This week in Neurology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

This week: a lower blood pressure target after ischemic stroke cuts major cardiovascular events, a new potassium channel opener sharply reduces focal seizures, and CAR T-cell therapy brings stiff person syndrome into immunotherapy-free remission.

Welcome to This Week in Neurology. We're covering 10 papers this week: 3 in depth, 4 in brief, 3 quick signals.

The New England Journal of Medicine published OPTIMAL-STROKE, an open-label randomised trial led by Sampaio Silva at 30 sites in Brazil [1]. It enrolled 4368 adults with elevated blood pressure and a previous ischemic stroke or transient ischemic attack. It compared a systolic target below 120 with a target below 140 millimetres of mercury, and the groups separated by about 13 millimetres, with the intensive group averaging around 122. Over a median of almost four years, the composite of cardiovascular death, nonfatal heart attack, nonfatal stroke, or hospitalisation for unstable angina or heart failure occurred in about 11 percent of the intensive group and about 13.5 percent of the standard group. That is a relative reduction of roughly a fifth. Serious adverse events overall were similar. The composite renal outcome, a sustained fall of at least half in kidney filtration or progression to end-stage disease, was more common with intensive treatment. The abstract does not report recurrent stroke or cognitive outcomes separately. The open-label design and a single-country population are the main limitations. This is randomised evidence in a stroke population specifically, and it supports a lower systolic target for secondary prevention, with a kidney cost alongside it.

JAMA Neurology published X-TOLE2, a phase 3 placebo-controlled trial of azetukalner, a once-daily KV7 potassium channel opener, led by French [2]. It randomised 374 adults with focal epilepsy at 138 sites across four continents. These patients were highly refractory: they had stopped a median of five earlier antiseizure medications, nearly nine in ten participants were taking two or three concomitant drugs, and the median baseline was about thirteen seizures a month. With no titration, over twelve double-blind weeks, median monthly seizure frequency fell by about 10 percent on placebo, about 35 percent on 15 milligrams, and just over half on 25 milligrams. Both doses were statistically significant against placebo. Tolerability was described as similar to the earlier phase 2b X-TOLE trial, though the abstract gives no adverse-event rates, and it does not report responder rates or seizure freedom. The main limitation is the short controlled period, so claims of durability rest on an open-label extension. As evidence, this is a second positive trial for a new mechanism in drug-resistant focal epilepsy, with a clear dose response.

Also in JAMA Neurology, Piquet and colleagues report a pivotal single-arm phase 2 trial of mivocabtagene autoleucel, a fully human CD19 CAR T-cell therapy, in antibody-positive stiff person syndrome with inadequate response to immunotherapy [3]. Twenty-six adults at three sites in the United States received low-dose lymphodepletion and a single infusion. At week 16, the timed 25-foot walk improved by a median of almost five seconds, close to half of the baseline time, and the gain held in the sixteen patients who reached 24 weeks. Disability, stiffness and heightened sensitivity scores all improved. Cytokine release syndrome occurred in almost every patient, all grade 1 or 2, and neurotoxicity in three patients, all grade 1, with everything resolving. Neutropenia was the most common severe event, in five patients. At last follow-up, every patient was off immunomodulatory therapy for the condition. The central limitations are the absence of a control group, an unblinded walking endpoint in a disease with fluctuating symptoms, and a median follow-up of only about six months. This is the first prospective trial evidence for CAR T-cell therapy in this disease, and durability is still unproven.

Now, four in brief.

In Neurology, Ingwersen and colleagues pooled 18 studies and just over 14,500 patients comparing direct transport to a comprehensive stroke center with transport to the nearest stroke center [4]. Only two of the studies were randomised trials, and risk of bias in the nonrandomised studies was high. Bypass brought groin puncture about an hour earlier and was associated with modestly more functional independence at 90 days, with no significant difference in rates of thrombectomy or thrombolysis. In the randomised trials, thrombolysis came about half an hour later with bypass. In intracerebral hemorrhage, bypass was associated with higher mortality. The authors conclude the evidence is insufficient to support routine bypass protocols.

Two papers this week concern demyelinating disease. In Neurology, Di Filippo and colleagues applied the 2024 McDonald revisions retrospectively to 236 people with radiologically isolated syndrome across 22 Italian centres [5]. At baseline, about two thirds of these people already met the new criteria for multiple sclerosis, driven largely by CSF positivity, since central vein sign data existed for only five people. Among the 42 people who later had a clinical event, close to nine in ten patients would have been diagnosed a median of two years earlier. The data are registry-based, with nonstandardised MRI, so they describe reclassification and not whether earlier diagnosis changes outcomes.

In the Multiple Sclerosis Journal, Langer-Gould and colleagues followed a population-based cohort of 69 people diagnosed with MOG antibody-associated disease at symptom onset between 2019 and 2024, close to half of them children [6]. Most patients received corticosteroids within about a week, and only 11 patients received empiric prophylaxis. Over a median of about two and a half years, just over four in five patients recovered fully. Relapse occurred in 10 percent of children and about 13 percent of adults, and no factor, including treatment, was significantly associated with relapse. The small cohort limits power to detect treatment effects, but the data point to a low relapse risk without routine prophylaxis.

In The Lancet Neurology, Barton and colleagues report a phase 2 trial of bepranemab, an antibody against the tau mid-region, in 466 people with prodromal-to-mild Alzheimer's disease across Europe, Canada and the United States [7]. Over 80 weeks, neither dose slowed decline on the Clinical Dementia Rating sum of boxes. The higher dose did numerically slightly worse than placebo, and neither difference was significant. Safety was acceptable, with no neuroinflammatory signals on MRI. The authors describe differences favouring the drug on tau PET and a cognitive subscale as hypothesis-generating only. The primary result is clearly negative for another tau antibody.

Three quick signals.

In Neurology, Piura and colleagues tested the Graus criteria for possible autoimmune encephalitis in 261 patients with suspected rapidly progressive dementia at two referral centres [8]. The criteria were about 89 percent sensitive and 96 percent specific. Among survivors with autoimmune encephalitis, about two thirds of patients had persistent cognitive impairment, linked to older age and slower treatment. This is Class III evidence from tertiary centres.

In Movement Disorders, Kuzkina and colleagues applied a skin alpha-synuclein seed amplification assay in 117 people [9]. It detected synucleinopathy with about 82 percent sensitivity and 90 percent specificity, and assay kinetics separated multiple system atrophy from Parkinson's disease in about four in five positive cases. The study was blinded but small and cross-sectional.

In Neurology, Melin and colleagues compared 111 suicides among Swedish people with epilepsy with 1,147 controls [10]. Alcohol and other substance abuse carried roughly elevenfold higher odds, well ahead of depression and anxiety. Frequent recent tonic-clonic seizures and medication nonadherence also raised risk, while treatment was associated with lower risk. The design is observational with modest case numbers.

Stepping back across this week's papers, the firmest evidence came from large randomised trials with hard or patient-centred endpoints, while several consequential diagnostic and therapeutic questions rest on single-arm, retrospective or referral-centre data. New criteria and biomarkers are moving diagnosis earlier, yet none of these studies shows that earlier labelling changes long-term outcomes, and the one randomised test of a tau antibody did not turn biomarker signals into clinical benefit.

If you only have time for one paper this week, make it OPTIMAL-STROKE [1]. It is a large randomised trial in a stroke population that addresses the long-open question of how low systolic targets should go after ischemic stroke or transient ischemic attack, and it answers it with fewer hard cardiovascular events.

That's your roundup for This Week in Neurology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.