Nature Medicine最新文献

Global healthcare providers are exploring the use of large language models (LLMs) to provide medical advice to the public. LLMs now achieve nearly perfect scores on medical licensing exams, but this does not necessarily translate to accurate performance in real-world settings. We tested whether LLMs can assist members of the public in identifying underlying conditions and choosing a course of action (disposition) in ten medical scenarios in a controlled study with 1,298 participants. Participants were randomly assigned to receive assistance from an LLM (GPT-4o, Llama 3, Command R+) or a source of their choice (control). Tested alone, LLMs complete the scenarios accurately, correctly identifying conditions in 94.9% of cases and disposition in 56.3% on average. However, participants using the same LLMs identified relevant conditions in fewer than 34.5% of cases and disposition in fewer than 44.2%, both no better than the control group. We identify user interactions as a challenge to the deployment of LLMs for medical advice. Standard benchmarks for medical knowledge and simulated patient interactions do not predict the failures we find with human participants. Moving forward, we recommend systematic human user testing to evaluate interactive capabilities before public deployments in healthcare.

Extracorporeal liver cross-circulation (ELC) using genetically modified pig livers may address an unmet need for temporary liver support in patients with acute or acute-on-chronic liver failure. This study used the ELC platform to evaluate early immune responses and assess xenogeneic liver physiological support in a human decedent model. Four human decedents underwent ELC using pig livers with a triple glycan knockout; insertion of seven human transgenes and inactivation of pig endogenous retroviruses. Intravenous methylprednisolone was administered for immunosuppression. In the case of decedents 1-3, ELC was performed for 72-84 h with the native livers of the decedents remaining in situ. In the case of decedent 4, hepatectomy was performed, followed by 48 h of xenogeneic liver support exclusively using ELC. Biopsies of xenogeneic livers demonstrated preserved parenchymal architecture, mild immune infiltration and IgM deposition. Xenogeneic livers produced bile and supplemented native hepatocellular function. In decedent 4, xenogeneic liver-only support after hepatectomy maintained hemodynamic stability, normal pH, lactate, ammonia, international normalized ratio and sustained metabolic function. This study shows that ELC is feasible using xenogeneic livers with minimal immunosuppression and can provide effective liver support.

Pulsed field ablation (PFA) has proven to be a safe and effective non-thermal ablation modality for the treatment of atrial fibrillation (AF), but little outcome data beyond 1 year has been reported. Here, we present results from the ADVENT-LTO study, which provides extended follow-up of the ADVENT trial, the first randomized trial comparing PFA with conventional thermal ablation. In ADVENT-LTO, 364 paroxysmal AF patients (183 PFA, 181 thermal; 237 men, 127 women) participated, and were followed for 1,332±147 days. For the primary endpoint of four-year treatment success, PFA demonstrated preserved effectiveness compared to thermal ablation (72.8% PFA, 64.1% thermal; P=0.12). Moreover, there was a trend favoring PFA as compared to thermal ablation for the pre-specified outcome of freedom from hospital-based arrhythmia intervention (85.6% PFA, 78.6% thermal; HR 0.64, 95%CI 0.38, 1.05), including fewer repeat ablations (10.4% PFA, 17.7% thermal; P=0.04), as well as a trend favoring PFA as compared to thermal ablation for the pre-specified outcome of progression to persistent AF (2.6% PFA, 4.6% thermal; HR 0.55, 95%CI 0.16, 1.88). Taken together, these data demonstrate that the favorable outcomes of PFA are maintained over the course of four years. Coupled with the safety advantages of PFA over thermal ablation, these long-term data support widespread adoption of PFA for the treatment of AF. Clinical Trial Registration: NCT06526546.

Currently, no licensed vaccine is available against chikungunya virus (CHIKV) in children aged less than 12 years. In this phase 2, observer-blind, randomized, dose-response trial we evaluated the tolerability (solicited adverse events (AEs)), safety (unsolicited AEs) and immunogenicity of a live-attenuated CHIKV vaccine (VLA1553) in healthy children aged 1-11 years in endemic countries. Here we provide a prespecified interim analysis to 28 days after vaccination. Participants (n = 304) received either a half dose (n = 119) or a full dose (n = 124) of VLA1553, or active control meningococcal vaccine (Nimenrix) (n = 61). The primary endpoint was the incidence and severity of solicited AEs within 14 days after vaccination. Secondary endpoints included unsolicited AEs, medically attended AEs, serious AEs, AEs of special interest and immunogenicity through 28 days after vaccination. For the primary endpoint, there were no statistically significant differences between the VLA1553 groups and control for each age group nor between age groups (1-2, 3-6 and 7-11 years). Overall, the most frequently reported solicited injection site AEs were tenderness (10.9%) and pain (8.6% of participants); the most frequently reported solicited systemic AEs were fever (12.5%) and headache (11.5%; participants aged 7-11 years and 3-6 years only, (headache was not solicited in participants aged 1-2 years)). For the secondary safety endpoints, the most common unsolicited AEs were infections and infestations (10.0-20.5%) and the most common medically attended AE overall was fever (2.6%), with no differences for unsolicited AEs or medically attended AEs regardless of vaccine and age group (serious AEs and AEs of special interest were too infrequent for meaningful comparisons). Anti-CHIKV neutralizing antibody titers were higher in the full-dose than the half-dose group at days 14 and 28 after vaccination. The trial met its prespecified endpoints and supported the selection of full-dose VLA1553 in future clinical trials in this population, addressing a critical unmet medical need. ClinicalTrials.gov registration: NCT06106581 .

