This is the section that matters most, and it is the one most consumer pages skip. Of the MOTS-c records in the source set behind this article, exactly one is indexed as involving human subjects, and reading it carefully shows why that label needs qualifying.
Kumagai and colleagues (2021, American Journal of Physiology: Endocrinology and Metabolism) reported that plasma MOTS-c levels are inversely correlated with myostatin levels in human subjects. That is an observational correlation in blood samples. The interventional parts of the same paper were done elsewhere: MOTS-c prevented palmitic-acid-induced atrophy in differentiated C2C12 myotubes, which is a mouse cell line in a dish, and MOTS-c administration decreased plasma myostatin in diet-induced obese mice. The authors describe a CK2-PTEN-mTORC2-AKT-FOXO1 pathway and conclude that MOTS-c “could be a potential therapy.”
No one was treated. No outcome was measured in a person. Calling that paper human evidence, without saying which part of it was human, is the single most common error in MOTS-c content online.