Arbitrage opportunity, the odds on the other question don't make sense: https://manifold.markets/StrayClimb/biological-test-for-early-detection-0a920c9cdd47

Biological test for early detection of Parkinson's Disease before 2026

62% chance. It doesn't have to show that they will get Parkinson's 100%, just that they progressing towards it and would very likely get it if they don't change their lifestyle and lived long enough. It is not a test of their genome - it needs to be of some active biological state of their body. Current theories suggest people who get diagnosed with Parkinson's have obvious (in retrospect) symptoms throughout their body, and that these are likely diagnosable chemically, even though actual diagnosis typically only happens when the patient reports the typical symptoms: one-sided stiffness, slowness of movement, etc.

Also:

• Possible early diagnosis through retinas: https://cureparkinsons.org.uk/2023/09/seeing-the-early-signs-of-parkinsons/

• Blood test: https://parkinsonsnewstoday.com/news/blood-test-detect-synucleinopathies-parkinsons-disease/ "IP/RT-QuIC’s ability to correctly identify people with Parkison’s was 94.6% (sensitivity), while its ability to rule out the condition was 92.1% (specificity). Also, a faster rate of aggregate formation correlated with longer disease duration and worse motor symptoms, as assessed by the Unified Parkinson’s Disease Rating Scale part 3."

MJFF just came out with a test that detects alpha synuclein in spinal fluid. Right now it's binary but later can hopefully be extended to also measure change over time.

Summary video https://www.youtube.com/watch?v=wFhpkMhFzuM

Lancet paper: https://www.thelancet.com/journals/laneur/article/PIIS1474-4422(23)00109-6/fulltext

This claim may already be true, depending on rollout and use of the test. I am reading the paper now but the key question is whether people with positive results on AS-SAA who don't have PD yet can be shown to later develop it at a high rate.

@StrayClimb e.g. this article

https://www.eurekalert.org/news-releases/985529#.ZDgQsrduYCM.twitter

"In people with REM sleep behaviour disorder, a sleep disturbance that is known to be a precursor to Parkinson’s disease, positive αSyn-SAA results were present in 85% (28/33) of cases."

So we have a disorder (REM Sleep behavior disorder) which is highly predictive of later PD clinical diagnosis. And now we found a biological test which shows that people with that disorder show up positive.

So it would be very natural to conclude that if you show up positive in this new test, you are likely going to get PD.

I took a look at the paper. The specificity of the test is 96.3%, i.e. a 3.7% false positive rate. For a disease as rare as Parkinson's, that means most of the people the test diagnoses with pre-Parkinson's would be false positives. The upper range of the confidence interval is 99.2%, which would be a little more promising, so it'll be interesting to see what future studies with greater statistical power find.

Note that the eurekalert.org link says "The authors acknowledge some limitations to their study. Greater numbers of samples would improve analyses by helping to overcome issues posed by factors including skewed data and low sample numbers for some participant groups. The analyses presented are all cross-sectional, but the availability of PPMI samples collected over time would enable future studies to assess changes over specific time periods. Longer-term studies are also needed to further investigate differences in αSyn-SAA results between people with different genetic forms of Parkinson’s disease."

I'm skeptical that could all be done by 2028. I'm also skeptical that after all of that is done that the results will look good for this test (because trials for early detection of neurological diseases almost never pan out).

@zQ4Z82W I'm still going through it. How do you interpret a false positive, though? Given that we may be evaluating prodromal patients. Also, we already know there are multiple mechanisms leading to what we call PD (genetic/idiopathic) and that we now know the idiopathic group also has people who don't show up positive on this test. To me the fact they have a clinical diagnosis of PD suggests that they have lost dopaminergic neurons, but via another cause than AS.

I don't see how you'll be able to say that this 3.7% false positive rate isn't just explained as people who are likely to develop PD in the future.

i.e. we have people with PD but no AS (alternate mechanism), and we have people with AS but no PD (prodromal).

It still seems to me that showing up positive to this test is highly predictive of getting PD soon. But that it will not ever be able to detect people with alternate mechanism PD. But we already know that, because the response of people with genetic PD is already different than others.

This all leads me to feel that this test could already function as a biological test for earlier detection of PD in many cases.

I'm still going through, it though, so will update more later.

Do you guys know something I don't?

How early do you have in my mind for the early detection? Months/years/decades? And does the test need to be FDA approved?

@StrayClimb anything that would let us know something before the patient had serious symptoms leading to a clinical diagnosis

This is outside my expertise, but 2028 just seems too soon to me. I don't see any recent significant progress in this area, and even if there was a big discovery tomorrow, it would take years to do the necessary trials that prove the test actually works. For diseases like Parkinson's that develop slowly over several years, evaluating tests for early detection is an inherently long process because you need to test trial participants and then wait years to see who develops diagnosable symptoms and who doesn't. Maybe you get lucky and there are images/tissue samples/whatever from years ago that you can test and then see who ended up getting Parkinson's and who didn't, but even then you'd probably still need long trials to settle the matter conclusively. And if for purposes of this market you require patients to have access to the test, then regulatory approval needs to happen, and that's also a slow process.

Interesting, there's something called the Parkinson's Progression Markers Initiative that has collected specimens for verifying Parkinson's biomarkers: https://www.ppmi-info.org/access-data-specimens/request-specimens

@zQ4Z82W Yes, I think they used that to source patients for the development of the αSyn-SAA test above. There are tons of precursor symptoms for PD which people mostly don't notice, which in combination can be quite strong predictors of eventual PD diagnosis. The problem has been inability to test for them biologically.