Circadian metabolic dysfunction in MASLD: Physiological insights and future directions Authors: You‐Cheol Hwang Published: 2026 Metabolic dysfunction‐associated steatotic liver disease (MASLD) 1 has traditionally been assessed using static, daytime measurements. However, growing evidence from circadian biology suggests that the timing of metabolic flux may be as important as its magnitude 2 . In a recent study published in Cell Metabolism, Marjot et al. 3 reported that key metabolic abnormalities in MASLD were more pronounced at night. Using paired daytime and nighttime hyperinsulinemic–euglycemic clamp studies combined with stable isotope tracers, the authors identified nocturnal hepatic insulin resistance, sustained de novo lipogenesis (DNL), unsuppressed adipose tissue lipolysis, and impaired skeletal muscle glucose disposal (R d). These day–night differences persisted after a 12‐week weight loss intervention that reduced hepatic steatosis, as assessed by the controlled attenuation parameter, indicating that nocturnal metabolic dysfunction may not simply be a consequence of liver fat accumulation. These paired daytime and nighttime clamp studies further revealed that nocturnal metabolic dysfunction in MASLD involves multiple organs. Specifically, although the basal insulin secretion rate (ISR) was higher in MASLD, consistent with the underlying insulin resistance, the insulin‐stimulated increase in ISR was reduced by more than 66% at night, whereas systemic insulin clearance increased. Together, these changes reduced insulin availability at night, at a time when insulin resistance in peripheral tissues and the liver was increased. Although nocturnal 3‐hydroxybutyrate (3OHB), a marker of β‐oxidation, was increased, plasma non‐esterified fatty acids (NEFA) remained persistently elevated, suggesting that the accompanying rise in fatty acid oxidation was insufficient to compensate for excessive nocturnal lipid flux. Targeted plasma proteomics also showed higher nighttime levels of growth differentiation factor 15 (GDF‐15), a candidate biomarker of nocturnal nutritional stress that correlated with circulating 3OHB, together with morning peaks in tissue‐type plasminogen activator (tPA) and plasminogen activator inhibitor‐1 (PAI‐1), and all these findings were independently replicated in the UK Biobank. A major strength of this study was the integration of several complementary physiological approaches. Specifically, the authors combined paired daytime and nighttime hyperinsulinemic–euglycemic clamps, stable isotope tracers, breath 13CO2 enrichment and substrate oxidation measurements, targeted plasma proteomics (a 691‐protein Olink panel), untargeted adipose and skeletal muscle proteomics, and a longitudinal 12‐week dietary intervention, with external validation in the UK Biobank dataset. The evidence of multisystem impairment came primarily from functional clamp‐derived data, which demonstrated nocturnal abnormalities in liver, adipose tissue, and skeletal muscle. In contrast, tissue proteomics in the MASLD group identified only one borderline‐significant protein in adipose tissue and none in skeletal muscle. The tissue proteomics findings should therefore be considered exploratory rather than independent evidence of multisystem dysfunction. The UK Biobank analysis provided additional population‐level data for the plasma proteomic findings but relied on morning and evening sampling windows rather than true day and night measurements and identified MASLD from ICD‐10 codes, which may underestimate its prevalence. METHODOLOGICAL CONSIDERATIONS AND UNANSWERED QUESTIONS Several unanswered questions and methodological limitations of this study should be considered:Biomarker correlations and lipid oxidation Biomarker correlations and lipid oxidation Biomarker correlations and lipid oxidation Plasma proteomics identified GDF‐15 as a candidate biomarker of nocturnal nutritional stress that correlated with 3OHB levels. However, this analysis focused primarily on the association between GDF‐15 and 3OHB and did not examine the relationship of GDF‐15 with key clamp‐derived indices, including suppression of hepatic glucose production, the M/I ratio, muscle R d , and NEFA suppression. The morning peaks in tPA and PAI‐1 also raise the question of whether these diurnal changes are associated with cardiovascular risk in MASLD. Future studies should determine whether nocturnal GDF‐15 elevation is merely a marker