Med最新文献

Background: Digital subtraction angiography (DSA) devices are commonly used in numerous interventional procedures across various parts of the body, necessitating multiple scans per procedure, which results in significant radiation exposure for both doctors and patients. Inspired by generative artificial intelligence techniques, this study proposes GenDSA, a large-scale pretrained multi-frame generative model-based real-time and low-dose DSA imaging system.

Methods: GenDSA was developed to generate 1-, 2-, and 3-frame sequences following each real frame. A large-scale dataset comprising ∼3 million DSA images from 27,117 patients across 10 hospitals was constructed to pretrain, fine-tune, and validate GenDSA. Two other datasets from 25 hospitals were used for evaluation. Objective evaluations included SSIM and PSNR. Five interventional radiologists independently assessed the quality of the generated frames using the Likert scale and visual Turing test. Scoring consistency among the radiologists was measured using the Kendall coefficient of concordance (W). The Fleiss' kappa values were used for inter-rater agreement analysis for visual Turing tests.

Findings: Using only one-third of the clinical radiation dose, videos generated by GenDSA were perfectly consistent with real videos. Objective evaluations demonstrated that GenDSA's performance (PSNR = 36.83, SSIM = 0.911, generation time = 0.07 s/frame) surpassed state-of-the-art algorithms. Subjective ratings and statistical results from five doctors indicated no significant difference between real and generated videos. Furthermore, the generated videos were comparable to real videos in overall quality (4.905 vs. 4.935) and lesion assessment (4.825 vs. 4.860).

Conclusions: With clear clinical and translational values, the developed GenDSA can significantly reduce radiation damage to both doctors and patients during DSA-guided procedures.

Funding: This study was supported by the National Key R&D Program and the National Natural Science Foundation of China.

The ALIGN trial¹ demonstrates that atrasentan, an endothelin A (ETA) receptor antagonist, reduces proteinuria in patients with IgA nephropathy (IgAN), a key goal to slow progressive renal disease. These results are consistent with those from sparsentan,^2,3 a combined ETA and angiotensin inhibitor, in IgAN, suggesting two-year data will show atrasentan improves renal outcomes.

LITESPARK-005 evaluated belzutifan against everolimus in advanced renal cell carcinoma (RCC),¹ demonstrating significant progression-free survival improvement but failing to meet the overall survival (OS) co-primary endpoint. Despite FDA approval, the trial highlights key obstacles in drug development in RCC, given the absence of OS improvement, lack of biomarker studies, high financial toxicity, and limited accessibility outside the United States.

Background: Fish oil (FO), a mixture of omega-3 fatty acids mainly comprising docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA), has been recommended for patients with type 2 diabetes (T2D) and hypertriglyceridemia. However, its effects on lipidomic profiles and gut microbiota and the factors influencing triglyceride (TG) reduction remain unclear.

Methods: We conducted a 12-week, randomized, double-blind, placebo-controlled trial in 309 Chinese patients with T2D with hypertriglyceridemia (ClinicalTrials.gov: NCT03120299). Participants were randomly assigned (1:1) to receive either 4 g FO or corn oil for 12 weeks. The primary outcome was changes in serum TGs and the lipidomic profile, and the secondary outcome included changes in the gut microbiome and other metabolic variables.

Findings: The FO group had significantly better TG reduction (mean [95% confidence interval (CI)]: -1.51 [-2.01, -1.01] mmol/L) compared to the corn oil group (-0.66 [-1.15, -0.16] mmol/L, p = 0.02). FO significantly altered the serum lipid profile by reducing low-unsaturated TG species and increasing those containing DHA or EPA. FO had minor effects on gut microbiota, while baseline microbial features predicted the TG response to FO better than phenotypic or lipidomic features, potentially mediated by specific lipid metabolites. A total of 9 lipid metabolites significantly mediated the link between 4 baseline microbial variables and the TG response to FO supplementation.

Conclusions: Our findings demonstrate differential impacts of omega-3 fatty acids on lipidomic and microbial profiles in T2D and highlight the importance of baseline gut microbiota characteristics in predicting the TG-lowering efficacy of FO.

