Hacker News new | past | comments | ask | show | jobs | submit login

This is a good question:

> Horvath used four different epigenetic clocks to assess each patient’s biological age, and he found significant reversal for each trial participant in all of the tests.

Presumably the "epigenetic clocks" are correlated with age, but it's not clear if changes to the epigenetic clocks causes aging, or aging causes changes to the epigenetic clocks. (Maybe it's both?) Is this like overhauling the engine in an old car to make it function like new, or is it more like rolling back the odometer to make it seem new?




Exactly. And this is why the FDA won't approve drugs based on changes to biomarker (which is what these "epigenetic clocks" are) unless a direct link between that biomarker and a beneficial outcomes has been proven.


unless its for Alzheimer's? [0]

'The FDA’s eventual decision to ignore the advice of its advisory committee and approve the drug, it says, was based on aducanumab’s ability to lower levels of amyloid plaques in the brain — protein clumps that some scientists think cause Alzheimer’s.'

...

With aducanumab, the agency has shown that it is willing to push the paradigm to a broader set of people. One reason that decision has attracted criticism is that decreased levels of amyloid plaques are an unvalidated and contentious marker of a drug’s activity.

In large trials of other Alzheimer’s drug candidates, amyloid lowering has not led to cognitive benefits, and this has made it a sticking point for researchers.' [0]

[0] https://www.nature.com/articles/d41586-021-01763-9


Indeed! The Alzheimer’s drug was an Accelerated Approval. Basically conditional, based on a “likelihood” that the biomarker is indicative of a benefit, but requiring further study to prove benefit.


I wonder if those amyloid lowering drugs could have a prophylactic effect while having no curative effects. I would love to see a trial in people at high risk of Alzheimer...


They are already doing those and they aren’t working out too well either.


And human lifespan, the best meaningful outcome, is too long for any study to use.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: