Let me say at the outset that in spite of the negative outcome that I think the creatine study was worth doing: both for its own sake (there was a reasonable a priori case as to why creatine might be effective); and also because it has created a large amount of longitudinal data that can be used in other studies. My concern is that it has taken so long and I presume (I don't have the figures) cost so much to reach this conclusion. Unless lessons are learnt to shorten trial lengths and reduce costs, we will only be able to trial a few of the many potential therapies for PD.
The paper describing the work in detail has been published [1]. Unfortunately this is behind a pay wall. However, the abstract is available, and this gives much interesting information.
Number of participants = 1741, enrolled in 45 investigative sites. With so many sites there may be confounding issues in areas such as the consistency of subjective assessment, differences between the efficacy of the various treatment regimes, any variations in geographical rates of progression.
They used a basket of five outcome measures. "All outcomes were coded such that higher scores indicated worse outcomes and were analyzed by a global statistical test. Higher summed ranks (range, 5-4775) indicate worse outcomes". Unless these measure independent features of PD, this aggregation may add little in terms of accuracy. (Separating the results from each test will give interesting results about the correlation of the tests to each other).
"The trial was terminated early for futility based on results of a planned interim analysis of participants enrolled at least 5 years prior to the date of the analysis (n = 955). The median follow-up time was 4 years. Of the 955 participants, the mean of the summed ranks for placebo was 2360 (95% CI, 2249-2470) and for creatine was 2414 (95% CI, 2304-2524)".
What strikes me here is that when the trial finished the results show creatine doing slightly worse than placebo. This is probably not statistically significant, but it makes it very unlikely that the true result is positive and even more unlikely that the true benefit is big enough to justify taking the drug, more than 5%, say. Could the call to end the trial have been made earlier?
Working with a cohort answers some questions, but doesn't answer the most important question: does creatine help me? It will be interesting to see if they have analysed any subgroups.
Working with only a placebo and creatine doesn't answer the question of whether the idea of taking creatine is beneficial. It could be that the placebo effect for creatine is large.
It seems to me the way forward involves:
- using proper biomarkers;
- cheap, objective tests;
- high frequency testing (daily);
- asking a different question.
Reference:
[1] "Effect of Creatine Monohydrate on Clinical Progression in Patients With Parkinson Disease"
JAMA. 2015;313(6):584-593. doi:10.1001/jama.2015.120.
http://jama.jamanetwork.com/article....icleid=2108890
John