Funding: This study was funded by the National Nature Science Foundation.

Background: Zorifertinib (AZD3759), an epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI) with high blood-brain barrier penetration capability, demonstrated promising intracranial and systemic antitumor activity in phase 1 and 2 studies in central nervous system (CNS)-metastatic patients.

Methods: In this phase 3 EVEREST trial (ClinicalTrials.gov: NCT03653546), patients with EGFR-sensitizing mutations, advanced treatment-naive non-small cell lung cancer (NSCLC), and non-irradiated symptomatic or asymptomatic CNS metastases were randomized (1:1) to zorifertinib or first-generation EGFR-TKI (gefitinib or erlotinib; control). The primary endpoint was blinded independent central review (BICR)-assessed progression-free survival (PFS) per RECIST1.1.

Findings: Overall, 439 patients were randomized (zorifertinib n = 220; control n = 219). Most patients had the EGFR L858R mutation (55%) or >3 CNS lesions (54%). Median PFS was significantly longer with zorifertinib versus control (9.6 versus 6.9 months; hazard ratio [HR], 0.719; 95% confidence interval [CI], 0.580-0.893; p = 0.0024). Zorifertinib significantly prolonged intracranial PFS versus control (BICR per modified RECIST1.1: HR, 0.467; 95% CI, 0.352-0.619; investigator per RANO-BM: HR, 0.627; 95% CI, 0.466-0.844). Overall survival (OS) was immature; the estimated median OS was 37.3 months with zorifertinib and 31.8 months with control (HR, 0.833; 95% CI, 0.524-1.283) in patients subsequently treated with third-generation EGFR-TKIs. Safety profiles were consistent with previously reported data for zorifertinib.

Conclusions: Zorifertinib significantly improved systemic and intracranial PFS versus first-generation EGFR-TKIs; adverse events were manageable. Sequential use of zorifertinib and third-generation EGFR-TKIs showed the potential to prolong patients' survival. The results favor zorifertinib as a novel, well-validated first-line option for CNS-metastatic patients with EGFR-mutant NSCLC.

Funding: This work was funded by Alpha Biopharma (Jiangsu) Co., Ltd., China.

The phase III EVEREST trial evaluating zorifertinib in the treatment of metastatic EGFR-mutant NSCLC was groundbreaking in its specific inclusion of patients with brain metastases.¹ Zorifertinib prolonged systemic and intracranial progression-free survival compared with first-generation EGFR inhibitors, yet questions remain about its efficacy and toxicity compared with osimertinib.

Background: De novo pathogenic variants in GNAO1-the gene encoding the major neuronal G protein Gαo-cause pediatric encephalopathies and other neurological deficiencies largely refractory to available therapies. Zn²⁺ emerged to restore guanosine triphosphate hydrolysis and cellular interactions of pathogenic Gαo; dietary zinc salt supplementation improves lifespan and motoric function in a Drosophila disease model.

Methods: Using biochemical, animal, and first-in-human studies, we provide support for the patient stratification and application of zinc acetate in GNAO1-associated disorders.

Findings: We show that 16 different pathogenic missense variants cluster in three distinct groups in their responsiveness to Zn²⁺, and we provide the safety study in a mouse disease model. We further describe treatment of a 3-year-old patient with the common pathogenic GNAO1 variant c607G>A, p.Gly203Arg with oral 50 mg zinc (in the form of zinc acetate) daily, as applied in Wilson's disease. During 11 months of treatment, the patient shows cessation of daily dyskinetic crises, improved Burke-Fahn Marsden Dystonia Rating Scale movement score, reduction in epileptic seizures, and an excellent safety profile.

Conclusions: Our findings warrant a large-scale clinical trial and might set the new standard of care for GNAO1-related disorders.

Funding: This work was funded by the Russian Science Foundation (grant #21-15-00138) and GNAO1 España.

Background: Primary biliary cholangitis (PBC) is a progressive autoimmune liver disease. An inadequate response to ursodeoxycholic acid (UDCA) poses a high risk of progression toward end-stage liver disease. Gut dysbiosis has been implicated in PBC. Here, we aimed to investigate microbial signatures that permit risk stratification and provide mechanistic insights into novel therapies for PBC.

