For how we live.
Trials, journals, oncology, genomics, and the long work of proving things.
In 5,520 children across 34 states, prenatal exposure to regulated public-water fluoride showed no association with emotional or behavioural problems overall — with one caveat in a secondary model.
In the 172-patient PARADIGM trial, azacitidine plus venetoclax more than doubled median event-free survival versus induction chemotherapy — but it was a phase 2 study not powered for overall survival.
Stargardt disease, the most common inherited macular dystrophy, has no approved treatment. In the 50-patient TEASE-1 trial, gildeuretinol slowed atrophic lesion growth by a fraction of a millimetre a year.
In an early-phase trial of 136 previously treated patients, daraxonrasib produced responses in 31–37% depending on dose — but 54% had a serious side effect, and this is a phase 1–2 study.
In the NSABP B-59 trial, adding the immunotherapy to neoadjuvant chemotherapy did not significantly improve event-free survival, while immune-related side effects more than doubled.
In 53 patients, oral fluconazole — but not a mouth-only antifungal — cut gut Candida and coincided with improved disease activity. The study was observational, brief and unblinded.
Ervebo was licensed against Zaire ebolavirus. WHO says it is not known whether it protects against Bundibugyo virus in humans — which is why a trial is running alongside the wider rollout.
Lorecivivint improved pain modestly and changed nothing structural. Sprifermin grew cartilage and the effect made no difference by year five. Both illustrate the gap between a biomarker and a benefit.
The REACT study imaged people aged 18 to 70 with no known cardiovascular disease. Atherosclerosis was already present in some people in their twenties, and it was usually in the legs, not the heart.
Two pragmatic randomised trials and a Medicare nursing home cohort landed in the same week. Together they show real protection against hospitalisation and a durability question nobody has answered.
PRESC1SE-MI covered 67,624 chest-pain presentations at 19 hospitals. The one-hour troponin pathway matched the three-hour one on safety, and median length of stay was identical: 309 minutes.
Carbon filters and ion exchange concentrate PFAS rather than breaking it down, which turns treated water into a residual-waste problem. A new review puts numbers on what destruction would cost in energy.
A Chinese cohort found serum PCBs associated with a higher calculated 10-year cardiovascular risk. The longitudinal results did not survive correction for multiple testing, and no events were counted.
The beliefs — that Black patients' skin is thicker, their nerve endings less sensitive — predicted lower pain ratings and less accurate treatment recommendations for a Black patient in the same experiment.
A first-in-human trial of Kylo-11 reports 97% reduction at 48 weeks after a single dose. A lepodisiran analysis reports near-total clearance of the oxidised phospholipids Lp(a) carries.
Cochrane's 2024 review found low-certainty evidence of a small effect since fluoride toothpaste became widespread. The IQ meta-analysis that reopened the argument reported a null below 1.5 mg/L in one key analysis.
POET II randomised 508 stabilised patients to stop antibiotics or complete the standard course. The safety endpoint met noninferiority; relapse of infection did not go the same way.
The first double-blind sham-controlled trial of pulmonary vein isolation found quality of life improved in both arms. Most of the benefit patients feel appears not to come from the ablation.
One found specific phenanthrene metabolites associated with fetal growth restriction and chased the mechanism. The other tested a 24-metabolite mixture against ADHD in 3,962 children and found nothing.
The Fifth Universal Definition of Myocardial Infarction drops the numbered types, keeps sex-specific troponin thresholds, and proposes ICD-11 codes so the categories can be counted.
LIBREXIA-ACS enrolled 14,194 patients before stopping for futility. Milvexian did not reduce events, but intracranial and fatal bleeding were identical to placebo, which is the class's whole premise.
CARDIO-TTRansform randomised 1,432 patients with transthyretin amyloid cardiomyopathy to eplontersen or placebo for 140 weeks. The confidence interval around the primary result crosses one.
A network meta-analysis found route of delivery decides everything, and that autologous transplant ranked below placebo. In IBS, 13 trials produced very low certainty of any effect.
Two birth cohorts totalling 9,801 children were sorted into eight developmental patterns. The classic progression was one of them, and one of the smallest.
A $12 million gift will test whether low-dose rapamycin changes the transition to frailty in adults over 65 — the first study of the drug designed around an outcome rather than a biomarker.
Suppressor tRNAs have been a promising idea for nonsense mutations for years, held back by weak activity and delivery. A Science paper reports one chemical tweak and one redesigned nanoparticle.
The longest human trial of low-dose rapamycin for healthy ageing found no change in visceral fat, its own primary outcome. The positive findings it is quoted for rest on eight women.
Twelve studies in 1,158 IBS patients found gut-directed hypnotherapy improved symptoms, with the firmest results in high-volume and group delivery. The microbiome route is weaker.
In two phase 3 trials, 38.8% and 35.9% on baricitinib 4 mg reached near-complete scalp regrowth at 36 weeks against 6.2% and 3.3% on placebo — the first drug class to do this for the condition.
An investigation into the flood of papers mining one public health database found something more awkward than organised fraud: an open marketplace of tools that lets anyone produce a publishable analysis in an afternoon.
A mandatory citywide programme sequenced 782 genes in 1,000 prospective couples. The at-risk rate was double that of a comparable Australian study, and most flagged couples still chose to marry.
A review in Gut says no intestinal disease can be cured by normalising barrier function, and that it is unproven whether restoring the barrier improves any clinical outcome at all.
A 100-child randomized trial in Uganda found a cheap supplement lowered infection rates on top of standard hydroxyurea therapy — though the authors say it needs a bigger trial before it changes practice.
WHO data put the normal window for walking alone at 8.2 to 17.6 months. The 2022 CDC checklists moved to the age by which 75% of children reach a milestone — to flag real delay, not redefine normal.
The approved tests are for patients already showing cognitive decline. Used on the worried well, a positive result mostly signals a risk that may never become disease — and no treatment follows from it.
Moderna and Merck's personalized vaccine slowed melanoma's return when added to Keytruda. It is the first mRNA cancer vaccine to succeed at scale — and the first real test of a decade-old idea.
Three of the four Yamanaka factors, delivered to one eye and switched on with an antibiotic for eight weeks. The mouse data are striking; the trial is a safety study and will not measure ageing.
Chronic granulomatous disease is usually X-linked in Western cohorts. A multicentre Middle East and North Africa registry found the opposite pattern, alongside late diagnosis and a median survival of 8.5 years.
VPM1002 did not show non-inferiority to BCG for preventing tuberculosis infection in a phase 3 trial across sub-Saharan Africa. Far fewer infections occurred than planned, leaving the comparison unresolved.
Farm-raised children really are protected from asthma, and the mechanism has been traced to microbial exposure and innate immunity. What has not been shown is that household hygiene caused the allergy epidemic.
Creatinine is muscle waste your kidneys clear; eGFR is an estimate of filtering capacity built from it. A single reading is a snapshot, not a diagnosis — and the equations have limits worth knowing.
Two national studies settle the question the trials could not: in England, cancer rates fell 87% among those vaccinated at 12-13; in Sweden, the risk was cut by nearly two-thirds.
A Brazilian laboratory study found both a prototype strain and an emergent reassortant killed neural progenitors and collapsed cortical structure. This is a dish, not a pregnancy.
Longevity trials want a measurement that moves in months instead of decades. A Berlin study ranked the candidates, and a four-week diet trial showed what a clock does when nothing much happens.
Analgesic trials keep failing, and one reason is that the comparison arm keeps improving. Where a trial recruits, how long it runs and how big it is all move the placebo response — and none of that is about the drug.
A systematic review in Cell called that rate implausible and said it overstates the gut microbiome's role in human disease. A separate audit found the statistical tools themselves disagree.
Two datasets report an age-independent fall in men's testosterone across birth cohorts. A separate study of nine thousand men found much of the variation between cohorts came from the tests themselves.
Raised liver enzymes are common and mean the liver may be under strain, not that a disease is confirmed. The pattern, degree and duration of the change are what a clinician reads, not a single flag.
A 223-study meta-regression reports sperm concentration falling 51.6% since 1973. Two published critiques argue the method cannot support that reading. Both sides are peer-reviewed.
Longitudinal data and a meta-analysis of 159,425 children find consistent associations between screen time and development — but the effects are small, observational, and thinner than the guidelines imply.
Blood samples from a West African vaccine trial show both licensed Ebola vaccines generate antibodies that bind Bundibugyo virus — the species with no vaccine of its own, now causing DRC's largest outbreak on record.
Peruvian Andeans are famous for high haemoglobin at altitude. A genome-wide scan finds the alleles under recent positive selection are the ones that lower it — a pattern closer to Tibetans than expected.
Gene-edited islet cells implanted in a person's forearm are still functioning more than a year later. It is a single case report, and will need many more patients to mean anything.
Two 2026 reviews make the case that human-derived models can replace animal testing. A third paper argues the field is checking the biology and the algorithm separately, and calling that assurance.
Years of observational data suggested 4CMenB cut gonorrhea risk by a third or more. A placebo-controlled trial in 654 people at high risk found essentially no effect at all.
Nous-209 targets 209 mutations shared across a whole class of tumours, so it can be manufactured in advance. What the trial has not yet shown is that anyone got less cancer.
In a 187-patient trial, chronic graft-versus-host-disease-free survival at one year was 78% with the new therapy versus 38% with conventional transplant. The FDA approved it 30 June.
A sham-controlled trial found FMT shifted bacterial composition but cleared drug-resistant organisms no better than a saline placebo. A second trial shows what a defined replacement product can and cannot yet prove.
A New England Journal of Medicine review of the ongoing outbreak lays out what is and isn't known about treating Bundibugyo virus disease — and why Ebola virus countermeasures may offer partial cover.
University of Florida researchers identify excess glycan production as a driver of Alzheimer's pathology, and find oral glucosamine worsened cognition in mice and tracked with faster decline.
Data from 20 European countries covering 2014 to 2024 tracks lung function, nutrition and infection through the arrival of triple modulator therapy — and the new complications arriving with age.
Thirty-five people received VERVE-102 in a phase 1 trial. The edit is permanent, the follow-up is short, and the endpoint measured was cholesterol, not heart attacks.
35% of patients on tezepelumab stopped oral steroids entirely, versus 21% on placebo, without losing asthma control — in a drug that targets inflammation upstream of the eosinophil count doctors usually check.
A modelling analysis puts 2.75 million children under five in sub-Saharan Africa living with sickle cell disease. Seven African countries have transplant programmes; approved gene therapies cost millions.
An 8-year-old boy had brain inflammation with no respiratory symptoms. Every respiratory, blood and rectal sample tested negative; the virus was in his cerebrospinal fluid alone.
Baxdrostat blocks the enzyme that makes aldosterone, rather than blocking aldosterone's receptor. In a 794-patient trial it lowered systolic pressure another 10 points in people already on multiple drugs.
ETX101 doesn't replace the faulty gene behind Dravet syndrome — it turns up the healthy copy that's still there. New data from 21 children show the effect holding for a year.
A linked-registry cohort of New South Wales residents from 2001 to 2018 finds cancer risk in people with intellectual disability inverts with age — higher when young, lower after 50.
Buntanetap showed dose-dependent cognitive benefit in biomarker-positive patients with mild disease. The company's announcement does not say what happened in the full 351-patient population.
The LiverScreen project ran transient elastography across nine countries, then sent everyone who screened positive to a hepatologist. The gap between the two numbers is the interesting part.
Of 1,363 evaluable patients matched to drugs by tumor genetics, 7% became exceptional responders. Activity across all 37 drugs was modest, and severe side effects hit over a quarter.
27 of 28 patients had no severe pain crises after treatment with reni-cel. But the sponsor ended the study early, and the reported analysis was not the one it was designed to test.
Results posted to ClinicalTrials.gov show 15- and 25-day nirmatrelvir-ritonavir regimens produced no meaningful edge over placebo on cognitive, autonomic, or exercise-intolerance symptoms.
In 2,077 children, higher processed-food intake at age 3 tracked with worse behavioral and emotional scores at age 5. It's an association from a birth cohort, not a trial, and can't establish that the food caused it.
The STABLED trial randomized 331 atrial fibrillation patients with a recent stroke to ablation plus anticoagulation or anticoagulation alone. The two groups ended up nearly identical.
Sequencing 6,141 people turned high consanguinity into a research advantage: it produced natural human knockouts that population studies elsewhere almost never see.
Researchers profiled nearly 356,000 nuclei from post-mortem hippocampi across five groups. Chromatin accessibility in neurogenic cells was already altered in people with preclinical Alzheimer's pathology.
A cluster-randomised trial across 15 population clusters found mosquito abundance collapsed in intervention areas and protective efficacy of 71–72% against symptomatic dengue.
A study in patients with VITT maps the trigger to a specific epitope on an adenoviral core protein, and to a single amino acid change that redirects the antibody onto a human platelet protein.
A Swedish birth cohort of 2.76 million people found the cumulative male-to-female ratio for autism incidence had fallen to 1.2 by age 20 — and projects parity by 2024.
Nine children received a single infusion of AAV9 carrying the beta-galactosidase gene. Enzyme activity rose and substrate fell in every participant; some developmental measures held steady and others declined.
Trained on 2,202 retracted paper mill articles, a text classifier flagged 9.87 percent of the cancer literature — including in the highest-impact journals. Flagged is not the same as fraudulent.
AlphaGenome takes a million base pairs of DNA and predicts what the sequence does. In Nature on January 28, it matched or beat the best existing models in 25 of 26 variant-effect evaluations.
