Body-first vs brain-first Parkinson's

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Per Borghammer and Nathalie Van Den Berge (J Parkinsons Dis 2019; the paper is titled "Brain-first versus gut-first" but speaks of peripheral-first and CNS-first subtypes); Jacob Horsager, Borghammer and colleagues at Aarhus University (Brain 2020). Funders were not checked.

What happened

• 2019 hypothesis. The authors split Lewy body disease into a peripheral-first and a brain-first type. In the first, damage to the autonomic nerves (which run gut, heart and bladder) comes first, REM sleep behaviour disorder (RBD, acting out dreams) is common in the prodrome, and dopamine loss follows. In the second, dopamine loss comes first and RBD is usually absent. Supporting data included cardiac nerve scans in RBD and colon biopsies with abnormal synuclein in 18 of 28 (64%) RBD-positive versus 2 of 15 (13%) RBD-negative Parkinson's patients, from a study the review cites.
• 2020 imaging study. Horsager et al. used PET, MIBG scanning and MRI on 37 new Parkinson's patients (24 RBD-negative, 13 RBD-positive) and 22 people with isolated RBD. The RBD-positive and isolated RBD groups had lower cardiac (MIBG) and colon (donepezil PET) nerve signals than the RBD-negative group, and larger colons with slower transit. Their data fit a body-first path (heart and gut nerves lost first, then putamen dopamine), while the RBD-negative group fit a brain-first path.
• A commentary noted limits: cross-sectional design, no direct pathology, and some exclusions that may exaggerate differences. Beach et al. 2021 (autopsy study of the vagus nerve and stomach) argued against a body-first route (landmark 18).

Why it moved us toward a cure

It may explain why one therapy fails in a mixed group, since subtypes could need different treatments. It builds on the ordered spread of landmark 6 and the gut-to-brain spread in mice of landmark 18. It suggests enrolling by subtype and, for the body-first type, intervening early in the gut.

Outcome

Open. It is still a model. Human proof of the starting site is lacking. Horsager, Knudsen and Sommerauer's 2022 review (Neurobiol Dis 164:105626, DOI 10.1016/j.nbd.2022.105626) collected clinical and imaging differences between the proposed subtypes and noted LRRK2 carriers mostly look brain-first and GBA carriers body-first; I did not read it in full. I found no correction or retraction for either key paper.

Sources

• Borghammer and Van Den Berge, J Parkinsons Dis 2019;9(s2):S281–S295, DOI 10.3233/JPD-191721 (PMC6839496)
• Horsager et al., Brain 2020;143:3077–3088, DOI 10.1093/brain/awaa238 (abstract read; commentary at Brain 143:2871)
• Horsager, Knudsen, Sommerauer, Neurobiol Dis 2022;164:105626
• Not yet verified: Horsager full text; funders of both papers; independent replication of the imaging pattern; the clinical-outcome differences between subtypes.