Elevated ATG13 in serum of pwME stimulates oxidative stress response in microglial cells , 2022, Gottschalk et al

Recent overview paper on autophagy: Life and Death Decisions-The Many Faces of Autophagy in Cell Survival and Cell Death (2022)

 

From the Simmaron Research blog: Cleaning Crisis? Is Defective Mitochondrial Cleanup Impairing Energy Production in ME/CFS? (Sep 2022).

 

From the Simmaron Research blog: Cleaning Crisis? Is Defective Mitochondrial Cleanup Impairing Energy Production in ME/CFS? (Sep 2022):

This is interesting but, after all the effort that has been put into highlighting the methodological problems and unscientific reasoning applied to BPS research, I wish people would be more careful in their reporting of recovery anecdotes. I’ve not read this story on HR, but it is misleading to say that “an MTOR inhibitor significantly helped at least one long-term ME/CFS patient”. All we can know is that one patient reported that they had recovered after taking rapamycin, an MTOR inhibitor. But without clinical trials we cannot know whether rapamycin helped.

Something to add to your Something in the Blood blog @Simon M?

 

I hope they have good reason to hype this up to this level.

"We’re onto something big: a treatable pathway for PEM. The implications are huge. Not only do we believe the chemical pathway involving the ATG-13 protein is a culprit in PEM, we believe it can be targeted for drugs. There is hope for treatment!"

 

They found something new in a small study...that a whopping two people with ME have elevated levels of ATG13 and that microglia don't like it. This is a brainstorm, not a real hope. While their research appears novel, all scientists should present their findings realistically, especially if they're a non-profit soliciting donations.

A reasonable threshold for a "real hope" of treating PEM would be a drug entering human trials.

 

Did anyone get a reply about this inconsistency?

Below is figure 3.A as used in the paper:


And this is my own plot of the data shown in table 1:

 

This is the sample size determination as stated in the paper.

"For 99% confidence interval and 0.05 significance, our sample size calculation is n = z2×p (1− p) / ε2 = 1.282×0.99 (1− 0.99) / 0.05^ 2 = 7. Z is the z score, which is 1.28 for power 0.8; p is the population proportion. For 99% confidence interval, p will be 0.99; ε is the margin of error = 0.05. Therefore, throughout the study we selected at least n = 7 per group when comparing results between HC and ME groups."
​

To me this does not make sense. A power calculation requires an effect size.

The author probably misinterpreted a formula he found online or in a textbook: p should be a proportion of the population, e.g how many people have a disease. Setting p to 99% because one wanted a confidence interval of 99% seems like a quite a big statistical misunderstanding.

If the authors conducted a proper power calculation, they would probably found that only incredibly large effects can be reliable detected with 7 participants per group.

 

Did you email them?

 

No did look into it at the time of publication.

 

See also Kaposi’s sarcoma-associated herpesvirus (KSHV) utilizes the NDP52/CALCOCO2 selective autophagy receptor to disassemble processing bodies (2023)