N6-methyladenosine (m⁶A) RNA methylation has recently been found involving in regulatory mechanism of the tumor progression. Our aim was to explore the biological function and clinical significance of the m⁶A methyltransferase METTL3 in intrahepatic cholangiocarcinoma (ICC). In this study, we revealed that METTL3 was upregulated and predicted poor prognosis of patients with ICC. Multivariate regression analysis demonstrated that METTL3 expression was an independent predictor for overall survival in patients with ICC. Moreover, METTL3 knockdown inhibited ICC progression, while METTL3 overexpression showed the opposite effect. METTL3 inhibitor STM2457 also showed anti-tumor effect in ICC. Mechanistically, METTL3 transcription was driven by H3K4me3 activation. Upregulation of METTL3 mediated m⁶A modification of IFIT2 mRNA and accelerated IFIT2 mRNA decay in a YTHDF2-dependent manner, which promoted the development of ICC and lead to poorer prognosis. In summary, our findings revealed that H3K4me3 activation-driven METTL3 transcription promotes ICC progression by YTHDF2-mediated IFIT2 mRNA degradation, suggesting that METTL3 may serve as a potential target for human ICC therapy.

Objectives: To develop a computed tomography (CT) radiomics-based interpretable machine learning (ML) model to predict the pathological grade of pancreatic neuroendocrine tumors (pNETs) in a non-invasive manner. Methods: Patients with pNETs who underwent contrast-enhanced abdominal CT between 2010 and 2022 were included in this retrospective study. Radiomics features were extracted, and five radiomics-based ML models, namely logistic regression (LR), random forest (RF), support vector machine (SVM), XGBoost, and GaussianNB, were developed. The performance of these models was evaluated using a time-independent testing set, and metrics such as sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AUC) were calculated. The accuracy of the radiomics model was compared to that of needle biopsy. The Shapley Additive Explanation (SHAP) tool and the correlation between radiomics and biological features were employed to explore the interpretability of the model. Results: A total of 122 patients (mean age: 50 ± 14 years; 53 male) were included in the training set, whereas 100 patients (mean age: 48 ± 13 years; 50 male) were included in the testing set. The AUCs for LR, SVM, RF, XGBoost, and GaussianNB were 0.758, 0.742, 0.779, 0.744, and 0.745, respectively, with corresponding accuracies of 73.0%, 70.0%, 77.0%, 71.9%, and 72.9%. The SHAP tool identified two features of the venous phase as the most significant, which showed significant differences among the Ki-67 index or mitotic count subgroups (p < 0.001). Conclusions: An interpretable radiomics-based RF model can effectively differentiate between G1 and G2/3 of pNETs, demonstrating favorable interpretability. Clinical relevance statement: The radiomics-based interpretable model developed in this study has significant clinical relevance as it offers a non-invasive method for assessing the pathological grade of pancreatic neuroendocrine tumors and holds promise as an important complementary tool to traditional tissue biopsy. Key Points: • A radiomics-based interpretable model was developed to predict the pathological grade of pNETs and compared with preoperative needle biopsy in terms of accuracy. • The model, based on CT radiomics, demonstrated favorable interpretability. • The radiomics model holds potential as a valuable complementary technique to preoperative needle biopsy; however, it should not be considered a replacement for biopsy.

Background: Intrahepatic cholangiocarcinoma (ICC) is an aggressive cancer with exceedingly poor prognosis, and chemoresistance is a huge challenge for treatment. N6‐methyladenosine (m⁶A) modification plays an important role in the progression and chemoresistance of cancers. We aimed to investigate the oncogenic function and therapeutic significance of the m⁶A binding protein, YTH domain family 2 (YTHDF2), in ICC progression and cisplatin‐based chemotherapy.Methods: Several independent data sets were used to assess the expression of YTHDF2 in ICC, particularly in chemoresistant ICC. Knockdown and overexpression were used to evaluate the effects of YTHDF2 on tumourigenesis and cisplatin response in ICC. Multi‐omics sequencing was performed to identify target genes. RIP, dual luciferase reporter, RNA stability experiment and loss‐of‐function assays were conducted to study the mechanisms underlying the oncogenic function of YTHDF2. Furthermore, patient‐derived xenograft (PDX) model was established to determine the effect of combination treatment of YTHDF2 siRNA and cisplatin in ICC.Results: Our study showed that YTHDF2 was upregulated in ICC tissues, particularly in chemoresistant ICC tissues, and correlated with poor prognosis. Furthermore, silencing YTHDF2 led to inhibited proliferation, promoted apoptosis and G0/G1 cell cycle arrest. Its downregulation also enhanced DNA damage and sensitised ICC cells to cisplatin. YTHDF2 overexpression exerted the opposite results. Integration analysis using RNA‐seq, MeRIP‐seq and anti‐YTHDF2 RIP‐seq elucidated the role of YTHDF2 in tumourigenesis and cisplatin‐desensitising function by promoting the degradation of cyclin‐dependent kinase inhibitor 1B (CDKN1B) mRNA in an m⁶A‐dependent manner. Downregulation of CDKN1B increased the YTHDF2 silencing‐induced influence on tumourigenesis and cisplatin response to ICC. In addition, the combination treatment of YTHDF2 siRNA and cisplatin significantly enhanced the anti‐tumour effect of cisplatin in a chemoresistant ICC PDX model.Conclusions: YTHDF2 exhibits tumour oncogenic and cisplatin‐desensitising properties, which may offer insight into the development of novel combination therapeutic strategies for ICC.

Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignant cancer with complex genomic variations, and no targetable genomic lesions have been found yet. Super‐enhancers (SEs) have been found to contribute to the continuous and robust oncogenic transcription. Here, histone H3 lysine 27 acetylation (H3K27ac) is profiled in PDAC cell lines to establish SE landscapes. Concurrently, it is also shown that PDAC is vulnerable to the perturbation of the SE complex using bromodomain‐containing protein 4 (BRD4) inhibitor, JQ1, synergized with cyclin‐dependent kinase 7 (CDK7) inhibitor, THZ1. Formulations of hydrophobic l‐phenylalanine‐poly (ester amide) nanoparticles (NPs) with high drug loading of JQ1 and THZ1 (J/T@8P4s) are further designed and developed. J/T@8P4s is assessed for size, encapsulation efficiency, morphology, drug release profiles, and drug uptake in vitro. Compared to conventional free drug formulation, the nanodelivery system dramatically reduces the hepatotoxicity while significantly enhancing the tumor inhibition effects and the bioavailability of incorporated JQ1 and THZ1 at equal doses in a Gemcitabine‐resistant PDAC patient‐derived xenograft (PDX) model. Overall, the present study demonstrates that the J/T@8P4s can be a promising therapeutic treatment against the PDAC via suppression of SE‐associated oncogenic transcription, and provides a strategy utilizing NPs to assist the drug delivery targeting SEs.