Thread by @MadhavThambiset

1/n Good point, Mike. I can use #solanezumab as a case study in selective reporting of RCT trial results which does not serve patient’s interests. Sola was discontinued despite pooled results from 2 ph-3 trials suggesting that ‘mild’ (but not ‘moderate’) stage #Alzheimer patients

https://twitter.com/michaelokun/status/1637453397363154947

2/n may benefit. But a 3rd ph-3 trial in only ‘mild’ AD was clearly negative. Presumed reason was that the Rx did not lower brain amyloid levels enough. Lilly published the MRI findings from these studies looking at brain volume changes. They considered these results important to

3/n report because they noted that “Rates of whole brain volume & ventricular volume changes are sensitive markers of neurodegeneration & used as biomarker outcome measures in trials of potential disease‐modifying therapies.” tinyurl.com/33a7dkh3 They also reported on the

4/n relationship between volume changes & cognitive outcomes in placebo & Rx groups. The main conclusions were that there were no differences in volume/ventricular changes in Rx relative to placebo. As the disease progressed brain volumes ↓ & ventricular volume ↑ in both groups

5/n & these changes were also associated with worse cognitive outcomes as seen here.👇 I highlighted whole brain & ventricular volume correlations with cognitive outcomes. These results are important because even though the drug “failed” we know that it did not accelerate atrophy

6/n & “neurodegeneration”. But we now have results from a “successful” ph-2 trial of another amyloid antibody from Lilly, #donanemab as well as 2 ph-3 trials of #aducanumab & ph-2 & ph-3 trials of Lecanemab- all designated as “breakthrough” Rx by @US_FDA. All these trials showed

7/n effective brain amyloid clearance but also ↓ in brain volume &/or ↑ in ventricular volume relative to placebo. BUT no publication has reported on whether these changes are associated with worse cognitive outcomes- a glaring omission. In #lecanemab ph-2 RCT 👇the

8/n brain volume loss at the highest & “effective” dose is 8ml > than placebo over 18 months which is approximately 3 tsp of brain volume and this increases over the course of Rx. It is unconscionable that the trials have not disclosed till now if these changes track with

9/n worse cognitive outcomes. If they don’t it would be reassuring to our patients to know this. If they do, it would be important to include a warning on the label about this as a potential adverse event. Accelerating brain atrophy is usually an outcome that would be considered

10/n by most to be undesirable in a progressive neurodegenerative disease. The fact that companies can selectively report on these MRI changes & their relationship with cognitive outcomes from some RCTs & not disclose them in others is a grave concern. Patients volunteered

11/n to participate in these experiments agreeing to be randomized to Rx or placebo over years. They deserve to know these results in full. Not disclosing them is an egregious betrayal of their trust & risks causing them long term harm. Failure to disclose these results also

12/n raises legitimate questions about why these findings are being concealed. It undermines the credibility of trial sponsors, regulators as well as physicians counseling patients about potential harms of these experimental Rx. We can and must do better.