We sincerely appreciate the thoughtful comments by Nishida et al. regarding our article “Steatotic liver disease: A key related risk factor in the emergence of metabolic syndrome–related disorders” 1. Their letter, “Steatotic Liver Disease and the ‘Point of No Return’ in the Development of Metabolic Disorders,” elegantly emphasizes that steatotic liver disease (SLD) is not merely a passive hepatic phenotype but an active upstream driver of metabolic dysregulation and introduces the important notion of a potential “point of no return” along the SLD–metabolic disease axis 2. First, we concur that our definition of SLD, based on ultrasonographic detection of hepatic steatosis, encompasses individuals both with and without overt metabolic dysfunction and therefore overlaps substantially with the contemporary construct of metabolic dysfunction–associated steatotic liver disease (MASLD) 1, 6. This is consistent with our recent large multicenter health-checkup study (MIRACLE-J), in which NAFLD and MASLD diagnoses were concordant in approximately 97%–99%, indicating that findings from older NAFLD cohorts remain largely valid under the MASLD framework and underscoring the tight clustering of steatosis with cardiometabolic risk factors in real-world Japanese populations 3. In our longitudinal cohort, baseline SLD prevalence (60.9%) exceeded that of type 2 diabetes (T2D) and hypertension, supporting Nishida et al.’s view that hepatic steatosis often precedes clinically manifest metabolic disorders and may serve as an accessible early marker in health checkup programs 1-3. Second, we fully agree that binary ultrasound classification is a limitation of our study design. Conventional sonographic features, such as hepatorenal contrast, vascular blurring, and posterior attenuation, reflect hepatic fat burden but are not inherently quantitative 1. As Nishida et al. suggest, semiquantitative grading (e.g., mild, moderate, and severe) could refine risk stratification for incident T2D, hypertension, and dyslipidemia and may serve as a pragmatic bridge toward quantitative techniques such as controlled attenuation parameter (CAP) or MRI-PDFF in health-check settings 2, 6. Building on the MIRACLE-J experience, which leveraged standardized ultrasonography across multiple centers to characterize MASLD/NAFLD subtypes and noninvasive fibrosis markers 3, future analyses within our and other cohorts will aim to incorporate semiquantitative ultrasound scoring in line with current MASLD guideline recommendations 6. Third, their emphasis on body-weight dynamics is highly relevant 2. In our original report, baseline SLD was an independent risk factor for the incidence and development of T2D and hypertension over 7 years, whereas baseline obesity was not an independent predictor of incident metabolic syndrome–related diseases 1. Conversely, obesity at baseline was the only independent risk factor for new-onset SLD, suggesting a sequential pathway from obesity to SLD and then to metabolic disorders 1. MIRACLE-J further showed, in a much larger cross-sectional health-checkup cohort, that cardiometabolic clustering and alcohol-related categories (MASLD, MetALD, and alcohol-associated liver disease with metabolic dysfunction) strongly influence noninvasive fibrosis indices, such as FIB-4, reinforcing the central role of metabolic load and lifestyle factors in SLD-related risk stratification 3. Although we did not analyze weight-change categories or SLD-by-weight-change interactions, the concrete proposals by Nishida et al.—classification into weight-loss, stable, and weight-gain groups within SLD strata and formal testing of interaction terms—are both feasible and clinically meaningful and will be considered in subsequent analyses of our dataset 2. Fourth, we appreciate the nuanced discussion by Nishida et al. on SLD regression and the “point of no return” 2. In our study, individuals whose SLD regressed over 7 years had lower incidences of T2D, hypertension, and dyslipidemia than those with persistent SLD, indicating that improvement in steatosis is associated with a more favorable metabolic trajectory 1. Nevertheless, as they highlight, a substantial proportion of participants with SLD regression still developed T2D or hypertension, raising the possibility that a history of SLD may identify a persistently high-risk phenotype even after apparent hepatic recovery 1, 2. Several nonmutually exclusive explanations are plausible, including (a) irreversible or slowly reversible pathophysiologic programming once the SLD–metabolic disease pathway is activated, (b) apparent irreversibility related to long intervals between measurements, and (c) residual constitutional or genetic susceptibility not captured by routine clinical variables 2, 9, 10. Because our design relied on two time points, we cannot distinguish these mechanisms or define a precise temporal “point of no return” and we concur that higher-frequency follow-up will be required to delineate windows of reversibility 2. Fifth, we share Nishida et al.’s view that the relationship between SLD and T2D is bidirectional and self-reinforcing 2, 5. In our cohort, baseline SLD significantly increased the risk of incident and new-onset T2D, whereas baseline T2D contributed little to the incidence of SLD within 7 years, suggesting that SLD lies upstream of T2D onset whereas diabetes exerts stronger effects on disease progression than on initial steatosis 1. Their clinical work on the prognostic value of the oral glucose tolerance test in cirrhosis, and subsequent reviews has clearly shown that even subclinical glucose intolerance adversely affects liver-related and overall survival in chronic liver disease 4, 5. Together with recent data from Japanese and international cohorts linking MASLD/NAFLD to incident hypertension and cardiovascular disease 3, 7, 8, 10, these findings support a model in which SLD both precedes and is later exacerbated by metabolic disorders in a vicious cycle. In this context, we agree, as Nishida et al. emphasize, that SLD should be regarded as a pivotal early intervention target rather than a mere hepatic bystander 2, 3, 5-7. Finally, our longitudinal data support the conceptual sequence proposed in our article—obesity → SLD → metabolic syndrome–related diseases—whereas MIRACLE-J complements this by showing, at a national health-checkup scale, that MASLD, MetALD, and alcohol-associated liver disease with metabolic dysfunction carry graded risks of advanced fibrosis as assessed by FIB-4 1, 3. These converging lines of evidence reinforce Nishida et al.’s important message that early identification and treatment of SLD in health-check programs may delay or prevent the onset of T2D, hypertension, and dyslipidemia and improve long-term hepatic outcomes 1-3, 5-7. We are grateful to Nishida et al. for their insightful comments and for highlighting the need to integrate hepatologic and cardiometabolic perspectives in the care of individuals with SLD 2, 3, 5. The author has nothing to report. This study was funded by JSPS KAKENHI grants (Grant Numbers 22H02967 and 25K11274) from the Japan Society for the Promotion of Science. This study was approved by the Institutional Review Board of the Osaka International Cancer Institute (approval number: A2022002). The author has nothing to report. Yoshihiro Kamada is an Editorial Board member of Hepatology Research. The author has nothing to report. The author has nothing to report. The data that support the findings of this study are available from the corresponding author upon request. The data are not publicly available due to privacy or ethical restrictions.
Yoshihiro Kamada (Sat,) studied this question.