We’ve learned a lot about Parkinson’s disease in the last 25 years (genetics, for example). However, the gold standard checklist for diagnosis — the respectable, if somewhat outdated UK Parkinson's Disease Society Brain Bank Clinical Diagnostic Criteria — hadn’t caught up.
New diagnostic criteria from an International Parkinson and Movement Disorder Society (MDS) task force comprising experts from four continents published recently in the journal Movement Disorders presents a tool that reflects our current understanding. The Michael J. Fox Foundation is currently funding a validation study of these MDS Clinical Diagnostic Criteria for Parkinson’s Disease.
"We now understand Parkinson’s as a multisystemic disorder affecting all parts of the nervous system, with a genetic component in many cases and a very slow progression of the neurodegenerative process, reflected in a long prodromal period. These aspects are now incorporated in the new criteria, which mimic the process of expert diagnosis," said task force co-chair Daniela Berg, MD, of University Hospital Tuebingen in Germany.
We spoke to task force co-chair Ronald B. Postuma, MD, MSc, of Montreal General Hospital about the process and promise of this new tool.
MJFF: What is the current state of diagnosis? What gaps do these criteria aim to fill?
Dr. Postuma: The way we diagnose Parkinson’s now is we document parkinsonism (the cardinal symptoms of tremor, rigidity and/or slowness) and then we decide it’s most likely Parkinson’s disease or not. There are criteria — the one that’s used most is the UK Brain Bank criteria — but they were developed over 25 years ago, and there have been many changes in the field.
But beyond that, what we do diagnostically is not fill out checkboxes as that tool is designed. Experts weigh things; they calculate things to make a decision. So with that in mind our goal is to codify what we do as experts and write it down in a systematic way so it can be applied by people who aren’t full-time Parkinson’s experts, and, importantly, to standardize diagnosis within clinical trials.
There are some absolute exclusion criteria that tell you it’s almost certainly not Parkinson’s. There are other things called red flags that make you worry about the diagnosis, but we know some people with Parkinson’s have that feature. And they’re weighted. You have to have more red flags than supportive features to rule out disease. That recapitulates the organic process when an expert diagnoses Parkinson's.
MJFF: Why is getting a diagnosis correct so important?
Dr. Postuma: A wrong diagnosis in clinical care is bothersome but not always a major problem. It’s not good for predicting prognosis, but you mostly treat the alternate conditions [multiple system atrophy (MSA), progressive supranuclear palsy (PSP), etc.] in the same way. But misdiagnosis is a disaster in clinical research. If you’re testing a drug for Parkinson’s disease, and 25 percent of your population doesn’t actually have Parkinson’s, you’re in a lot of trouble. You lose a lot of power. The study has to be way bigger than it otherwise would have to be, for example.
Especially early in disease, misdiagnosis rates are as high as 30 percent. Time helps you make the diagnosis. But most studies, especially for disease modification, are looking for people early in disease.
MJFF: So what changes in our understanding of Parkinson’s do these criteria address?
Dr. Postuma: Well, recognition of the non-motor aspect is important. If you see a patient who, after five years of disease, has no non-motor features of disease at all, that’s a red flag that you might be wrong in your diagnosis. So that’s actually been added to the criteria as a red flag. Secondly, the non-motor aspects are the most important way to diagnose prodromal disease. We have published a companion piece of criteria for prodromal selection for clinical trials.
Also, if you have abnormal olfaction (ability to smell) that can be used as evidence for Parkinson’s disease now. It’s a supportive criterion because we know that 80 percent of Parkinson’s patients have lost their sense of smell. And up to 80 percent of people who have alternate causes (PSP, MSA, essential or dystonic tremor) have normal smell. So it’s actually a pretty good diagnostic tool.
Family history of Parkinson’s disease is an exclusion criteria in many old criteria, but now there are clearly familial cases.
MJFF: Why isn’t DaTscan (brain scan to assess dopamine loss) part of the process?
Dr. Postuma: A normal DaTscan is in our criteria as an absolute exclusion. Other diseases with parkinsonism (MSA, PSP) also have abnormal DaTscan, so DaTscan is not useful in the differential diagnosis of true parkinsonism. And it’s not generally available. We want these criteria to be able to be applied everywhere.
MJFF: What can you tell us about the prodromal (before appearance of symptoms) diagnostic criteria?
Dr. Postuma: This is an effort to diagnose those early stages of Parkinson’s before people are aware of parkinsonism. When we develop neuroprotective therapies, we need to intervene earlier. We consider this a research tool only for identifying patients for neuroprotective trials or for identifying patients who need to be followed further.
It’s a very unique approach. What it does is combines a variety things that increase the risk of disease — sleep patterns, smell ability, subtle motor slowing, constipation — but often only do so slightly. It’s a systematic way to add these features together mathematically to come up with the likelihood that the person is in that prodromal state of disease. Not when they’re going to convert to full PD, but to test if they have pathology in their nervous system.
The diagnostic strength of the information varies immensely. Constipation doubles your risk of Parkinson’s, maybe triples it at best. You have to weight that differently than something like REM sleep behavior disorder, which is associated with a 50 to 100 times increased risk of developing Parkinson’s. So you add all the information together, and it gives you an output: this person has a 40 percent risk of being in the prodromal stage of Parkinson’s disease therefore you probably don’t want to include them in your neuroprotective trial, which may have side effects. This person has a 95 percent so definitely is a candidate, for example.
MJFF: How did your task force develop and test these criteria?
Dr. Postuma: The executive of the MDS decided that enough things have changed that we need to work on the definition of Parkinson’s. Once we had outlined our definition, we realized that we need some criteria to reflect the new definition. So the task force quite early on decided that new criteria would part of this job, and because we are so interested in the early stages, we decided to separate those criteria into diagnostic criteria for clinical Parkinson’s and research criteria for prodromal Parkinson’s.
Dr. Daniela Berg and I held early teleconferences then wrote the initial draft. It went back and forth among the task force members, an iterative process. And then we ran it through some clinics with people who had not seen the criteria before and improved the wording, things that weren’t clear, etc.
MJFF: What are the next steps?
Dr. Postuma: Now we’re testing against true experts in diagnosis. A resident, not expert in Parkinson’s diagnosis, applies the criteria to a patient. The same patient is then seen by an expert. The expert diagnosis is then compared to the diagnosis according to the new criteria. What do the concordance rates look like? Did we get it pretty close or are there big sources of divergence we should fix?
Even today, we think that every study that uses criteria should use these ones. Why wouldn’t you? There are no other criteria that can be used in a clinical trial now that reflect our updated definition of Parkinson’s.