Volume 30 Issue 5, May 2024

Nature Medicine explores the latest translational and clinical research news, with clinical trials testing glycosylated autoantigens as immunotolerance therapies.

Quality assurance labs should independently test AI health tools.

Automated insulin delivery systems have transformed the care of people living with type 1 diabetes and continue to move closer to the ultimate goal of a fully autonomous artificial pancreas.

The European Union’s new AI Act focuses on risk without considering benefits, which could hinder the development of new technology while failing to protect the public.

Scientific and medical conferences should not be held in states that ban abortion, as such bans put the lives of women at risk.

New data confirm that APOE4 homozygosity is a major genetic cause of Alzheimer’s disease, warranting the development of specialized research strategies, treatment approaches and clinical trials.

A strategy that controls confounders in quantitative microbiome data challenges the validity of previously reported microbial markers in colorectal cancer and serves as a wake-up call for the microbiome research field.

Vision–language models can be trained to read cardiac ultrasound images with implications for improving clinical workflows, but additional development and validation will be required before such models can replace humans.

A large study of older adults in China points to physical and cognitive function — not age — as key predictors of heat-related mortality, highlighting the need for climate adaptation policies to prioritize accessibility across all age groups.

Using single-cell RNA and T cell receptor sequencing along with microscopy, we identified the cell types and genes associated with immune checkpoint inhibitor therapy-related colitis. Our study will help to identify targets for early diagnosis and lays the groundwork for the development of safer immunotherapy regimens.

Exome-sequencing analysis in a Chinese cohort of 1,578 children with cerebral palsy established a genetic etiology in 387 of the patients (24.5%). Children with cerebral palsy for whom perinatal asphyxia had been recorded at birth were found to be more likely to carry pathogenic or potentially pathogenic variants.

Dapagliflozin improved a hierarchical composite outcome, including death, a worsening kidney disease event, and estimated glomerular filtration rate slope, compared with placebo, in patients with heart failure. This hierarchical outcome — analyzed with win statistics — might provide the statistical power to evaluate the effect of treatments on kidney function in heart failure trials.

QR4 is a new cardiovascular disease (CVD) risk score developed and evaluated in 16.9 million people that has better performance than other commonly used CVD risk scores. It includes nine new risk factors associated with increased risk of developing CVD (for example, a heart attack or stroke) over the next 10 years.

This Review outlines the state of the art of artificial intelligence in the preoperative, intraoperative and postoperative settings, where it has enormous potential to improve patient outcomes, surgical education and system efficiencies.

AI-enabled wellness apps exist in a regulatory gray area and may pose risks if used to manage mental health issues; this Perspective outlines the possible risks to users and the implications for app developers and regulators.

Externally applied electrical stimulation over the cervical spinal cord improves arm and hand functions in people with chronic tetraplegia due to spinal cord injury.

The study on APOE4 homozygosity indicates a genetic variant of Alzheimer’s disease with early symptom onset and distinct biomarker progression, highlighting the need for specialized treatment approaches.

The posterior hippocampal tail is an important seizure node in humans and may be a promising interventional target in epilepsy.

An antibody screen of two distinct multiple sclerosis cohorts reveals an autoantibody signature that is detectable years before symptom onset and linked to a common microbial motif.

Developed on cytology images of hydrothorax and ascites from 57,220 cases at four hospitals, a deep-learning model shows high accuracy in tumor origin prediction and presents prognostic value when patient treatment is consistent with the cancer origin predicted by the model.

In an interim analysis of an ongoing phase 1 trial of CAR T cells targeting EGFR and IL13Ra2 in patients with multifocal, recurrent glioblastoma, intrathecal delivery is feasible and well tolerated, with some reductions seen in tumor size.

In a phase 2 trial, nivolumab achieved a response rate of 58% in patients with mismatch-repair-deficient gynecological cancers, meeting the primary endpoint, and genomic and immunologic features correlated with response.

Controlling for confounders calls into question the robustness of some associations of microbiota with colorectal cancer stages.

Single-cell multi-omic analysis of 300,000 cells from 29 patients representing peripheral immune cells and colon mucosal immune, epithelial and mesenchymal cells reveals crosstalk between circulating and tissue-resident immune cells with epithelial cells in checkpoint inhibitor colitis and identifies potential therapeutic targets.

A self-amplifying mRNA vaccine shows promise in this new modality by eliciting neutralizing antibodies against the SARS-CoV-2 Omicron (BA.1) variant in a phase 2/3 trial.

The dynamics and durability of immune responses associated with protection against symptomatic infection in children offer insights to guide vaccination policies in pediatric populations.

Genome-wide analyses of vaccine antibody responses in 2,499 infants from Uganda, South Africa and Burkina Faso identify associations between specific HLA genes and response to eight vaccines, providing insights that could be considered for population-adjusted vaccine design strategies.

Using exome sequencing data from one of the largest cohorts of children with cerebral palsy, the genetic diagnostic rates of single-nucleotide and copy number variants were assessed and a sizeable fraction found to be clinically actionable.

Interim results from the low-dose cohort of the PROCLAIM trial show that AAV-based delivery of the progranulin gene in patients with frontotemporal dementia was safe and was associated with increased CSF progranulin.

In a large multinational cohort study, maternal, gestational or pregestational diabetes was associated with only a small-to-moderate risk of ADHD in offspring, contrary to previous estimates that showed stronger effect sizes, attributing the differences in findings to confounding by shared genetic and familial factors.

The analysis of continuous glucose monitoring measurements from a large cohort of nondiabetic individuals uncovered large inter- and intraindividual variabilities, with potential implications for current diagnostic cutoffs for diabetes diagnosis and several cardiometabolic clinical measures.

Post hoc analysis of the DAPA-HF and DELIVER trials reports on the approach of win statistics to evaluate the effect of dapagliflozin on a hierarchical composite kidney outcome in patients with heart failure.

The QR4 algorithm for prediction of 10-year cardiovascular disease risk, developed, tested and externally validated in datasets comprising 16.8 million people from the United Kingdom, improves upon the QRISK3 algorithm that is in current use by incorporating new risk factors.

Multi-omics profiling of the blood and heart of two human decedents receiving pig heart xenografts, including single-cell studies, reveals early immune responses and perioperative cardiac xenograft dysfunction in one of the two decedents, which may be due to mismatched heart size and/or insufficient immunosuppression.

In a randomized clinical trial, alerts based on the detection of abnormalities in electrocardiograms using a deep learning algorithm reduced all-cause mortality at 90 days in patients admitted to hospital emergency or internal medicine departments.

A two-step, video-based deep learning model is developed to first screen for cardiac anomalies using noncontrast magnetic resonance imaging, followed by diagnosis of 11 types of cardiovascular disease using gadolinium enhancement-based imaging.

A vision–language foundation model, trained on a dataset of more than 1 million echocardiogram video–text pairs, is able to assess various cardiac structural and functional parameters despite not having been directly trained on any specific image interpretation task.

In a prospective cohort of Chinese participants aged 65 years and older, heatwaves doubled the risk of mortality, especially in adults with functional impairments and dependency on daily living activities.