New study offers insights into reliable Alzheimer’s diagnosis

A research group from LMU University Hospital has investigated how Alzheimer’s disease can be diagnosed more reliably in the future.

They will soon be approved in Germany: the first drugs against Alzheimer’s disease that are intended to slow the progression of the condition. On November 14, 2024, the EMA granted approval for Lecanemab in the European Union. The drugs attack so-called amyloid plaques in the brain. But how can we reliably and cost-effectively diagnose the presence of amyloid plaques in patients presenting to a memory clinic with mild cognitive impairment or mild dementia – i.e. the target group for the drugs?

A new study by physicians at LMU University Hospital provides answers that can be incorporated into patient treatment. The research was initiated by Professor Matthias Brendel, acting director of the Department of Nuclear Medicine, Dr. Nicolai Franzmeier from the Institute for Stroke and Dementia Research, and Professor Günther Höglinger, Director of the Neurological Clinic – all three of whom are also members of the SyNergy Cluster of Excellence. Now the results of the study have been published in the Journal of the Alzheimer’s Association, Alzheimer’s and dementia: diagnosis, assessment and disease monitoring.

Which method is better?

Basically, there are two approved methods to identify the presence of dangerous amyloid plaques in the brains of Alzheimer’s patients. Method number 1 involves analyzing the patient’s cerebrospinal fluid (CSF). However, this requires an epidural with a cannula – an invasive procedure with rare complications. And for some patients, such as people taking blood thinners, this test is not suitable. In addition, CSF analysis provides indirect, nonquantitative evidence of amyloid deposits in the brain.

Method number 2 involves a brain imaging technique called positron emission tomography (PET). This non-invasive method provides direct, semi-quantitative evidence of amyloid deposits in the brain. However, at 1,500 to 3,000 euros per scan, the method is still unaffordable and is currently not reimbursed by health insurance. Furthermore, depending on the equipment and expertise available in the centers, the use of amyloid imaging and CSF analysis varies in Germany, with CSF analysis currently even more widespread.

To determine the diagnostic accuracy of CSF test results compared to gold standard PET imaging, the Munich researchers evaluated data from more than 400 patients with suspected Alzheimer’s disease who underwent both a CSF amyloid test and a PET received a brain scan at LMU University Hospital between 2013 and 2024.

The results showed that patients with amyloid levels above 7.1 in their cerebrospinal fluid had PET scans that usually didn’t detect anything abnormal – meaning they tested negative for Alzheimer’s disease. Meanwhile, patients with an amyloid level of less than 5.5 in their cerebrospinal fluid mainly had PET scans that also came out as abnormal – meaning they tested positive for Alzheimer’s with a high degree of probability. Most significantly, however, there was a gray area between 5.5 and 7.1 in the cerebrospinal fluid – in about 15 to 20 percent of patients. “Half of these study participants had abnormal amyloid results in their PET scans,” says Brendel, “and therefore the CSF here is not reliable enough.” In an independent patient cohort at the University of Vienna, the researchers achieved exactly the same outcome. The results are therefore robust.

Possible implications for clinical practice

Once the new drugs for the treatment of amyloid plaques are approved, the study findings can be incorporated into diagnostic practice. Amyloid PET, if available, would be the diagnostic method of choice. However, depending on the expertise and equipment at a given location, many patients in Germany currently have easier access to CSF ​​analysis than to amyloid PET.

From a medical and economic point of view, it therefore seems reasonable to initially give these patients a CSF analysis, unless there are specific medical reasons that indicate otherwise.”

Professor Matthias Brendel, acting director of the Department of Nuclear Medicine

This concerns about 70 to 80 percent of the patients.

“Of these patients, only those whose results are in the gray area between 5.5 and 7.1 would require an additional PET scan. Matthias Brendel: “Certainly if the costs of amyloid PET decrease in the future and broader access becomes possible, amyloid PET would be the first choice and avoid duplication of efforts and costs where currently two tests – CSF and PET – are required.”