Jon's article is well worth a read. If you want to look further, it is useful to see the data in the paper [1] that includes the results for the 40 - 80 week period. The first period, 0 - 40 week, was double blinded, placebo controlled. The second period, 40 - 80 week was open label, with the original treatment group continuing as before and the original placebo group now also taking the treatment.

This GDNF trial should become a classic case study of statistical analysis. Every time a read it I come away with a different picture.

The ones that interest me today are:

LEDD mg, GDNF/GDNF, baseline= 1011, week 80= 1071
LEDD mg, Placebo/GDNF, baseline= 954, week 80= 1243

Timed tapping: OFF-state, GDNF/GDNF, baseline= 43.1, week 80= 63.8
Timed tapping: OFF-state, Placebo/GDNF, baseline= 42.4, week 80= 59.1

The increase in the levodopa equivalent daily dose especially in the Placebo/GDNF group is likely, in my opinion, to be a confounding issue.

Reference

[1] "Extended Treatment with Glial Cell Line-Derived Neurotrophic Factor in Parkinson’s Disease"
Alan L. Whonea et al.
Journal of Parkinson's Disease 2019, Corrected Proof
https://content.iospress.com/downloa...se%2Fjpd191576

John