Two notices, one day apart, end one line of federally funded research and freeze another pending review. NIH is now asking scientists which experiments could not be done without embryonic stem cells.
FDA approved copper histidinate on the strength of two single-arm trials compared with an external control group. Children treated within four weeks of birth had a 78 percent lower risk of death.
The agency says it has been relaxing chemistry and manufacturing expectations for cell and gene therapies case by case. The change is that every sponsor now knows it.
Narsoplimab clears the FDA on a single-arm study of 28 adults plus 19 patients from an expanded access programme. Sixty-one per cent responded. Seven per cent had a fatal adverse reaction.
The NEJM paper reports 47 children in an open-label trial with no randomised control. The biomarker effect is large; the clinical claim rests on softer ground.
The biomarker effect reported at CTAD is large and sustained over two years. The sample is seven, the design is open dose-escalation, and imaging was available for three.
At a Mayo Clinic cohort of 252 patients, 40% of amyloid-negative people tested positive. The manufacturer says it has found a manufacturing problem and put the product on hold.
Shared cell time raises transmission odds sharply — six months of it doubles them. But cell-level exposure explains only part of the genomic clustering, which undercuts contact tracing as a strategy.
A Buenos Aires surveillance cohort estimates 80.8% effectiveness against RSV hospitalisation in infants under three months — close to what the licensing trial found, in a real programme.
An 18,476-person phase 3 trial and a small human challenge study landed the same day. Both favour the mRNA vaccine against influenza A; neither settles what it does against influenza B.
An individual-patient meta-analysis of 17,801 patients found a hazard ratio of 0.97 for death, reinfarction or heart failure — a result precise enough to make the question look settled.
VESALIUS-CV randomised 12,257 patients with atherosclerosis or diabetes but no prior MI or stroke. Evolocumab reduced the three-point event rate from 8.0% to 6.2% over a median 4.6 years.
The first phase 1 results for CTX310 show gene editing can switch off a lipid gene in the liver. The trial was designed to look for harm, not benefit, and one participant died suddenly.
PISCES randomised 1,228 haemodialysis patients in Canada and Australia. The headline hazard ratio of 0.57 counts every event, not just the first — and that choice explains part of the gap.
The experimental recombinant antivenom outperformed a plasma-derived product in mice and cut tissue necrosis. It has not been tested in a single human, and the paper does not address cost or manufacturing at scale.
The Nature paper pairs mouse mechanism with retrospective human cohorts: vaccination within 100 days of starting a checkpoint inhibitor was associated with better survival. Retrospective is the load-bearing word.
Two Nature papers describe the Taiwan Precision Medicine Initiative and the risk scores built from it. The genetic effects identified explained up to 10.3% of health variation in the cohort.
Two Nature papers apply duplex sequencing with fewer than five errors per billion bases to cheek swabs and sperm — finding 46 genes under selection in mouth lining and a fatherhood risk that rises with age.
Three laureates, three papers, and a 30-year gap between the first observation and the therapies it enabled. The prize rewards work that was against the consensus when it was published.
Researchers showed open-source protein design software could produce variants of proteins of concern that biosecurity screening tools missed, then wrote and deployed patches before publishing.
Thirty-seven had the necrotising form, of whom 41% died. Just 16% of vaccine-eligible children with encephalopathy had received that season's flu vaccine. There is no national surveillance for the condition.
A 13-year-old observational hypothesis finally got its randomised test. In ALASCCA, 160 mg of daily aspirin cut three-year recurrence from 14.1% to 7.7% in biomarker-selected patients.
BCG and oral polio vaccine given within 48 hours to 5,420 babies under 2,000 g reduced deaths from infections other than tuberculosis. The effect is off-target, and the mechanism is unresolved.
In a 400-pair trial in Uganda where every family already had a bed net, insecticide-treated wraps reduced clinical malaria in infants from 2.14 to 0.73 cases per 100 person-weeks.
Forzinity was approved under accelerated approval to improve muscle strength in patients weighing at least 30 kg. The confirmatory trial is not due to report until 2030.
In 99 patients with normal-pressure hydrocephalus, an open shunt valve improved gait velocity and balance versus a valve set so high it did nothing. Cognition and bladder control did not improve.
Brensocatib blocks an enzyme that arms neutrophils rather than killing bacteria. In a 1,721-patient trial it cut yearly flare-ups by about a fifth — a modest effect in a disease with nothing else approved.
A Nature paper maps the route cerebrospinal fluid takes to neck lymph nodes, finds the route degrades with age, and reports a mechanical device that reopened it in old mice.