Critiques of Braak staging

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

Robert Burke, William Dauer, Jean Paul Vonsattel (a critical review in Ann Neurol 64:485–491, 2008); Kurt Jellinger (a review in Biochim Biophys Acta 1792:730–740, 2009); Laura Parkkinen, Tuomo Pirttilä, Irina Alafuzoff (an autopsy study in Acta Neuropathol 115:399–407, 2008); Kalaitzakis, Graeber, Gentleman and Pearce (a letter in Acta Neuropathol 116:125–128, 2008). PubMed lists US NIH extramural support for Burke; the other papers' funders were not checked.

What happened

Heiko Braak's six stages put Lewy pathology first in the lower brainstem and smell centre, then in the nigra, then in the cortex (landmark 6). Others tested the pattern in other brain collections.

• Burke 2008. Critical review. Outside people diagnosed with Parkinson's, the pattern is often not seen, especially in dementia with Lewy bodies. Some elderly people without neurological signs had synuclein pathology up to Braak stages 4 to 6, and Braak stage did not relate to the clinical severity of Parkinson's. They concluded that the link between synuclein staining and the disease "remains to be determined".
• Parkkinen 2008. From 1,720 autopsies, 248 (14%) had synuclein deposits in a key brain region, and 226 were staged. 83% fitted a staging system; about 17% deviated from the bottom-up path. About 55% of those at Braak stages 5 to 6 had neither dementia nor movement signs.
• Jellinger 2009. Review: studies of staging found that 6.3% to 43% of cases did not follow the proposed order; in 7–8.3% of clinical Parkinson's cases the medulla was spared.

Why it moved us toward a cure

It sets the limits of the spreading idea behind antibody and gut-first work (landmark 18). Pathology can be present without symptoms, and a single order does not fit all patients. This is part of why researchers now ask which people carry a "brain-first" or "body-first" pattern (Horsager 2020, Brain 143:3077–3088, a multimodal imaging study).

Outcome

Open. Most cases fit, but not all, and staging is not a validated predictor. The critics did not disprove spread, only the idea that one fixed route explains every case. Later work agrees: in the Arizona brain bank, 70 of 280 brains (25.3%) could not be classified by Braak's system, while a unified staging system placed all 280 and correlated with symptoms (Adler 2019, J Neuropathol Exp Neurol; first described by Beach 2009, Acta Neuropathol). No correction or retraction was found for the four papers (Crossref, Europe PMC).

Sources

• Burke et al., Ann Neurol 2008, DOI 10.1002/ana.21541
• Parkkinen et al., Acta Neuropathol 2008, DOI 10.1007/s00401-008-0346-6
• Jellinger, BBA 2009, DOI 10.1016/j.bbadis.2008.07.006
• Kalaitzakis et al., Acta Neuropathol 2008, DOI 10.1007/s00401-008-0381-3
• Adler et al., J Neuropathol Exp Neurol 2019;78:891–899, DOI 10.1093/jnen/nlz080 · Beach et al., Acta Neuropathol 2009;117:613–634, DOI 10.1007/s00401-009-0538-8 · Horsager et al., Brain 2020;143:3077–3088, DOI 10.1093/brain/awaa238
• Not yet verified: the Horsager study's findings (title only read); the Kalaitzakis letter's content (no abstract; it is listed in Europe PMC as a comment on the staging papers).