Mazdutide is a once-weekly glucagon and glucagon-like peptide 1 receptor dual agonist developed for the treatment of type 2 diabetes (T2D)1. Here we report on a randomized phase III trial assessing the efficacy and safety of mazdutide, compared with dulaglutide, in participants with T2D who were also treated with background oral anti-diabetic drugs. In this study, 731 participants with T2D were randomized 1:1:1 to receive 4 mg mazdutide, 6 mg mazdutide or 1.5 mg dulaglutide for 28 weeks. Both doses of mazdutide showed non-inferiority and superiority to the 1.5-mg dose of dulaglutide in terms of the mean change in the diagnostic marker glycated haemoglobin A1c (HbA1c) from baseline to week 28, with a least-squares mean treatment difference of -0.24% (P = 0.0032) for 4 mg mazdutide and -0.30% (P = 0.0003) for 6 mg mazdutide, relative to 1.5 mg dulaglutide. Significantly greater reductions in body weight were achieved with mazdutide than with dulaglutide, with a least-squares mean treatment difference of -3.78% for 4 mg mazdutide and -5.76% for 6 mg mazdutide (both P < 0.0001), relative to dulaglutide. Moreover, significantly more participants who received mazdutide 4 mg or 6 mg reached the composite end point of HbA1c < 7.0% with a body-weight reduction of at least 5% at week 28 (both P < 0.0001), compared with those who received dulaglutide. The most common treatment-emergent adverse events were diarrhoea, nausea and vomiting. In summary, we found that in Chinese participants with T2D, 28 weeks of treatment with mazdutide (4 mg and 6 mg) provided reductions in HbA1c and body weight that were superior to those attained with 1.5 mg dulaglutide. Mazdutide was generally safe, although the incidence of gastrointestinal adverse events was higher for mazdutide than for dulaglutide.

Despite advances in type 2 diabetes (T2D) management, unmet needs remain for therapies that effectively control hyperglycaemia while addressing comorbid metabolic disorders1,2. Here we assessed the efficacy and safety of the dual glucagon receptor (GCGR)/glucagon-like peptide-1 receptor (GLP-1R) agonist mazdutide monotherapy versus placebo in Chinese adults with T2D controlled inadequately with diet and exercise alone. In this phase 3 trial, 320 participants (mean glycated haemoglobin A1c (HbA1c) of 8.24%, body mass index of 28.2 kg m-2 and diabetes duration of 1.9 years) were randomized 1:1:1 to receive weekly subcutaneous injections of mazdutide (4 mg or 6 mg) or placebo for 24 weeks, followed by a 24-week extended mazdutide treatment. At week 24, mazdutide significantly reduced HbA1c versus placebo (primary endpoint): -1.57% with mazdutide 4 mg and -2.15% with mazdutide 6 mg, versus -0.14% with placebo, with treatment differences of -1.43% and -2.02% (both P < 0.0001). Weight loss from baseline at week 24 occurred with -5.61% (4 mg) and -7.81% (6 mg) versus -1.26% (placebo) (both P < 0.0001). Furthermore, more participants with mazdutide achieved HbA1c < 7.0%, weight loss ≥ 5% (all P < 0.0001) and composite endpoints (HbA1c < 7.0% and weight loss ≥ 5%) versus placebo (P = 0.0006 for 4 mg; P < 0.0001 for 6 mg) at week 24. The most common adverse events-diarrhoea, decreased appetite and nausea-were consistent with GLP-1R agonists. These results establish mazdutide monotherapy as an effective intervention providing clinically meaningful glycaemic control and weight reduction alongside a favourable safety profile in this population.

Decades of research have demonstrated that recovery from serious neurological injury will require synergistic therapeutic approaches. Rewiring spared neural circuits after injury is a long-standing goal of neurorehabilitation1,2. We hypothesized that combining intensive, progressive, task-focused training with real-time closed-loop vagus nerve stimulation (CLV) to enhance synaptic plasticity3 could increase strength, expand range of motion and improve hand function in people with chronic, incomplete cervical spinal cord injury. Here we report the results from a prospective, double-blinded, sham-controlled, randomized study combining gamified physical therapy using force and motion sensors to deliver sham or active CLV (ClinicalTrials.gov identifier NCT04288245). After 12 weeks of therapy composed of a miniaturized implant selectively activating the vagus nerve on successful movements, 19 people exhibited a significant beneficial effect on arm and hand strength and the ability to perform activities of daily living. CLV represents a promising therapeutic avenue for people with chronic, incomplete cervical spinal cord injury.

At the heart of medicine lies physician-patient dialogue, where skillful history-taking enables effective diagnosis, management and enduring trust1,2. Artificial intelligence (AI) systems capable of diagnostic dialogue could increase accessibility and quality of care. However, approximating clinicians' expertise is an outstanding challenge. Here we introduce AMIE (Articulate Medical Intelligence Explorer), a large language model (LLM)-based AI system optimized for diagnostic dialogue. AMIE uses a self-play-based3 simulated environment with automated feedback for scaling learning across disease conditions, specialties and contexts. We designed a framework for evaluating clinically meaningful axes of performance, including history-taking, diagnostic accuracy, management, communication skills and empathy. We compared AMIE's performance to that of primary care physicians in a randomized, double-blind crossover study of text-based consultations with validated patient-actors similar to objective structured clinical examination4,5. The study included 159 case scenarios from providers in Canada, the United Kingdom and India, 20 primary care physicians compared to AMIE, and evaluations by specialist physicians and patient-actors. AMIE demonstrated greater diagnostic accuracy and superior performance on 30 out of 32 axes according to the specialist physicians and 25 out of 26 axes according to the patient-actors. Our research has several limitations and should be interpreted with caution. Clinicians used synchronous text chat, which permits large-scale LLM-patient interactions, but this is unfamiliar in clinical practice. While further research is required before AMIE could be translated to real-world settings, the results represent a milestone towards conversational diagnostic AI.

Cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6i) in combination with endocrine therapy improve the outcomes of patients with hormone-receptor (HR)-positive, HER2-negative advanced breast cancer and can be used early as first-line treatment or deferred to second-line treatment1-7. Randomized data comparing the use of CDK4/6i in the first- and second-line setting are lacking. The phase 3 SONIA trial (NCT03425838) randomized 1,050 patients who had not received previous therapy for advanced breast cancer to receive CDK4/6i in the first- or second-line setting8. All of the patients received the same endocrine therapy, consisting of an aromatase inhibitor for first-line treatment and fulvestrant for second-line treatment. The primary end point was defined as the time from randomization to disease progression after second-line treatment (progression-free survival 2 (PFS2)). We observed no statistically significant benefit for the use of CDK4/6i as a first-line compared with second-line treatment (median, 31.0 versus 26.8 months, respectively; hazard ratio = 0.87; 95% confidence interval = 0.74-1.03; P = 0.10). The health-related quality of life was similar in both groups. First-line CDK4/6i use was associated with a longer CDK4/6i treatment duration compared with second-line use (median CDK4/6i treatment duration of 24.6 versus 8.1 months, respectively) and more grade ≥3 adverse events (2,763 versus 1,591, respectively). These data challenge the need for first-line use of a CDK4/6i in all patients.

Working from home has become standard for employees with a university degree. The most common scheme, which has been adopted by around 100 million employees in Europe and North America, is a hybrid schedule, in which individuals spend a mix of days at home and at work each week1,2. However, the effects of hybrid working on employees and firms have been debated, and some executives argue that it damages productivity, innovation and career development3-5. Here we ran a six-month randomized control trial investigating the effects of hybrid working from home on 1,612 employees in a Chinese technology company in 2021-2022. We found that hybrid working improved job satisfaction and reduced quit rates by one-third. The reduction in quit rates was significant for non-managers, female employees and those with long commutes. Null equivalence tests showed that hybrid working did not affect performance grades over the next two years of reviews. We found no evidence for a difference in promotions over the next two years overall, or for any major employee subgroup. Finally, null equivalence tests showed that hybrid working had no effect on the lines of code written by computer-engineer employees. We also found that the 395 managers in the experiment revised their surveyed views about the effect of hybrid working on productivity, from a perceived negative effect (-2.6% on average) before the experiment to a perceived positive one (+1.0%) after the experiment. These results indicate that a hybrid schedule with two days a week working from home does not damage performance.

PARTNER is a prospective, phase II-III, randomized controlled clinical trial that recruited patients with triple-negative breast cancer1,2, who were germline BRCA1 and BRCA2 wild type3. Here we report the results of the trial. Patients (n = 559) were randomized on a 1:1 basis to receive neoadjuvant carboplatin-paclitaxel with or without 150 mg olaparib twice daily, on days 3 to 14, of each of four cycles (gap schedule olaparib, research arm) followed by three cycles of anthracycline-based chemotherapy before surgery. The primary end point was pathologic complete response (pCR)4, and secondary end points included event-free survival (EFS) and overall survival (OS)5. pCR was achieved in 51% of patients in the research arm and 52% in the control arm (P = 0.753). Estimated EFS at 36 months in the research and control arms was 80% and 79% (log-rank P > 0.9), respectively; OS was 90% and 87.2% (log-rank P = 0.8), respectively. In patients with pCR, estimated EFS at 36 months was 90%, and in those with non-pCR it was 70% (log-rank P < 0.001), and OS was 96% and 83% (log-rank P < 0.001), respectively. Neoadjuvant olaparib did not improve pCR rates, EFS or OS when added to carboplatin-paclitaxel and anthracycline-based chemotherapy in patients with triple-negative breast cancer who were germline BRCA1 and BRCA2 wild type. ClinicalTrials.gov ID: NCT03150576 .

