Cts-Clinical and Translational Science最新文献

Artificial intelligence (AI) is making a significant impact across various industries, including healthcare, where it is driving innovation and increasing efficiency. In the fields of Quantitative Clinical Pharmacology (QCP) and Translational Sciences (TS), AI offers the potential to transform traditional practices through the use of agentic workflows—systems with different levels of autonomy where specialized AI agents work together to perform complex tasks, while keeping “human in the loop.” These workflows can simplify processes, such as data collection, analysis, modeling, and simulation, leading to greater efficiency and consistency. This review explores how these AI-powered agentic workflows can help in addressing some of the current challenges in QCP and TS by streamlining pharmacokinetic and pharmacodynamic analyses, optimizing clinical trial designs, and advancing precision medicine. By integrating domain-specific tools while maintaining data privacy and regulatory standards, well-designed agentic workflows empower scientists to automate routine tasks and make more informed decisions. Herein, we showcase practical examples of AI agents in existing platforms that support QCP and biomedical research and offer recommendations for overcoming potential challenges involved in implementing these innovative workflows. Looking ahead, fostering collaborative efforts, embracing open-source initiatives, and establishing robust regulatory frameworks will be key to unlocking the full potential of agentic workflows in advancing QCP and TS. These efforts hold the promise of speeding up research outcomes and improving the efficiency of drug development and patient care.

Interleukin-1 receptor–associated kinase 4 (IRAK4), a key component of the Myddosome complex, mediates signaling through toll-like and interleukin-1 receptors. KT-474, a heterobifunctional IRAK4 degrader, was evaluated in a randomized, double-blind, placebo-controlled Phase 1 trial (NCT04772885) in single (25, 75, 150, 300, 600, 1000, and 1600 mg) and multiple (25, 50, 100, and 200 mg once daily [QD] for 14 days; or 200 mg twice weekly) ascending doses in healthy subjects. The pharmacokinetics of KT-474 and its diastereomers, the pharmacodynamics of KT-474, and the effect of food on KT-474 pharmacokinetics and the pharmacokinetic–pharmacodynamic analysis are presented as additional analyses to supplement the Ackerman et al. publication. KT-474 showed delayed absorption and prolonged elimination. Plasma exposure increased less than dose-proportionally, with single-dose exposure plateauing after the 1000 mg dose. Steady state was achieved after 7 days of daily dosing and resulted in a 3- to 4-fold accumulation in exposure. A significant food effect was observed at the 600 mg dose, with exposure increasing up to 2.57-fold when KT-474 was administered with a high-fat meal. Urinary excretion of KT-474 was < 1%. KT-474 demonstrated robust IRAK4 degradation in blood, with mean reductions of up to 98% observed at the 50–200 mg QD doses, as well as inhibition of ex vivo induction of a broad array of cytokines and chemokines by stimulants lipopolysaccharides and R848. Analysis of the relationship between plasma KT-474 concentration and IRAK4 reduction in blood indicated that plasma concentrations of 4.1–5.3 ng/mL would yield 80% IRAK4 reductions.

A target heart rate of 55–60 beats per minute is a goal for propranolol in both primary and secondary prophylaxis of variceal hemorrhage (VH). However, dose adjustments are often needed based on baseline heart rates. This study analyzed the effect of heart rate reduction from baseline with propranolol therapy on VH in patients with cirrhosis. A retrospective study was conducted on cirrhotic patients receiving propranolol for primary and secondary prophylaxis, 2008–2023. Patients were categorized as responders or non-responders based on the achievement of a heart rate reduction of ≥ 25% from baseline. The primary outcome was the incidence of VH. A survival analysis with propensity score-inverse probability treatment weighting was performed to associate heart rate reduction and the outcome. Among the 215 patients treated with propranolol for primary prophylaxis, 72 (33.5%) were responders and 143 (66.5%) non-responders. In secondary prophylaxis, 157 patients were included, with 52 (33.1%) classified as responders and 105 (66.9%) as non-responders. The median Child–Pugh score was 6 (range 5–12) for primary and 7 (range 5–12) for secondary prophylaxis. Responders and non-responders showed a similar incidence of VH in both primary (adjusted hazard ratio (HR) 1.70, 95% CI: 0.82–3.49) and secondary prophylaxis (adjusted HR 1.00, 95% CI: 0.34–2.90). Our analysis did not support achieving a heart rate reduction of ≥ 25% from baseline as a response to propranolol for the primary and secondary prophylaxis of VH in cirrhosis.

