By Dr Peter Schueler, Senior Vice President, Drug Development Services and Haichen Yang, Vice President, Global Therapeutic Lead, both at CNS, ICON plc
Parkinson’s disease is the second most common neurodegenerative disorder after Alzheimer’s. The anti-Parkinson medications available primarily promote dopamine levels in the brain to compensate for the loss of dopamine due to neurodegeneration. Beside the still-existing need to better manage symptoms in advanced stages of the disease, there is an even greater need to develop neuro- protective or neuroregenerative therapies that will slow, stop or reverse disease progression.
The design of such neuroprotective studies may benefit from the lessons learned in studies in the most common neurodegenerative disease, Alzheimer’s. Target early disease stages: It is probably true for all neurodegenerative diseases that symptoms only occur when more than 70% of the relevant nerve cells are already gone. This seems to be a ‘point of no return’ for neuroprotection. Neuroregenerative therapies that may still work at such a late stage are not yet on the horizon.
Make sure the study population is sufficiently homogeneous, eg through imaging such as DaT-SPECT: with more precise diagnostic tools, it becomes evident that a clinical diagnosis of Parkinson’s disease (PD) may be wrong in about 15% of cases. In very targeted therapies, these 15% would most probably be non-responders, thus decreasing the effectiveness of a clinical trial.
Don’t rush into costly, large, phase 3 studies. Take the time to explore preliminary evidence of efficacy, suitable therapeutic dosage ranges and proper patient population in phase 2 clinical development. Adaptive design at phase 2 development can be a very effective approach to explore dosage range and provide insightful information. For instance, initiate a study with multiple doses, then remove inefficacious or intolerable dosage levels, and/or adjust patient numbers in different dosage groups based on interim analysis.
Explain to study subjects and caregivers they can’t expect improvements of PD symptoms, although they may benefit from a slower disease progression over time. In contrast to most other drug trials, the study subjects would not see any effect, even if they are in the active treatment group. However, they may experience adverse reactions. If this is not clearly communicated upfront, high study drop-out rates or (even worse) undisclosed low adherence to any oral therapy would be the consequence.
Take special measures to ensure adherence to the investigation product for non- injectable therapies. This is mandatory for all long-lasting prophylactic treatments. Published adherence in migraine-prophylaxis (as a comparable indication) indicates 92.6% adherent at one month, 40.7% at two months, 30.9% at three months, 24.8% at four months, 24.3% at five months and 20.8% at six months. Imagine a clinical trial with only 20% taking the medication as planned prior to the final assessments in month six – and even less after 12 months. It’s no surprise that many such studies fail. Preventive measures include: a) a daily medication reminder via text messaging; b) at each visit, site staff should discuss the importance of adherence with the subject and caregiver; c) a Patient Workbook that covers the background of a prophylactic treatment. Based on literature, education will help to maintain high adherence. Patients who received education were 96% adherent at six months; with no education, adherence was 58.5% at six months. A more technical solution is the use of control tools, such as drug dispensers with a pill count chip, or smartphone apps which document pill intake.
Do not rely on biomarkers (such as pathological forms of α-Synuclein), but put clinical outcomes as an endpoint in phase 2 trials. If the validity of a biomarker is not fully proven, it should not be used to take milestone-related decisions in the development plan. The amyloid hypotheses let many companies invest billions in phase 3 studies (which all failed), based only on a reduction of amyloid burden in phase 2. Today, innovative, digital endpoints make it much easier to reliably assess motor function as a clinical outcome, also in phase 2, with sufficient power. There are a number of devices that have been developed to measure symptoms such as tremor, bradykinesia and dyskinesia, eg Great Lake Neurotechnology’s Kinesia One and Kinesia 360, but also from MC10, BioSensics, Kinesis, Moticon, Mc Roberts etc. Taking all these lessons learned into consideration should markedly lower the risk of a false negative result in clinical trials for disease-modifying therapies against Parkinson’s.
