Sharing thoughts on my recent misstep: Due to concerns about the risk/benefit of monoclonal antibodies in #Alzheimers, I preprinted an analysis (now withdrawn) quantifying deaths among treated patients using FDA’s FAERS data. (1/10) 2/ Challenges and errors in the analysis emerged after posting. I take full responsibility for not anticipating these issues and preprinting without full coauthor agreement, leading to the concerns highlighted by Jennifer Couzin-Frankel in @ScienceMagazine
science.org/content/articl…

An analysis of FDA’s Adverse Event Reporting (FAERS) data, just posted on @researchsquare, provides early evidence of real-world increased mortality among #Alzheimers patients treated with anti-Aβ monoclonal antibodies, aducanumab & lecanemab. 🧵 (1/6)
researchsquare.com/article/rs-528… (2/6) Death rates for patients treated with lecanemab & aducanumab are 2-4x higher than the background mortality in a similar Alzheimer’s population.

An analysis of 24 randomized trials involving 25,000 Alzheimer's patients suggests an alternative explanation for how anti-Aβ monoclonal antibodies work: they may slow cognitive decline by raising Aβ42 levels. 🧵(1/12)
doi.org/10.1093/brain/… (2/12) The hypothesis. If dementia in amyloid-positive people only occurs when Aβ42 levels are very low, regardless of the levels of brain amyloid, then high levels of Aβ42 should lessen the risk of dementia.
sciencedirect.com/science/articl…

Why do we describe 20 years of biomarker data as “divergence” and summarize the findings as what "goes up" when? What we lose matters. CSF Aβ42 levels remain high in controls but drop by ~300 pg/ml before #Alzheimers begins. @NEJM (1/5) nejm.org/doi/full/10.10…

https://twitter.com/AlbertoEspay/status/1701256845896327381

(2/5) The Jia et al @NEJM data align with the finding that in a PET-positive sporadic cohort (ADNI), normal cognition correlates with higher levels of CSF Aβ42 across all brain amyloid levels. At any PET level, lower Aβ42 levels are associated with AD. thelancet.com/journals/eclin…

@MadhavThambiset was the voice of reason at the #AANAM. He delivered a Neurology Year in Review Plenary about the dubious record of the anti-proteinopathy framework inspiring therapies in #Alzheimers, #Parkinsons, and #ALS --Some highlights (1/4)

https://twitter.com/AlbertoEspay/status/1650873538688024576

(2/4) Unblinding due to infusion-related reactions or ARIA-H was not accounted for in the sensitivity analysis of the #lecanemab trial. This and other issues have led to an inflated sense of efficacy and underestimated risk for anti-amyloid therapies.

The success of #COVID19 vaccines required heavy investments into #HIV research and the resourcefulness and creativity of many. The 60-year history summarized by @nytimes highlights important lessons. (1/12)

nytimes.com/2022/01/15/hea… (2/12) 1st milestone: 1960 – “Molecule X” carried copies of DNA segments to ribosomes to make proteins. It is posited to be the "messenger" RNA, or mRNA. Could this molecule be instructed to make tiny pieces of viruses to strengthen the immune system?

#Donanemab for #Alzheimer, misleadingly spun as positive: “donanemab resulted in a better […] ability to perform activities of daily living than placebo at 76 weeks”. The data suggest otherwise. (1/8)

#EliLilly @NEJM #ADPD2021

nejm.org/doi/full/10.10…

https://twitter.com/NEJM/status/1370702429461782529

(2/8) First off, donanemab accomplished exactly what it was designed for: At 76 weeks, the reduction in the amyloid plaque level as assessed by florbetapir PET was 85.06 centiloids greater in the donanemab group than in the placebo group (−84.13 vs. 0.93 centiloids)