Metabolic dysfunction-associated steatotic liver disease (MASLD) affects up to 38% of the population globally. Liver biopsy, the historical reference standard, is invasive, subject to sampling error and not scalable. MRI-based biomarkers provide non-invasive, quantitative whole liver assessment. MRI proton density fat fraction (MRI-PDFF) is the most accurate noninvasive measure of steatosis (thresholds of 6.4%, 17.4% and 22.1% for grades ≥1, ≥2 and 3), and a ≥30% relative reduction is associated with histologic response. Liver stiffness by magnetic resonance elastography (MRE-LS) is the most accurate imaging biomarker for fibrosis: MRE-LS of 3.1-4.4 kPa corresponds to F2-F3 and ≥5 kPa to cirrhosis, and each 1-kPa increase is associated with a 43% higher risk of hepatic decompensation. Iron-corrected T1 is a sensitive but nonspecific marker of disease activity whose prognostic value requires further study. Composite scores, MAST, MEFIB, and MASH Resolution Index, combine imaging and blood tests for advanced fibrosis diagnosis and response assessment. MRE is cost-effective compared with vibration-controlled transient elastography for detecting advanced fibrosis. Artificial intelligence may broaden access by automating analysis and enabling PDFF from routine MRI. We review the evidence supporting these biomarkers for diagnosis, prognosis and regulatory use, and their role in clinical trials and in treatment decisions for approved MASH therapies.
MRI proton density fat fraction (MRI-PDFF) is the most accurate noninvasive measure of steatosis (thresholds of 6.4%, 17.4% and 22.1% for grades ≥1, ≥2 and 3), and a ≥30% relative reduction is associated with histologic response.
heterogeneous biopsy-validation cohorts and small treatment trials support diagnostic accuracy and histologic-response associations
Liver stiffness by magnetic resonance elastography (MRE-LS) is the most accurate imaging biomarker for fibrosis
prospective and pooled biopsy comparisons favor MRE, although head-to-head confidence intervals overlap
The risk of hepatic decompensation rises by 43% for each 1-kPa increase in MRE-LS.
six-cohort individual-participant meta-analysis with 119 events directly supports the stated association
Iron-corrected T1 is a sensitive but nonspecific marker of disease activity whose prognostic value requires further study.
variable diagnostic validation and event-limited outcome cohorts support the nonspecific characterization and explicit prognostic uncertainty
MRE is cost-effective compared with vibration-controlled transient elastography for detecting advanced fibrosis.
selected US and Canadian economic models do not establish unrestricted cost-effectiveness across health systems
Derived from the full evaluation — not a separate score.
Strengths
The review contrasts derivation estimates with prospective LITMUS findings and gives external validation greater weight. Tables expose lower validation performance rather than presenting only favorable estimates.
↳ Comparative-performance LITMUS discussion; Tables 1–2
Diagnostic accuracy, pharmacodynamic response, prognostic association, and surrogate-endpoint qualification are treated as different evidentiary questions. This distinction clarifies what current MRI evidence supports and what remains unestablished.
↳ MRI-PDFF section; cT1 section; regulatory section
The review maps MRE evidence against the FDA's three surrogate-endpoint considerations and distinguishes acceptance of a letter of intent from qualification. Missing outcome-linked change thresholds are explicitly identified.
↳ Imaging Biomarkers in Clinical Trials – Regulatory Considerations and Unmet Need
Limitations
No search strategy, eligibility criteria, or source-selection account accompanies the comparative synthesis. Readers cannot assess completeness despite the visible engagement with contrary evidence.
↳ Full text, including Introduction and review sections
The abstract states MRE cost-effectiveness without qualification, while the supporting economic models concern selected US and Canadian pathways. The body acknowledges that capacity and reimbursement can make those conclusions inapplicable elsewhere.
↳ Abstract; Cost Effectiveness for Identifying Advanced Fibrosis and Availability
Treatment-induced imaging changes have not been linked to clinical benefit through established prospective thresholds. Existing diagnostic validation therefore does not establish readiness to replace clinical outcomes with imaging surrogate endpoints.
↳ Imaging Biomarkers in Clinical Trials – Regulatory Considerations and Unmet Need; Future Directions
The review's main contribution is its separation of MRI biomarkers' established diagnostic roles from their incompletely validated roles in treatment monitoring and surrogate endpoints. Prospective LITMUS findings, external-validation attenuation, and negative treatment results receive substantive attention, supporting the contextual assessment. The absence of a source-selection method limits confidence that the narrative evidence base is complete. Clear organization and synthesis tables aid interpretation, but the abstract's cost-effectiveness statement extends beyond the selected US and Canadian models supporting it. The NIS2+ acceptance-rule ambiguity and denifanstat citation error are reporting notes rather than contradictions of the central MRI findings.
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Strong3.8
Confidence mediumThe review integrates diagnostic performance, longitudinal response, prognosis, implementation, and regulatory qualification into distinct biomarker roles. Its contribution is a current interpretive synthesis rather than a new method or resolution of surrogate-endpoint uncertainty.
The synthesis weighs prospective validation against derivation studies and incorporates null findings, sparse events, and conference-abstract limitations. The absence of a source-identification or selection account limits assessment of completeness.
Modality-based sections and synthesis tables distinguish diagnostic, pharmacodynamic, prognostic, and regulatory uses. The abstract's unrestricted economic claim is broader than the geographically bounded models described in the body.
Prospective LITMUS results, attenuated external validation, null treatment findings, and unresolved surrogate qualification are traced into interpretation. Economic pathway coverage and AI appraisal are less developed than the review's diagnostic and regulatory contextualization.
Caveats4 of 4 checks
One reported value does not reconcile, but the key claims do not rest on it.
Table 5 assigns the denifanstat trial the wrong reference number. Financial interests and funding are explicitly disclosed.
Conflict of interest: declared
157 of 157 checkable references verified
167 references in manuscript 10 have no canonical index record — counted, but not index-checkable 1 reference confirmed by manual review
References not verified automatically are resolved manually within 24 hours.
No retraction or expression of concern found in Crossref (includes Retraction Watch).
Sources: Crossref (incl. Retraction Watch) ✓
Where this paper’s evidence sits on the path from initial observation to real-world use.
The Validated readiness stop reflects prospective and biopsy-linked validation and existing clinical guidance for MRI biomarkers. Unstandardized serial thresholds and the absence of treatment-change thresholds linked to clinical benefit prevent equating this validation with qualified surrogate-endpoint use.
AI-generated, human-governed. Something look off? Contact us to request a review.
Keep reading
Other papers Nabu has evaluated blind, starting with the same field.
Majani, Chillo, Akida, et al. · 2025 · Clinical medicine
POX demonstrated a sensitivity of 50.0% (95% CI: 32.1-67.9), a specificity of 99.5% (95% CI: 99.4-99.7), a false-positive rate of 0.4%, and an overall accuracy of 99.5% (95% CI: 99.2-99.5)
TentativeNakamura, Yamazaki, Namikawa, et al. · 2026 · Clinical medicine
ORR by BICR was 22.6% (7/31), and the disease control rate was 54.8% (17/31).
SupportedDance, Weyman, Adams, et al. · 2026 · Health sciences
Systemic constraints, related to limitations of data availability and access and organisational capacity for prevention, were perceived by participants to affect the development of public health interventions to reduce alcohol use.
Supported