Conference Coverage

Baseline subtypes predict dementia and death in patients with Parkinson’s disease



Dementia and death were about four times more likely after 11 years in patients with Parkinson’s disease and cognitive or psychiatric symptoms at baseline, versus those with motor symptoms alone, according to the results of a longitudinal study of 162 patients at Washington University, St. Louis.

Dr. Peter Myers of Washington University, St. Louis

Dr. Peter Myers

Parkinson’s disease is often only staged as mild, moderate, or severe, and sometimes cognitive and psychiatric symptoms are not assessed. The St. Louis team wanted to see if there are actual subtypes, and if they have clinical relevance, said lead investigator Peter Myers, PhD, a postdoctoral research associate at Washington University.

The magnitude of the findings was “surprising. ... We don’t think we are seeing stages” because the different symptom patterns were apparent at baseline. Instead, “we are seeing really distinct clinical subtypes that could inform patient prognosis and help prepare families and caregivers. It is important that you assess more than just motor symptoms,” he said at the annual meeting of the American Neurological Association.

Some of the subjects were newly diagnosed and others diagnosed years earlier. At baseline, they had symptoms for an average of 6 years, and none had dementia.

After a battery of cognitive, psychiatric, and movement tests, Dr. Myers and his colleagues found that their subjects fell into three patterns: motor symptoms only, with normal cognitive and psychiatric performance (63 subjects); motor symptoms plus prominent anxiety or depression (17); and motor symptoms plus cognitive deficits (82).

A total of 42 patients developed dementia – a score of at least 1 on the Clinical Dementia Rating evaluation – including three in the motor-only group (5%), two in the psychiatric/motor group (12%), and 37 in the cognitive/motor group (45%).

After controlling for age, sex, and symptom duration, the risk of dementia conversion was far higher in the cognitive/motor group (relative risk, 4.16, 95% confidence interval, 1.18-14.65) and psychiatric/motor group (RR, 3.08; 95% CI, 0.72-13.28), compared with motor-only subjects.

Thirty-eight patients died, the majority because of Parkinson’s disease-related causes, including five in the motor-only group (8%), two in the psychiatric/motor group (12%), and 31 in the cognitive/motor group (38%).

The risk of death, relative to motor-only patients, was higher with both the cognitive/motor (RR, 4.06; 95% CI, 1.37-12.03) and psychiatric/motor subtypes (RR, 4.17; 95% CI, 0.70-24.91).

It is unclear what leads to the progression differences, but the researchers’ hypothesis is that the broader scope of symptoms indicates a greater extent of brain pathology. The St. Louis team is looking at brain-imaging data to see if that is true, and if the subtypes can be distinguished on imaging. They are also interested in seeing if genetics plays a role in susceptibility to the different subtypes, and if the baseline subtypes are stable over time or if patients convert between them.

Subjects were, on average, 66 years old, and 62% were men. The mean levodopa-equivalent daily dose was 613 mg in the motor-only group, 1,004 mg in the psychiatric/motor group, and 783 mg in the cognitive/motor group (P = .03). Mean symptom duration was shorter in motor-only patients (5.1 years) versus the psychiatric/motor group (7.2 years) and cognitive/motor group (6.9 years, P less than .01).

Although relative risks crossed 1 in the psychiatric/motor group, it was probably because there were only 17 patients. “We do think there is a very strong likelihood that” the psychiatric/motor findings, like the cognitive/motor findings, “are valid,” Dr. Myer said.

Dr. Claire Henchcliffe MD, DPhil, director of the Parkinson’s Disease and Movement Disorders Institute at Cornell Weill Medical Center, Manhattan

Dr. Claire Henchcliffe

“I think [the findings] could be real,” said Clair Henchcliffe, MD, DPhil, vice chair for clinical research in the department of neurology at Weill Cornell Medical College in New York.

“Other publications have shown that people who have cognitive symptoms at onset are more likely to go on to develop dementia down the line. This is much in line with that,” she said when asked for comment.

The field has “always been interested in finding data that help us personalize treatment, and a lot of people are looking to subtype Parkinson’s disease to help us plan with our patients.” It could also help with frequency of follow-up, trial participation, and maybe someday treatment selection. “We are always thinking a few years ahead” with Parkinson’s disease, she said.

There was no industry funding, and the investigators did not have any relevant disclosures.

Next Article: