent prognostic factor for both OS and PFS. The most common grade 3 or worse adverse event was anemia (28 [20.1%]), decreased white blood cell count (24 [17.2%]), and decreased neutrophil count (20 [14.4%]). The most prevalent genetic variant was PIK3CA, highlighting the importance of the PI3K-AKT pathway in the antitumor mechanism of cadonilimab.Conclusions: Cadonilimab shows an encouraging tumor response rate, with a manageable safety profile in patients with R/M CC. Notably, cadonilimab is also effective for those with PD-L1 CPS <1, suggesting a broad range of application prospects in R/M CC.Clinical Trial Registration: https://www.clinicaltrials.gov, identifier NCT06140589.

The outcome of patients with recurrent/metastatic cervical cancer (R/M CC) is poor, with a 5-year survival rate of only 10%–20%. Recent advances in immunotherapy renewed its interest in R/M CC treatment. It has been suggested that cadonilimab, a novel bispecific antibody targeting programmed death 1 (PD-1) and cytotoxic T-lymphocyte antigen-4 (CTLA-4), significantly improved the survival outcomes of the R/M CC. In the present study, we reported a programmed death ligand 1 (PD-L1) and human epidermal growth factor receptor 2 (HER-2) positive CC case at stage IV who was treated with cadonilimab and achieved a surprising radiographic complete response (CR) for 10 months, even in the PD-L1 negative metastatic site. Demographic, clinical, histopathological, laboratory, treatment regime and imaging data were recorded. Unfortunately, the patient progressed rapidly during maintenance therapy when cadonilimab was replaced by sintilimab, the monoclonal antibody against PD-1, indicating the more powerful anti-tumor activity of dual blockade immunotherapy. To conclude, cadonilimab offers a promising and effective therapeutic approach for R/M CC. Notably, HER-2 is also expected to be a new reference target for cadonilimab therapy.

Platinum-based neoadjuvant chemotherapy (NACT) followed by radical hysterectomy has been proposed as an alternative treatment approach for cervical cancer (CC) in stage Ib2-IIb, who had a strong desire to be treated with surgery. Our study aims to develop a model based on multimodal MRI by using radiomics and deep learning to predict the treatment response in CC patients treated with neoadjuvant chemoradiotherapy (NACRT). From August 2009 to June 2013, CC patients in stage Ib2-IIb (FIGO 2008) who received NACRT at Fujian Cancer Hospital were enrolled in our study. Clinical information, contrast-enhanced T1-weighted imaging (CE-T1WI), and T2-weighted imaging (T2WI) data were respectively collected. Radiomic features and deep abstract features were extracted from the images using radiomics and deep learning models, respectively. Then, ElasticNet and SVM-RFE were employed for feature selection to construct four single-sequence feature sets. Early fusion of two multi-sequence feature sets and one hybrid feature set were performed, followed by classification prediction using four machine learning classifiers. Subsequently, the performance of the models in predicting the response to NACRT was evaluated by separating patients into training and validation sets. Additionally, overall survival (OS) and disease-free survival (DFS) were assessed using Kaplan-Meier survival curves. Among the four machine learning models, SVM exhibited the best predictive performance (AUC=0.86). Among the seven feature sets, the hybrid feature set achieved the highest values for AUC (0.86), ACC (0.75), Recall (0.75), Precision (0.81), and F1-score (0.75) in the validation set, outperforming other feature sets. Furthermore, the predicted outcomes of the model were closely associated with patient OS and DFS (p = 0.0044; p = 0.0039). A model based on MRI images with features from multiple sequences and different methods could precisely predict the response to NACRT in CC patients. This model could assist clinicians in devising personalized treatment plans and predicting patient survival outcomes.