The Alzheimer's blood test, and the drugs it points toward

The edition in brief

A meta-analysis of 21 phase 3 trials (12,610 patients) quantified amyloid-related imaging abnormalities (ARIA) with anti-amyloid antibodies in early Alzheimer's disease: ARIA-E occurred in 5.4% and ARIA-H in 10.3%, mostly asymptomatic, but risk rose sharply with APOE4 — homozygotes had about five times the odds of ARIA-E — and with higher dose and baseline microhaemorrhages. A population-wide Ontario cohort of 113,081 people found TIA carried a sustained higher risk of dementia (overall hazard ratio 1.34, highest in the first year at 1.75) that converged toward the risk seen after stroke by five years, with dementia far outnumbering recurrent stroke. A meta-analysis of four randomised trials (224 patients) found CSF shunting for idiopathic normal-pressure hydrocephalus improved gait velocity (standardised mean difference 0.73) and functional independence (odds ratio 3.92), with smaller cognitive gains; positional headache was more frequent and a subdural haematoma excess could not be excluded. A clinic pearl: confirm amyloid biology and genotype APOE before an anti-amyloid antibody, and set the MRI monitoring plan. Finally, a Mayo Clinic study of 706 patients found plasma p-tau217 discordant with amyloid-PET or CSF in 8.5%, with false negatives (5.6%) more common than false positives (2.8%); discordance tracked with age, renal function, body-mass index, thyroid disease and clinical stage — so a plasma result should be confirmed with PET or CSF in these groups before it drives a diagnosis or a disease-modifying drug.