Less than 30% of people in Africa received a dose of the COVID-19 vaccine even 18 months after vaccine development1. Here, motivated by the observation that residents of remote, rural areas of Sierra Leone faced severe access difficulties2, we conducted an intervention with last-mile delivery of doses and health professionals to the most inaccessible areas, along with community mobilization. A cluster randomized controlled trial in 150 communities showed that this intervention with mobile vaccination teams increased the immunization rate by about 26 percentage points within 48-72 h. Moreover, auxiliary populations visited our community vaccination points, which more than doubled the number of inoculations administered. The additional people vaccinated per intervention site translated to an implementation cost of US $33 per person vaccinated. Transportation to reach remote villages accounted for a large share of total intervention costs. Therefore, bundling multiple maternal and child health interventions in the same visit would further reduce costs per person treated. Current research on vaccine delivery maintains a large focus on individual behavioural issues such as hesitancy. Our study demonstrates that prioritizing mobile services to overcome access difficulties faced by remote populations in developing countries can generate increased returns in terms of uptake of health services3.

Evidence is accumulating that perturbed postnatal development of the gut microbiome contributes to childhood malnutrition1-4. Here we analyse biospecimens from a randomized, controlled trial of a microbiome-directed complementary food (MDCF-2) that produced superior rates of weight gain compared with a calorically more dense conventional ready-to-use supplementary food in 12-18-month-old Bangladeshi children with moderate acute malnutrition4. We reconstructed 1,000 bacterial genomes (metagenome-assembled genomes (MAGs)) from the faecal microbiomes of trial participants, identified 75 MAGs of which the abundances were positively associated with ponderal growth (change in weight-for-length Z score (WLZ)), characterized changes in MAG gene expression as a function of treatment type and WLZ response, and quantified carbohydrate structures in MDCF-2 and faeces. The results reveal that two Prevotella copri MAGs that are positively associated with WLZ are the principal contributors to MDCF-2-induced expression of metabolic pathways involved in utilizing the component glycans of MDCF-2. The predicted specificities of carbohydrate-active enzymes expressed by their polysaccharide-utilization loci are correlated with (1) the in vitro growth of Bangladeshi P. copri strains, possessing varying degrees of polysaccharide-utilization loci and genomic conservation with these MAGs, in defined medium containing different purified glycans representative of those in MDCF-2, and (2) the levels of faecal carbohydrate structures in the trial participants. These associations suggest that identifying bioactive glycan structures in MDCFs metabolized by growth-associated bacterial taxa will help to guide recommendations about their use in children with acute malnutrition and enable the development of additional formulations.

Many communities in low- and middle-income countries globally lack sustainable, cost-effective and mutually beneficial solutions for infectious disease, food, water and poverty challenges, despite their inherent interdependence1-7. Here we provide support for the hypothesis that agricultural development and fertilizer use in West Africa increase the burden of the parasitic disease schistosomiasis by fuelling the growth of submerged aquatic vegetation that chokes out water access points and serves as habitat for freshwater snails that transmit Schistosoma parasites to more than 200 million people globally8-10. In a cluster randomized controlled trial (ClinicalTrials.gov: NCT03187366) in which we removed invasive submerged vegetation from water points at 8 of 16 villages (that is, clusters), control sites had 1.46 times higher intestinal Schistosoma infection rates in schoolchildren and lower open water access than removal sites. Vegetation removal did not have any detectable long-term adverse effects on local water quality or freshwater biodiversity. In feeding trials, the removed vegetation was as effective as traditional livestock feed but 41 to 179 times cheaper and converting the vegetation to compost provided private crop production and total (public health plus crop production benefits) benefit-to-cost ratios as high as 4.0 and 8.8, respectively. Thus, the approach yielded an economic incentive-with important public health co-benefits-to maintain cleared waterways and return nutrients captured in aquatic plants back to agriculture with promise of breaking poverty-disease traps. To facilitate targeting and scaling of the intervention, we lay the foundation for using remote sensing technology to detect snail habitats. By offering a rare, profitable, win-win approach to addressing food and water access, poverty alleviation, infectious disease control and environmental sustainability, we hope to inspire the interdisciplinary search for planetary health solutions11 to the many and formidable, co-dependent global grand challenges of the twenty-first century.

Artificial intelligence (AI) has been developed for echocardiography1-3, although it has not yet been tested with blinding and randomization. Here we designed a blinded, randomized non-inferiority clinical trial (ClinicalTrials.gov ID: NCT05140642; no outside funding) of AI versus sonographer initial assessment of left ventricular ejection fraction (LVEF) to evaluate the impact of AI in the interpretation workflow. The primary end point was the change in the LVEF between initial AI or sonographer assessment and final cardiologist assessment, evaluated by the proportion of studies with substantial change (more than 5% change). From 3,769 echocardiographic studies screened, 274 studies were excluded owing to poor image quality. The proportion of studies substantially changed was 16.8% in the AI group and 27.2% in the sonographer group (difference of -10.4%, 95% confidence interval: -13.2% to -7.7%, P < 0.001 for non-inferiority, P < 0.001 for superiority). The mean absolute difference between final cardiologist assessment and independent previous cardiologist assessment was 6.29% in the AI group and 7.23% in the sonographer group (difference of -0.96%, 95% confidence interval: -1.34% to -0.54%, P < 0.001 for superiority). The AI-guided workflow saved time for both sonographers and cardiologists, and cardiologists were not able to distinguish between the initial assessments by AI versus the sonographer (blinding index of 0.088). For patients undergoing echocardiographic quantification of cardiac function, initial assessment of LVEF by AI was non-inferior to assessment by sonographers.