GLP-1 receptor agonist medications are currently being evaluated in major Alzheimer's disease clinical trials in non-overweight and non-obese populations.
"now we have the GLP-1 drugs like Ozempic, Mounjaro that are being tested in big Alzheimer's trials in thin people. You know, these are not obese or overweight; these are thin people, and because they have such potency of reducing brain inflammation" (said at 0:26:43)
Major Phase 3 clinical trials (evoke and evoke+) evaluated the GLP-1 receptor agonist semaglutide in large populations of patients with early-stage symptomatic Alzheimer's disease, enrolling participants based on amyloid positivity and cognitive impairment rather than elevated BMI or diabetes, thus including non-obese and normal-weight individuals. However, the Phase 3 trial results demonstrated that semaglutide did not slow clinical cognitive progression compared to placebo.
- supports: evoke and evoke+: design of two large-scale, double-blind, placebo-controlled, phase 3 stu… (Alzheimer's research & therapy 2025) · cited 139x in the literature
"evoke and evoke+ are randomized, double-blind, placebo-controlled phase 3 trials investigating the efficacy, safety, and tolerability of once-daily oral semaglutide versus placebo in early-stage symptomatic AD." (abstract, methods, passage verified)
pubmedfull study (doi) - context: Efficacy and safety of oral semaglutide 14 mg (flexible dose) in early-stage symptomatic A… (Lancet (London, England) 2026) · cited 36x in the literature
"Oral semaglutide was not efficacious in slowing clinical progression in participants with early Alzheimer's disease." (abstract, conclusions, passage verified)
pubmedfull study (doi)
In an American Journal of Cardiology study, 39% of the population had blood lead levels over 2 µg/dL, and those individuals had an increased risk of heart attack and stroke comparable to or higher than elevated cholesterol.
"I remember reading a paper, I think it was the American Journal of Cardiology years ago, where they looked at anybody who had lead levels over two, which is considered normal because level in the reference range is 1 to 10... That their risk of having a heart attack was higher or as high as those who had elevated cholesterol and an increased risk of strokes, and it was a big risk factor. And it was 39% of the population that had a lead level over two" (said at 0:50:45)
The speaker conflates specific statistics and misnames the journal, but correctly reflects the general findings of landmark prospective epidemiological research on low-level lead exposure and cardiovascular mortality. In an analysis of 14,289 adults from the NHANES III cohort followed for nearly 20 years (published in The Lancet Public Health, not the American Journal of Cardiology), geometric mean blood lead was 2.71 µg/dL. Increasing blood lead from 1.0 to 6.7 µg/dL was associated with a more than two-fold increase in ischemic heart disease mortality (HR 2.08, 95% CI 1.52–2.85) and a 70% increase in cardiovascular mortality (HR 1.70, 95% CI 1.30–2.22). The population attributable fraction for ischemic heart disease mortality was estimated at 37.4% (roughly 185,000 deaths annually in the US), a disease burden comparable to major traditional risk factors such as smoking and dyslipidemia. The speaker appears to have conflated this attributable fraction (37.4%) or the proportion of cardiovascular deaths in the cohort (38%) with the prevalence of exposure.
- context: Low-level lead exposure and mortality in US adults: a population-based cohort study. (The Lancet. Public health 2018) · cited 620x in the literature
"An increase in the concentration of lead in blood from 1·0 μg/dL to 6·7 μg/dL (0·048 μmol/L to 0·324 μmol/L), which represents the tenth to 90th percentiles, was associated with all-cause mortality (hazard ratio 1·37, 95% CI 1·17-1·60), cardiovascular disease mortality (1·70, 1·30-2·22), and ischaemic heart disease mortality (2·08, 1·52-2·85). The population attributable fraction of the concentration of lead in blood for all-cause mortality was 18·0% (95% CI 10·9-26·1), which is equivalent to 412 000 deaths annually. Respective fractions were 28·7% (15·5-39·5) for cardiovascular disease mortality and 37·4% (23·4-48·6) for ischaemic heart disease mortality, which correspond to 256 000 deaths a year from cardiovascular disease and 185 000 deaths a year from ischaemic heart disease." (abstract, results, passage verified)
pubmedfull study (doi)
A Swedish trial of over 100,000 women demonstrated that AI-assisted mammography screening detected 25% more cancers than radiologists alone with no increase in false positives.
"100,000-plus women in Sweden, the AI picked up 25% more cancers compared to radiologists alone, you know, significant cancers, and no increase in false positives." (said at 1:21:40)
The statement references the Swedish Mammography Screening with Artificial Intelligence (MASAI) randomized controlled trial published in The Lancet Oncology, but contains minor numerical discrepancies. In the published interim safety analysis of 80,033 women (analyzing 39,996 in the AI-supported group and 40,024 in the standard double reading group), AI-supported screening detected 20% more cancers than standard double reading (244 vs 203 screen-detected cancers; detection rate 6.1 vs 5.1 per 1,000 participants; ratio 1.2, 95% CI 1.0-1.5, p=0.052), rather than 25%. The false-positive rate was identical in both groups at 1.5% (95% CI 1.4-1.7%). Most detected cancers in both groups were invasive (75% in the AI group vs 81% in the control group).
- supports: Artificial intelligence-supported screen reading versus standard double reading in the Mam… (The Lancet. Oncology 2023) · cited 555x in the literature
"Between April 12, 2021, and July 28, 2022, 80 033 women were randomly assigned to AI-supported screening (n=40 003) or double reading without AI (n=40 030)... AI-supported screening among 39 996 participants resulted in 244 screen-detected cancers... Standard screening among 40 024 participants resulted in 203 screen-detected cancers... Cancer detection rates were 6·1 (95% CI 5·4-6·9) per 1000 screened participants in the intervention group, above the lowest acceptable limit for safety, and 5·1 (4·4-5·8) per 1000 in the control group-a ratio of 1·2 (95% CI 1·0-1·5; p=0·052)... The false positive rate was 1·5% (95% CI 1·4-1·7) in both groups." (abstract, results)
pubmedfull study (doi)