Quote:
Originally Posted by soccertese
would you provide the links to the studies showing coq10 was neuroprotective in human cell lines?
|
http://www.sciencecodex.com/patientd...rkinsons-94469
Next, the researchers attempted to rescue the toxin-exposed cells with various drug treatments that have shown promise in animal models of Parkinson's, including the antioxidant coenzyme Q10 and the immunosuppressant rapamycin. All patient-derived neurons – whether they carried LRRK2 or PINK1 mutations – had beneficial responses to coenzyme Q10. However, the patient-derived neurons differed in their response to rapamycin; the drug helped prevent damage to neurons with LRRK2 mutations, but it did not protect the neurons with PINK1 mutations.
Quote:
Originally Posted by soccertese
if the latter was the case then i wouldn't call that simulating real pd progression but just screening possible protective compounds. can't call that proving neuroprotection.
|
The latter wasn't the case, I think. Though for me this is not clear after reading the article.
Quote:
Originally Posted by soccertese
i think the normal progression is pretreatment in cells - rat, mice, or human, then exposing the cells to a toxin, then mice/rats, then humans.
|
You forgot nonhuman primates. Besides that, your sequence has always been likte that. But now they managed to convert skin cells of patients with PD into human neurons. So from now on drug research will be different.
Quote:
Originally Posted by soccertese
as far as sample size in statistics, results always take this into account. statisticians always include in their reports sample size and the affect that had on the outcome of the results. but scientists alway make note of outliers in a clinical trial, those that had above average affects, either negative or positive. if a few patients had amazing results it isn't ignored.
|
Indeed. And because there are so many things they don't know, clinical trials on neuroprotectiveness are a bit useless unless the neuroprotectiveness is really huge. If the uncertainty on your variables is really big, then it is almost impossible to prove anything statistical if you take everything into account. In this specific case ... you have this PD score for patients. If you have to proof something is 20 % neuroprotecive ... suppose your PD score is 20. 20 % of 20 is 4. This means that if you would take something neuroprotective you should have been more or less on 16 instead of 20. But this is all open to interpretation. A patient with score 16, could be rated 18 or 22 by another doctor. And someone with a score of 20 could be rated 16 by another doctor. And then, it also depends on the patient itself. Maybe he goes to a doctor on a day in which his symptoms are less than the day before.
So if you want to prove statistically that something is neuroprotective, then the neuroprotectiveness has to be maybe 50-70 % before you can statistically claim that it is neuroprotective. And the placebo effect only adds to it.
Quote:
Originally Posted by soccertese
placebo affect is so strong in pd'ers that biomarkers would be nice to have, maybe the newer mri studies measuring dopamine neurons activity would help. you can get a baseline and measure the decline over time.
|
That would drastically improve the results of the clinical trials.
Linear Mode
