Trials of neuroprotective therapies for Parkinson's disease: problems and limitations.

Since the initial results of the DATATOP study, considerable effort has been devoted over the past 20 years to test neuroprotective therapies for Parkinson's disease (PD). Two trials (CALM-PD-CIT and REAL-PET studies) used neuroimaging dopamine changes as a surrogate marker for PD progression, and concluded that pramipexole and ropinirole could have a neuroprotective effect compared to levodopa. However, it should be recognized that all the presynaptic dopamine markers currently available for SPECT and PET studies are potentially subject to regulatory changes. Consequently, the results of these two trials can also be interpreted in terms of drug-related differences in dopamine regulation. More recently, the delayed-start design was applied to test whether rasagiline could have a neuroprotective effect in PD (ADAGIO study). Unfortunately, a major limitation of the delayed-start design is that, whenever the active treatment has a symptomatic effect, the blinding can be broken. This can lead to unequally-distributed placebo responses during phase 2, and is also a potential source of raters' biases. None of the trials on neuroprotective therapies for PD has yet provided solid evidence for neuroprotection. Copyright 2010 Elsevier Ltd. All rights reserved.