Anetumab ravtansine, like other ADC drugs, has high inter-patient variability in its pharmacokinetic (PK) and pharmacodynamic (PD) outcomes, which raises concerns about whether current dosing regimens are optimal for patients. The objective of this study was to evaluate covariates, especially body habitus and the innate immune system (IIS), which may affect anetumab ravtansine PK and PD as part of two clinical trials in patients with ovarian cancer and mesothelioma. Biomarkers of Fcγ receptors(FcγR) CD64 on IIS cells, total body weight (TBW), body surface area (BSA), and other covariates, such as sex and age, were analyzed for an association with anetumab ravtansine PK. Higher FcγR CD64, TBW, and BSA were associated with higher clearance (CL) of anetumab ravtansine (p < 0.05). However, there was no relationship between TBW or BSA and FcγR CD64. Female patients had a lower anetumab ravtansine CL (0.030 ± 0.007 L/h) compared to male patients (0.042 ± 0.006 L/h) (p < 0.05). In both studies, patients with stable disease (SD) and partial response (PR) had higher anetumab ravtansine AUC_0-inf compared to patients with progressive disease (PD). Individualizing the dose of anetumab ravtansine and potentially other ADCs based only on TBW is not optimal, whereas precision dosing of an ADC based on the inclusion of novel metrics of IIS biomarkers, body habitus, and sex may be more appropriate to reduce variability in PK exposure, reduce toxicity, and improve response.

Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide, with a rising incidence. The most common therapeutic choice for HCC is transarterial chemoembolization (TACE). While the standard protocol of TACE adopts cisplatin, the application of cisplatin needs hydration before and after the procedure to alleviate adverse effects on kidney function. Miriplatin, a lipophilic platinum complex, enables the omission of periprocedural hydration compared to cisplatin-based TACE. This study aimed to compare the survival benefit between miriplatin-based TACE and cisplatin-based TACE. Briefly, a retrospective cohort study in a single hospital was designed. Patients with HCC complicated by vascular invasion or distant metastasis were excluded. Background variability was adjusted using a propensity score matching; then, overall survival rates were compared using the Gehan-Breslow-Wilcoxon test. As a result, cisplatin and miriplatin were administered to 166 and 120 patients in TACE procedures. After adjusting baseline characteristics using a propensity score including age, sex, tumor burden, functional hepatic reserve, baseline year, and HbA1c, a pair of 99-patient cohorts was generated. Overall survivals did not differ significantly, despite poorer serum creatinine at baseline (0.89 vs. 0.74 mg/dL, p < 0.0001) and fewer patients being prepared for TACE through prehydration (18 patients vs. 38 ones, p = 0.0025) in the miriplatin group than in the cisplatin group. The median survival time was 1490 days for the miriplatin group and 1,830 days for the cisplatin group (p = 0.4022; ratio = 0.814; 95% confidence interval 0.546–1.215). In conclusion, miriplatin will benefit patients with HCC who cannot tolerate perioperative hydration.

Low-dose aspirin is recommended for prevention of hypertensive disorders of pregnancy (HDP) and preterm birth (PTB) in high-risk pregnancies. There is limited data on factors impacting aspirin response in pregnancy. We aimed to evaluate predictors of aspirin response and association with pregnancy outcome with a prospective study of high-risk pregnancies taking 81 mg aspirin daily. Aspirin response was evaluated with Platelet Function Assay-100 (PFA-100) epinephrine closure time at baseline (< 16 weeks' gestation), follow-up 1 (2–4 weeks after aspirin initiation), and follow-up 2 (28–32 weeks gestation). Multivariable regression was used to identify factors associated with PFA-100 at each visit, and results presented with beta coefficient (B) and confidence interval. The median difference (MD) in PFA-100 in those with and without HDP or PTB was compared. Results included 108 who completed follow-up 1 and 96 who completed both visits with > 75% adherence. PFA-100 was increased from baseline at follow-ups 1 and 2 (MD 37 (27–49); MD 26 (15.5–38.5) respectively). At follow-up 1, obesity (B = −30 (−53 to −7) seconds), diabetes (B = −39 (−75 to −2) seconds), and age (B = 2.2 (0.3–4.0) seconds per year increased) were associated with PFA-100 response. Those with HDP in the current pregnancy versus not had similar aspirin response, but those with PTB versus term birth in the current pregnancy had reduced aspirin response at 28–32 weeks (MD −27 (−54 to −3) seconds). A daily dose of 81 mg aspirin results in platelet inhibition throughout gestation. Obesity, diabetes, and younger age are associated with reduced aspirin response in pregnancy.

There are several disease-modifying antirheumatic drugs currently available to treat rheumatoid arthritis (RA). However, the optimal combination therapy with methotrexate for treating RA remains unclear. We aimed to identify combination therapies with high-efficacy and safety by employing the Bayesian method in a network meta-analysis. We systematically searched PubMed, Embase, CENTRAL, Ichushi web, and PMDA review reports and application materials through October 2020, and found 86 randomized controlled trials. The primary efficacy outcome was the 50% improvement rate according to the American College of Rheumatology criteria (ACR50), and the primary safety outcome was the incidence of serious adverse events. We calculated odds ratios (ORs) and its 95% credible intervals (CrIs) between each treatment, and the surface under the cumulative ranking curve (SUCRA) score for each treatment to rank disease-modifying antirheumatic drug combinations. Individually, most disease-modifying antirheumatic drugs combined with methotrexate are more likely to achieve ACR50 than methotrexate monotherapy, with significant differences (p < 0.05), whereas the incidence of serious adverse events was not significantly different compared with methotrexate monotherapy (p > 0.05). Infliximab combined with methotrexate had the highest efficacy ranking (OR = 10.53, 95% CrI: [3.20, 42.87], SUCRA score: 0.884), and etanercept combined with methotrexate had the highest safety ranking (OR = 0.29, 95% CrI: [0.03, 2.04], SUCRA score: 0.893). Comprehensive cluster analysis revealed that the combination of etanercept, an Fc-fusion protein targeting tumor necrosis factor α, with methotrexate demonstrated higher efficacy and safety. These findings could support the selection of combination therapies for the treatment of RA.

In March 2024, ponatinib received accelerated FDA approval for the treatment of newly diagnosed Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph + ALL) in combination with chemotherapy based on the Phase 3 PhALLCON study (NCT03589326), which demonstrated a higher rate of minimal residual disease (MRD)-negative complete remission (CR) at the end of induction (EOI) with ponatinib (34.4%) versus imatinib (16.7%; p = 0.002). Patients received ponatinib (30 mg QD with reduction to 15 mg QD upon achievement of MRD-negative CR at EOI) or imatinib (600 mg QD) combined with 20 cycles of reduced-intensity chemotherapy (induction: 3 cycles; consolidation: 6 cycles; and maintenance: 11 cycles). Ponatinib pharmacokinetics (PK) were similar in patients in PhALLCON and patients in a previous population PK analysis. Bayesian re-estimation of the previously developed population PK model adequately described PhALLCON PK data. Exposure–efficacy analyses did not identify a significant relationship between ponatinib exposure and the probability of MRD-negative CR at EOI (p = 0.619), suggesting a consistent efficacy benefit across exposures. Ponatinib exposure was not a significant predictor of arterial occlusive events, venous thromboembolic events, thrombocytopenia, or lipase increase (p > 0.05). However, higher exposures were associated with a higher probability of hypertension (p = 0.0340) and alanine aminotransferase (ALT) increase (p = 0.0034). Dose reduction from 30 to 15 mg was predicted to decrease the odds of experiencing hypertension by 37.7% and ALT increase by 44.2%. Collectively, exposure–response analyses support a favorable benefit–risk profile of the approved ponatinib dosage (30 mg QD reduced to 15 mg QD upon achievement of MRD-negative CR at EOI), combined with chemotherapy, for frontline treatment of Ph + ALL.

Despite interest in clinical trials with decentralized elements (DCTs), analysis of their trends in trial registries is lacking due to heterogeneous designs and unstandardized terms. We explored Llama 3, an open-source large language model, to efficiently evaluate these trends. Trial data were sourced from Aggregate Analysis of ClinicalTrials.gov, focusing on drug trials conducted between 2018 and 2023. We utilized three Llama 3 models with a different number of parameters: 8b (model 1), fine-tuned 8b (model 2) with curated data, and 70b (model 3). Prompt engineering enabled sophisticated tasks such as classification of DCTs with explanations and extracting decentralized elements. Model performance, evaluated on a 3-month exploratory test dataset, demonstrated that sensitivity could be improved after fine-tuning from 0.0357 to 0.5385. Low positive predictive value in the fine-tuned model 2 could be improved by focusing on trials with DCT-associated expressions from 0.5385 to 0.9167. However, the extraction of decentralized elements was only properly performed by model 3, which had a larger number of parameters. Based on the results, we screened the entire 6-year dataset after applying DCT-associated expressions. After the subsequent application of models 2 and 3, we identified 692 DCTs. We found that a total of 213 trials were classified as phase 2, followed by 162 phase 4 trials, 112 phase 3 trials, and 92 phase 1 trials. In conclusion, our study demonstrated the potential of large language models for analyzing clinical trial information not structured in a machine-readable format. Managing potential biases during model application is crucial.

Patients' longitudinal adherence to antidiabetic medication in routine clinical care remains unexplored. This study aimed to identify adherence groups among individuals with type 2 diabetes with up to 1 and 5 years of follow-up. This was a register-based cohort study using data from Swedish national health and population registers and the National Diabetes Register (2006–2022). New users of blood glucose–lowering drugs (other than insulin) were identified. Trajectories of the proportion of days covered (PDC) by any antidiabetic medication, including insulin, over 1- and 5-year periods were clustered using k-means for longitudinal data. Analyses up to 1- and 5-year follow-up periods included 75,421 individuals with an overall mean PDC of 0.7 and 283,795 individuals with an overall mean PDC of 0.3, respectively. K-means clustering identified two main adherence groups. For the 1-year follow-up, 70.6% of individuals were in the cluster with a higher mean PDC (0.9) and 29.4% in the cluster with a lower mean PDC (0.4). The corresponding figures for the 5-year follow-up were 36.9% (higher mean PDC [0.9]) and 63.1% (lower mean PDC [0.3]). Clusters with higher mean trajectories of PDC included more men, older individuals, patients using drugs from only one antidiabetic medication class, and noninsulin users during follow-up. Mean trajectories of adherence decreased mainly during the first year. This study identified a substantial problem with longitudinal adherence to any antidiabetic medication, with a low proportion of individuals clustered as having higher adherence during the 5-year follow-up. Results suggest the need for interventions via follow-up strategies aiming at monitoring and improving continuous treatment management while considering tailored treatment strategies.