Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial

[Mike Lustgarten]

https://www.nature.com/articles/s43587-022-00357-y

[Saul]

The subjects were overweight; it's more a study of the positive effects of reducing obesity.

  --  Saul

[Mike Lustgarten]

Arguably the most important data in that study is that the DunedinPACE test is the only epigenetic test to identify a biological age reduction for people on CR, relative to other epigenetic tests. That data is presented in this video:

 

[Mike Lustgarten]

23 hours ago, Saul said:

The subjects were overweight; it's more a study of the positive effects of reducing obesity.

  --  Saul

If "the subjects" means all, then that's not true: 107 subjects had a BMI of 22 - 24.9 (not overweight), whereas 113 had a BMI of 25 - 27.9 (overweight).

So about half were overweight in both the CR and control groups.

[kpfleger]

The idea that only the DunedinPACE test detected the benefit is wrong or at least misleading. The effect size was so small that detecting it via point-estimates (the kind of other tests used) was within the noise. Ie, such a small benefit is not detectable by point-estimate tests. One would need a bigger effect size (eg by doing the intervention longer or using higher % CR or using a better intervention). See this thread on Twitter (which Mike was also a participant in): 

 

[Mike Lustgarten]

idk why you've only posted where I commented in that thread, but your argument is with Belsky, not me, which happened in February. Why are you rehashing this now?

Full thread is here:
 

 

[Alex K Chen]

Mice have some epigenetic age reductions from CR... (see John Speakman ARDD). Idk which clocks

Edited December 8, 2023 by InquilineKea

[kpfleger]

Mike, I posted here just for completeness. I don't check these forums often but above in this thread you wrote, "Arguably the most important data in that study is that the DunedinPACE test is the only epigenetic test to identify a biological age reduction for people on CR, relative to other epigenetic tests" which gives the impression that only DunedinPACE was "good" (if one believes in CR, as many here do), so I thought it was important for those here to have a pointer to that Twitter thread in which these issues were well hashed out. Twitter thread forks make it hard to find the exact right Tweet to link to so I did best I could linking to one where the key points were right below but one could scroll up for full context. The key points in the next few Tweets from the one I linked were Olafur agreeing with the Tweet I linked that PCGrimAge wouldn't be able to detect a real CR effect in this trial design, me saying 'So doesn't that invalidate the "I think the fact that PC GrimAge was not improved by 24 month of CR is strongly indicative of that clock not being good" sentiment?' in response to an upthread Tweet and then him agreeing. That was the main point.