PET scanning is giving Mass General Hospital researchers a whole new view of neurodegenerative diseases. Previously, definitive diagnoses of these diseases could only be achieved by post-mortem pathological exam. Now, the newest radiotracers are giving physicians a live feed on the molecular events in the brain that drive disease progression. For Parkinson’s disease and related conditions, PET promises more accurate and earlier diagnoses, paving the way for therapies targeted to specific molecular pathologies. In the field of amyotrophic lateral sclerosis (ALS), a new marker for inflammation in the brain may help speed clinical trials.

Parsing Pathology

Parkinsonian syndromes present a diagnostic challenge for neurologists. They are often but not always caused by one of five neurodegenerative diseases: Parkinson’s disease (PD), dementia with Lewy bodies (DLB), multiple systems atrophy (MSA), progressive supranuclear palsy (PSP), or corticobasal degeneration (CBD). Although each disease has a classical presentation, there is sufficient clinical variation that it can be difficult to get the diagnosis right, and this problem is especially severe early in the course of these illnesses.

But on a molecular level, these conditions are distinct. PD, DLB and MSA are all synucleinopathies, so-called because they are caused by accumulation of aggregated alpha-synuclein. PSP and CBD are tauopathies, marked by accumulation of abnormal tau proteins. Some Parkinsonian patients also show amyloid plaques, a pathology associated with Alzheimer’s disease.

“The problem is that the clinical syndromes that bring patients in for evaluation are imperfect predictors of the underlying pathological process, and we lack good tests that can help determine who’s got what,” says Mass General neurologist Stephen N. Gomperts, MD, PhD. “In fact, in some of these conditions, we are wrong a significant fraction of the time.”

To correct that, Gomperts and his collaborators are testing whether a combination of amyloid-PET and tau-PET imaging yields enough information to support a pathology-based diagnosis in these disorders. In an ongoing longitudinal study of patients with the classical presentations of these Parkinsonian syndromes, Gomperts is aiming to relate molecular imaging findings to clinical diagnosis and to the course of cognitive and motor decline. In the study, patients receive MRI scans and PET scans, and are followed with annual cognitive testing, neurologic evaluation, and blood and genetic testing.

Early results look promising, says Gomperts. The amyloid-PET ligand identifies cases associated with Alzheimer’s pathology. For amyloid-negative cases, the tau-PET ligand, originally developed to detect neurofibrillary tangles in AD patients, appears to light up regions associated with tau deposits in patients with typical presentations of PSP and CBD, the two PD-related tauopathies. Gomperts plans to compare different tau-PET ligands to find the most robust markers of tau deposition in PSP and CBD. He anticipates that differences in the topographical distribution of tau in the brain in PSP versus CBD may provide the power to definitively diagnose each condition, and to distinguish them from either the synucleinopathies or AD.

Getting it right is important, because new therapies that target disease-specific pathological processes, such as tau or synuclein deposits, are in early stage clinical testing. Being able to select the right patients for the right therapies will help speed the evaluation of these new drugs and will ultimately be critical for future treatments. “Making the correct molecular diagnosis is a critical step if we are to find medicines that work in people,” says Gomperts.

Another goal of the research is to understand why some patients with Parkinsonian syndromes develop dementia, and specifically how amyloid and tau deposition might contribute to cognitive symptoms in these diseases. Using PET imaging, Gomperts and his collaborators have shown that high levels of amyloid are often seen in people with DLB compared to those with PD. In addition, their longitudinal data further show that the more amyloid pathology present in PD, the steeper the course of cognitive decline.

Several amyloid-clearing drugs are now in late-stage clinical trials for AD, and Gompert’s results suggest they might also be applicable to treat dementia in PD or DLB. “If the trials are successful, there is good reason to consider applying the same amyloid-targeting technology to cases of PD and DLB where amyloid deposits are present,” he says.

Imaging Inflammation

Nazem Atassi, MD, Mass General neurologist, is working with another novel PET tracer, one that can detect inflammation in the brains of people with ALS. Part of his motivation is streamline clinical trials of candidate treatments with anti-inflammatory mechanisms.

Currently, trials for ALS are slow affairs, requiring hundreds of patients be followed for at least a year to see if the drug affects physical functioning or not. “These are very expensive studies, they take a very long time for both the patients and the researchers,” Atassi explains.

Having a biomarker for inflammation in the brain would allow a much more efficient trial. Starting with a much smaller group of subjects, researchers could look for a decrease in inflammation after a short course of treatment. Knowing that the medication entered the brain and acted as expected in terms of mechanism would give researchers more confidence to embark on a longer trial with more patients, Atassi says.

The technology could also help in choosing patients for trials “There might be a subset of patients where inflammation is driving the disease, and so are more likely to respond to inflammatory treatments,” Atassi says. Enriching for those patients may increase the chances of a positive result.

To define such a biomarker, Atassi and colleagues are using the PBR28, which binds to TSPO, a protein present in activated glia and a marker for brain inflammation. In a first study of PBR28 uptake in people with ALS, the researchers showed increased uptake of the PET ligand compared to healthy people. Localization of brain inflammation corresponded to the clinical presentation: people with limb onset ALS showed more uptake in the motor cortex, while those with bulbar onset, uptake was more pronounced in bulbar regions.

Tracer uptake was also related to function: people who had more tracer uptake, hence more glial activation, had lower scores on the ALS-Functional Rating Scale, the most commonly measure in clinical trials for ALS. Now, Atassi and colleagues are incorporating PBR28 scans into several small clinical trials of new anti-inflammatory treatments for ALS.

The early data suggest that PET scans could also accelerate the diagnosis of ALS. “If you look at any one scan, without being a radiologist or a neurologist, you can tell right away who has ALS and who does not,” says Atassi. His own work shows that currently it takes about 12 months on average for patients to be diagnosed, from the time of first symptoms. “This could be a way to reduce that diagnostic timeline and help with early diagnosis,” he said.

To help expand the use of the tracer, Atassi and his colleagues are leading a multicenter study of a second generation TSPO-PET ligand. The new tracer, GE-180, incorporates an isotope with a half life of nearly 2 hours, compared to 20 minutes for PBR28. That will make the technology more portable and easier to use. The new tracer works the same platform as FDG-glucose, a PET ligand that is already widely used in most hospitals. “We are thinking ahead in terms of how you can scale this up, and how to use it in multiple centers, both in the clinical setting and in clinical trials,” he says.