Classic psychedelics typically act at the serotonin 5-HT_2A receptor to profoundly alter brain function and consciousness. Research on these compounds has accelerated. Major strides have been made in understanding their unique mechanisms of action and clinical potential. This Review outlines the state of psychedelic science, spanning cellular mechanisms, systems neuroscience and clinical investigation. We show that preclinical and human research findings converge on two complementary processes: acute neural desynchronization, which destabilizes entrenched network patterns, and subacute neuroplasticity, which opens a window for psychological and behavioral change. We review evidence of therapeutic response across neuropsychiatric indications and consider how this integrates with mechanistic findings. We also explore challenges and opportunities, including discrepancies between preclinical evidence that non-hallucinogenic psychedelic analogs engage putative therapeutic mechanisms, and clinical evidence linking the subjective experience to therapeutic response; the risks inherent to enhanced neuroplasticity; and questions surrounding trial design, scalability and regulatory approval. The growth of psychedelic science and medicine may compel a fundamental rethinking of the relationship between subjective experience and biological change in psychiatry.

The scarcity of subspecialist medical expertise poses a considerable challenge for healthcare delivery. This issue is particularly acute in cardiology, where timely, accurate management determines outcomes. We explored the potential of Articulate Medical Intelligence Explorer (AMIE), a large language model-based experimental medical artificial intelligence system, to augment clinical decision-making in this challenging context. We conducted a randomized controlled trial comparing large language model-assisted care with the usual care of complex patients suspected of having a genetic cardiomyopathy, and we curated a real-world dataset of complex cases from a subspecialist cardiology practice. Nine participating general cardiologists were provided with access to both clinical text reports and raw diagnostic data-including electrocardiograms, echocardiograms, cardiac magnetic resonance imaging scans and cardiopulmonary exercise testing-and were randomized to manage these cases, either with or without assistance from AMIE. We developed a ten-domain evaluation rubric used by three blinded subspecialists to evaluate the quality of triage, diagnosis and management. In our randomized controlled trial with retrospective patient data, subspecialists favored large language model-assisted responses overall, and for the management plan and diagnostic testing domains, with the remaining domains considered a tie. Overall, subspecialists preferred AMIE-assisted cardiology assessments 46.7% of the time, compared with 32.7% for cardiologists alone (P = 0.02), with 20.6% rated as a tie. Subspecialists also quantified errors, extra and missing content, reasoning and potential bias. Cardiologists alone had more clinically significant errors (24.3% versus 13.1%, P = 0.033) and more missing content (37.4% versus 17.8%, P = 0.0021) than cardiologists assisted by AMIE. Lastly, cardiologists who used AMIE reported that AMIE helped their assessment more than half the time (57.0%) and saved time in 50.5% of cases.

Chimeric antigen receptor (CAR) T cell therapy was recently proposed as a treatment for adults with B-cell-mediated autoimmune diseases (ADs) refractory to conventional immunomodulatory therapy. We present a case series of eight children with severe/refractory AD (four systemic lupus erythematosus, three dermatomyositis, one systemic sclerosis) treated at Ospedale Pediatrico Bambino Gesù, Rome, and University Hospital Erlangen with a single infusion of 1 × 10⁶ kg^-1 point-of-care manufactured autologous CD19 CAR T cells (zorpocabtagene autoleucel), in a hospital exemption (HE) program. In Europe, the HE pathway offers the opportunity to treat patients with life-threatening or seriously debilitating disorders who lack valid therapeutic options, using an advanced therapy medicinal product (ATMP) authorized on a nonroutine, single-patient basis. In contrast to the 'compassionate use' pathway, the ATMP does not necessarily need to have undergone clinical trials or marketing authorization applications. Manufacturing was successful in all patients, yielding several drug product bags. Once infused after lymphodepletion, zorpocabtagene autoleucel cells expanded in vivo, promoting prompt B cell clearance. Grade 1 cytokine release syndrome was reported in six patients, and grade 1 immune effector cell-associated neurotoxicity syndrome was reported in one patient. Late-hematotoxicity was limited to grade 1 in two patients. All these adverse events were manageable and no severe infections occurred. With a median follow-up of 16.5 months (range = 9-24 months), all patients experienced a clinically substantial improvement/resolution of AD, as evidenced by reduction in disease activity scores and signs of reversal of organ damage. This improvement enabled sustained discontinuation of immunomodulators, even after B cell reconstitution. The activation of formal clinical trials enrolling children and adolescents is urgently needed to confirm these preliminary results and to assess the long-term safety of this approach.