of increased fatty acid oxidation or is associated with nocturnal metabolic abnormalities that could contribute to the development of MASLD.2Causality and the weight loss findings 2Causality and the weight loss findings Causality and the weight loss findings Nocturnal metabolic abnormalities persisted despite weight loss of approximately 6.3% and an associated improvement in hepatic steatosis, suggesting that nocturnal metabolic dysfunction is not simply secondary to liver fat accumulation. Indeed, this finding raises the possibility that nocturnal metabolic dysfunction may contribute to the development of MASLD. Notably, weight loss improved daytime but not nighttime R d, widening the day–night gap in muscle glucose disposal, indicating that nocturnal skeletal muscle dysfunction may persist despite improvement in hepatic steatosis. Establishing causality requires interventions that specifically improve nocturnal metabolic dysfunction, together with assessment of whether hepatic steatosis improves as a result, ideally without substantial changes in body weight or adiposity.3Study design considerations 3Study design considerations Study design considerations Although short naps were permitted, breath sampling every 30 min kept participants largely awake during the nighttime clamp, precluding normal sleep architecture. Consequently, the reported “nocturnal” defects reflect metabolism during a largely wakeful nighttime protocol rather than physiological sleep, and their relevance to usual overnight physiology remains uncertain. Both clamp visits were conducted after a 12‐h fast; therefore, the findings cannot be extrapolated to postprandial responses, an important limitation when considering the metabolic effects of evening meals. Finally, because insulin was infused peripherally rather than via the portal vein, hepatic insulin exposure during the study may have differed from physiological conditions. As the authors acknowledge, this may have blunted the suppression of endogenous glucose production and lowered measured insulin clearance relative to physiological portal insulin exposure.4Cohort and statistical power 4Cohort and statistical power Cohort and statistical power The control group had a substantially lower body mass index (BMI) than the MASLD group (27.5 vs 35.5 kg/m2). Although the authors acknowledge that BMI‐matched non‐steatotic controls are difficult to recruit, this imbalance nonetheless confounds the effects of adiposity with those of steatosis. Liver histology was not required for enrollment; therefore, the contribution of inflammation and fibrosis to the diurnal phenotype remains uncertain. The modest sample size (n = 12 per group), although adequately powered for the primary DNL outcome, limits sex‐specific analyses and leads to difficulties in distinguishing true equivalence from insufficient power in nonsignificant secondary comparisons. Larger sex‐balanced studies that include patients with biopsy‐proven metabolic dysfunction‐associated steatohepatitis or fibrosis and allow physiological sleep, ideally with objective sleep assessment, are required to address these limitations. CLINICAL IMPLICATIONS AND FUTURE DIRECTIONS Because these metabolic abnormalities are most pronounced at night, interventions timed to circadian metabolic rhythms warrant further investigation. Dietary approaches, such as early time‐restricted eating or avoidance of late‐night meals, may reduce overnight hepatic DNL and adipose lipolysis, consistent with evidence that meal timing influences glucose and lipid metabolism 4 . The effects of exercise timing should also be investigated to determine whether evening exercise improves overnight skeletal muscle glucose disposal. Future studies should also explore pharmacological strategies that could preferentially improve nocturnal insulin resistance, including optimal timing of drug administration 5 . Overall, this study suggests that the pathophysiology of MASLD depends on the magnitude of metabolic dysfunction and its timing. The next step will be to determine whether selectively improving nocturnal insulin resistance could alter the natural history of MASLD (Figure 1). Nocturnal metabolic dysfunction in MASLD: from observation to causality. DISCLOSURE The author declares no conflicts of interest related to this work. Approval of the research protocol: N/A. Informed consent: N/A. Approval date of registry and registration no.: N/A. Animal studies: N/A.