Dissecting fake news and violations of scientific ethics in scientific publications: analysis of the article “Hyperglycosylation is a metabolic driver of Alzheimer's disease”

Glucosamine in the form of glucosamine sulfate is used in the treatment of osteoarthritis, including in patients with concomitant diabetes, obesity, chronic ischemia, and neurodegenerative diseases (such as Alzheimer's disease). In addition to its chondroprotective effects, glucosamine sulfate is well known to exert hepatoprotective and neuroprotective effects. Despite the considerable body of basic and clinical research on this subject, “unexpected” articles occasionally appear in scientific journals published by reputable publishers that contradict not only the existing body of evidence but also common sense. Such publications often generate undesirable media attention, which may ultimately have a profoundly negative impact on patient care. This article presents the results of a detailed analysis of one such article, published in “Nature Metabolism” in 2026. Expert analysis identified the following: (1) the absence of verifiable data indicating that the reported experiments were actually performed; (2) apparent ethical violations in the reported medical procedures, including claims that frontal cortex samples were collected from patients with Alzheimer's disease; (3) numerous logical and quantitative inconsistencies in the reported results of biochemical and experimental studies; (4) major errors in the clinical characterization and analysis of the reported patient cohorts (e.g., 10-year dynamics over 5 years of observation); (5) formulaic use of vague or overly general terms throughout the article, including in the Results (“hyperglycosylation”, “glycosylation phenotype”, etc.); (6) apparent statistical artifacts (e.g., increases of exactly 25% in a wide variety of outcomes, including mortality from Alzheimer's disease due to glucosamine use); and (7) insufficient knowledge of the biochemistry of protein glycosylation on the part of the authors. It should be emphasized that entire sections of the article under review contain numerous logical errors and internally contradictory statements that can be summarized as “the toxic effect of a nontoxic substance” or “the detoxifying effect of a highly toxic effect”. It is hypothesized, based in part on the text of the publication itself, that the article published in “Nature Metabolism” was prepared using information technologies such as ChatGPT.