Whilst researches elucidating a diversity of intracellular mechanisms, platinum-resistant epithelial ovarian cancer (EOC) remains a major challenge in the treatment of ovarian cancer. Here we report that Exo70, a key subunit of the exocyst complex, contributes to both innate and acquired cisplatin resistance of EOC. Upregulation of Exo70 is observed in EOC tissues and is related to platinum resistance and progression-free survival of EOC patients. Exo70 suppressed the cisplatin sensitivity of EOC cells through promoting exocytosis-mediated efflux of cisplatin. Moreover, cisplatin-induced autophagy-lysosomal degradation of Exo70 protein by modulating phosphorylation of AMPK and mTOR, thereby reducing the cellular resistance. However, the function was hampered during prolonged cisplatin treatment, which in turn stabilized Exo70 to facilitate the acquired cisplatin resistance of EOC cells. Knockdown of Exo70, or inhibiting exocytosis by Exo70 inhibitor Endosidin2, reversed the cisplatin resistance of EOC cells both in vitro and in vivo. Our results suggest that Exo70 overexpression and excessive stability contribute to innate and acquired cisplatin resistance through the increase in cisplatin efflux, and targeting Exo70 might be an approach to overcome cisplatin resistance in EOC treatment.

Synergistic chemo-phototherapy has offered tremendous potential in cancer treatment. Nevertheless, nanosystems usually suffer from the complexity of multicomponents (polymeric or inorganic materials), which results in carrier-related toxicity issues. Moreover, the GSH over-expression of tumor cells seriously compromises ROS therapeutic efficiency. Herein, we designed a self-delivered nanodrug via Cu(II) coordination-driven co-self-assembly of celastrol (CST, a chemo-drug with anti-angiogenesis activity) and indocyanine green (ICG, a photosensitizer) for synergistic chemo-phototherapy with GSH depletion. The nanodrug was further cloaked by an erythrocyte membrane (RBC) to prolong the circulation time. Within the tumor microenvironment, the nanodrug would be disassembled upon intracellular GSH triggering. Moreover, the released Cu(II) could efficiently deplete the GSH, thus damaging the ROS-scavenging system and amplifying the phototherapeutic efficiency upon laser irradiation. The in vivo experiments validated the highly effective accumulation at tumor sites, potent tumor growth inhibition, and inappreciable systemic toxicity. The tumor microenvironment-responsive coordination-driven self-assembled biomimetic nanodrug may hold potential applications in tumor theranostics.

Cell instances segmentation in cervical cytology images have crucial importance for cancer screening and automatic biomedical image analysis. However, that the highly overlapping cells cause ambiguous boundaries makes accurate cell segmentation still a challenging task. This paper proposes a novel transformer-based neural network to segment overlapping cervical cells, by using the technique of denoising with Mask2Former. The model optimization includes two aspects. One is that class embeddings of ground truth categories are used as extra content queries to transformer decoder. The other is that noised ground truth masks are fed into each layer of transformer decoder and then the model is trained to reconstruct the original masks. Compared with the best result on ISBI2014 dataset, the proposed model obtains performance improvement by \(3.4\%\) on DSC, \(2.3\%\) on TPRp and \(2\%\) on FNRo, respectively. Especially, the masks predicted by our model are more precise than other methods and our previous models. Those results show that the proposed method could be effective in segmenting overlapping cells and help pathologists detect cell lesions.

Introduction: Global human papillomavirus (HPV) prevalence among women rose significantly from 14% (2019) to 24% (2024), underscoring the need to understand transmission dynamics and public health impact. Although multi-genotype infections are increasingly documented, evidence remains limited on their combined effect on cervical lesion severity and transmission, especially in regional populations. Using clinical datasets from Xiamen, China, this study evaluates co-infection patterns through integrated statistical and dynamical models to quantify associations with histological severity and transmissibility. Methods: Data were sourced from positive HPV nucleic acid tests at two hospitals in Xiamen. Genotyping employed multiplex PCR-based flow fluorescence hybridisation. Cumulative link models (CLMs) were used to assess associations between multi-genotype infections and cervical lesion severity. Concurrently, an ordinary differential equation transmission model was developed to estimate reproduction numbers, comparing transmission potential across infection types. Results: Of 1 33 438 samples, 15 939 (11.9%) were HPV-positive, covering 27 genotypes. HPV 16, 52 and 58 were the most prevalent high-risk types. Co-infections involving these genotypes showed strong inter-hospital correlation in pairing patterns (Pearson’s r =0.851). Co-infection severity association was context-specific: the number of genotypes predicted severity in the screening population (eg, quadruple infections OR=1.47, 95% CI 1.23 to 1.71, p<0.01) but not in the referral population. Conversely, co-infections exhibited consistently higher relative transmissibility indices in both settings (eg, high-low-risk co-infection median model-derived R 0 : 2.57–6.82). Conclusions: HPV co-infection impacts are modulated by patient population: relative transmission potential is broadly elevated, whereas histological severity effects are marked in screening cohorts but minimal in referral groups. Context-aware public health strategies—adapting co-infection screening and interventions to clinical setting—are urged for more effective and efficient disease control.