Methods: We prospectively recruited UDCA treatment-naive patients with PBC and performed metagenomic sequencing and metabolomic profiling using stool and serum samples obtained before (n = 132) and after (n = 59) treatment. PBC microbiome subtypes were identified using unsupervised machine learning methods and validated in two independent cohorts.

Findings: PBC baseline metagenomes clustered into two community subtypes characterized by varying abundances of Clostridia taxa. Compared with Clostridia^low microbiomes, Clostridia^high microbiomes were more similar to healthy controls. Notably, patients in the Clostridia^low subtype exhibited a 2-fold higher UDCA non-response rate compared to those in the Clostridia^high subtype (41% vs. 20%, p = 0.015). Integrative analysis of metagenomic and metabolomic data revealed divergent functional modules and metabolic activities between the two metacommunities. In particular, anaerobic fermentation and the production of bioactive metabolites, including tryptophan derivatives and secondary bile acids, crucial for immune regulation and gut barrier maintenance, were markedly diminished in the Clostridia^low subtype. Moreover, UDCA administration reconfigured the fecal microbial and metabolic profiles only in the Clostridia^high group. Importantly, the microbiome subtypes and their associations with UDCA response were reproducible in two independent treatment-naive PBC cohorts.

Conclusions: Characterizing baseline microbiota patterns may enable the prediction of treatment outcomes in PBC and facilitate personalized treatment strategies.

Funding: This research was mainly supported by the National Natural Science Foundation of China.

Professor Lin Shen, MD, graduated from Xuzhou Medical College in 1984 and Beijing Medical University in 1995. She trained at the US National Institutes of Health in 2000, focusing on therapies for gastrointestinal tumors. Currently, she is director of the Department of Gastrointestinal Oncology and Department of Early Drug Development Center, Peking University Cancer Hospital. Specializing in gastrointestinal oncology, she focuses on gastric, esophageal, colorectal, pancreatic, and gastrointestinal neuroendocrine cancers. She has led and participated in various national and international drug trials and steering committees.

Background: The efficacy and feasibility of pembrolizumab combined with chemotherapy in frontline management of advanced high-grade epithelial ovarian cancer (EOC) is unknown. Additionally, modification of the tumor microenvironment following neoadjuvant therapy is not well understood.

Methods: In this single-arm phase 2 trial (this study was registered at ClinicalTrials.gov: NCT02520154), eligible patients received up to 4 cycles of neoadjuvant chemotherapy followed by interval cytoreduction, 3 cycles of adjuvant intravenous carboplatin/weekly paclitaxel/pembrolizumab, and finally maintenance pembrolizumab until progression or toxicity (maximum 20 cycles). The primary endpoint was progression-free survival (PFS). Secondary endpoints included feasibility, toxicity, and overall survival (OS). PD-L1 staining, multiplex immunofluorescence staining, RNA sequencing, reverse-phase protein array analyses were performed on pre- and post-chemotherapy samples.

Findings: Thirty-one eligible patients were enrolled. Median PFS and OS was 14.88 (95% CI 12.39-23.00) and 57.43 months (95% CI 30.88-not reached), respectively. Among those with PD-L1 combined positive score (CPS) ≥10, the median PFS and OS were not reached compared to those with CPS <10 (10.50 and 30.90 months, respectively). Feasibility was met, with all patients completing their planned adjuvant cycles. Treatment discontinuation due to immune-related toxicity occurred in 6 patients (20%). Chemotherapy resulted in an infiltration of anti-tumor immune cells in the tumor microenvironment. Samples of patients with the best PFS demonstrated increased expression of NF-κB, TGF-β, and β-catenin signaling.

Conclusions: Pembrolizumab with chemotherapy was feasible and resulted in PFS within the historical range for this EOC population. Patients with CPS ≥10 may benefit more from this regimen, and future studies should investigate this potential biomarker.

Funding: This investigator-initiated trial was funded by Merck.