Nocturnal Melatonin Amplitude Collapse Is Associated with Age-Independent Convergence of Microbiome and Glymphatic...

Circadian desynchronization is increasingly linked to metabolic, immune, neurocognitive, and oncological disease, but integrated clinical phenotyping across endocrine, microbiome, metabolic, and neuroimaging domains remains limited. We conducted a prospective, single-centre, observational study in 179 symptomatic patients referred for chronic multisystem features consistent with circadian dysregulation and 107 practically healthy controls. Circadian melatonin status was assessed using fractionated 24 h urinary 6-sulfatoxymelatonin (aMT6s) and standardized dim-light plasma sampling at daytime (14:00-16:00) and nocturnal (02:00-04:00) windows. Microbiome composition was assessed by 16S rRNA sequencing, urolithin A by targeted metabolomics, and putative glymphatic/perivascular clearance by MRI-derived DTI-ALPS index, perivascular space scoring, and white-matter-hyperintensity (WMH) volumetry. Patients showed markedly reduced nocturnal melatonin output and loss of day-night contrast (night aMT6s 10.2 vs. 40.6 ng/mL; urinary aMT6s day/night ratio 0.81 vs. 0.14; plasma nocturnal melatonin 12.7 vs. 54.4 pg/mL; all p < 0.0001), accompanied by altered cortisol day-night patterning. Patients also showed reduced microbiome diversity, depletion of Gordonibacter and Ellagibacter, lower plasma urolithin A, higher WMH volume and perivascular space scores, and a lower DTI-ALPS index. Age distributions broadly overlapped in the individual-level dataset, and key biomarkers were not significantly correlated with chronological age within the patient cohort; however, this finding is interpreted as an exploratory absence of detectable age gradient within the symptomatic cohort, not as proof of biological age-independence. Overall, the data support a coherent cross-sectional association among blunted nocturnal melatonin rhythmicity, dysbiosis/urolithin depletion, and MRI markers compatible with impaired perivascular clearance. The MGM axis framework should be regarded as hypothesis-generating; causal direction, melatonin receptor involvement, and AQP4-related mechanisms require longitudinal and mechanistic validation.