Rezdiffra (Resmetirom) Evidence Base Graded by GRADE

What Is the Overall GRADE Evidence Quality for Resmetirom?

The GRADE framework assigns evidence to four levels: high, moderate, low, or very low, based on study design, risk of bias, inconsistency, indirectness, imprecision, and publication bias. For resmetirom's two co-primary histological endpoints, the evidence sits at moderate-to-high quality. The key MAESTRO-NASH trial was a well-conducted phase 3 RCT with low risk of bias, and both co-primary endpoints reached pre-specified thresholds with P<0.001. [1]

The headline caveat is that the FDA approved resmetirom on histological surrogates, not on hard clinical outcomes such as liver-related mortality, transplant-free survival, or cardiovascular events. Under GRADE rules, surrogate-endpoint-only evidence rarely escapes a one-level downgrade for indirectness, which is why some endpoints land at GRADE moderate rather than high.

Why Surrogate Endpoints Matter in GRADE

GRADE explicitly differentiates patient-important outcomes (mortality, cirrhosis decompensation, transplant) from surrogate markers (histological scores, imaging biomarkers). Histological NASH resolution and fibrosis regression are reasonably well-validated surrogates, but the correlation with hard outcomes has not yet been prospectively confirmed for resmetirom specifically. [2]

The FDA's accelerated approval pathway for MASH drugs accepts "histological endpoints reasonably likely to predict clinical benefit," acknowledging this evidentiary gap. [3]

Applying the GRADE Domains to MAESTRO-NASH

MAESTRO-NASH was a randomized, double-blind, placebo-controlled trial across 243 sites in 22 countries. The 966-patient sample provided >90% power to detect the pre-specified 10 percentage-point between-arm difference for each co-primary endpoint. Allocation concealment was adequate, and central biopsy reading used a blinded committee, both of which reduce risk of bias. Consistency was supported by directionally concordant results across liver enzymes, MRI-PDFF fat fraction, and lipid panels, though a second independent RCT has not yet been published for confirmation. [1]

MAESTRO-NASH: The Key Phase 3 Trial

MAESTRO-NASH enrolled adults with biopsy-confirmed MASH, NAS ≥4, and fibrosis stage F1B-F3. Patients were randomized 1:1:1 to resmetirom 80 mg, resmetirom 100 mg, or placebo daily for 52 weeks. [1]

Co-Primary Histological Endpoints

NASH resolution without worsening of fibrosis occurred in 25.9% of the 80 mg group and 29.9% of the 100 mg group, compared with 9.7% in the placebo group (P<0.001 for both doses). [1] Fibrosis improvement of at least one stage without worsening of NAS occurred in 24.2% (80 mg) and 25.9% (100 mg) vs. 14.2% placebo (P<0.001 for both). [1]

These absolute risk differences of approximately 16 and 12 percentage points for the two endpoints, respectively, translate to numbers needed to treat of roughly 6 and 8, which are clinically meaningful figures for a disease with limited prior treatment options.

Secondary and Exploratory Endpoints

Resmetirom also reduced MRI-PDFF-measured hepatic fat fraction by 34% (80 mg) and 38% (100 mg) relative to baseline versus 5% with placebo at week 24. [1] LDL-cholesterol fell by 13.5% and 16.3% in the 80 mg and 100 mg arms, which is relevant because MASH commonly co-exists with dyslipidemia. [1]

Liver stiffness measured by MRE (magnetic resonance elastography) decreased significantly in both active arms, adding mechanistic coherence to the biopsy-based findings. [4]

What MAESTRO-NASH Does Not Answer

The trial ran 52 weeks. Long-term durability of histological benefit, rate of cirrhosis prevention, and effect on liver-related clinical events remain unknown. A long-term outcomes extension (MAESTRO-NASH OUTCOMES) is ongoing, targeting cirrhosis-free survival as its primary endpoint. Until that trial reports, clinicians must prescribe resmetirom knowing that a definitive outcomes anchor is absent.

GRADE Rating Table: Endpoint-by-Endpoint Breakdown

The following framework grades each major resmetirom endpoint using the five GRADE domains applied to currently available evidence.

NASH resolution (histological, 52 weeks)

• Study design: RCT (starts at HIGH)
• Risk of bias: Low (blinded central biopsy reads, adequate allocation concealment)
• Inconsistency: No second RCT yet, one-level consideration; mitigated by consistent secondary data
• Indirectness: Surrogate, not mortality. One-level downgrade applied.
• Imprecision: N=966, narrow confidence intervals
• Final GRADE: MODERATE

Fibrosis improvement ≥1 stage (histological, 52 weeks)

• Study design: RCT (starts at HIGH)
• Risk of bias: Low
• Inconsistency: Consistent across MRE and liver enzyme data
• Indirectness: Surrogate. One-level downgrade applied.
• Imprecision: Adequate sample size, tight CIs
• Final GRADE: MODERATE

LDL-C reduction (lipid biomarker)

• Study design: RCT (starts at HIGH)
• Indirectness: Lipid surrogate, no CVOT data
• Final GRADE: LOW (two surrogate layers: lipid as surrogate for CV events, CV events as surrogate for mortality)

Hard liver outcomes (cirrhosis decompensation, transplant, liver-related death)

• No RCT data available. MAESTRO-NASH OUTCOMES ongoing.
• Final GRADE: VERY LOW (currently extrapolated from surrogate histology)

Cardiovascular hard outcomes

• No CVOT published. TR-β selectivity was designed to avoid cardiac TR-α stimulation.
• Final GRADE: VERY LOW (preclinical selectivity data only)

Mechanism of Action and Why TR-β Selectivity Affects Evidence Interpretation

Resmetirom acts as a selective agonist of thyroid receptor beta (TR-β), the dominant thyroid receptor isoform in the liver. By activating TR-β rather than TR-α (the predominant cardiac and bone isoform), resmetirom aims to reproduce the beneficial hepatic effects of thyroid hormone, including increased fatty acid oxidation, decreased lipogenesis, and enhanced LDL clearance, while minimizing cardiac and skeletal risks. [5]

TR-β Selectivity: Preclinical vs. Clinical Evidence

In preclinical models, TR-β selectivity correlated with hepatic fat reduction and minimal heart-rate or bone-resorption effects. In MAESTRO-NASH, heart rate and bone mineral density were monitored as safety signals. No clinically significant changes in resting heart rate were observed at either dose over 52 weeks, supporting but not confirming the selectivity hypothesis in humans at the approved dose range. [1]

This distinction matters for GRADE: mechanistic plausibility can upgrade evidence only when the mechanistic pathway is well-established and the surrogate has been validated. TR-β/hepatic fat is plausible but not yet validated as a causal chain to hard liver outcomes in large prospective data.

Thyroid Axis Effects

TSH levels remained stable in MAESTRO-NASH, with free T4 showing a modest suppression consistent with peripheral thyroid hormone mimicry. [1] The FDA label notes that resmetirom is contraindicated in patients with decompensated cirrhosis and in pregnancy (Category X equivalent under the new labeling format). [3]

Safety Evidence and GRADE Considerations

Safety data from a single RCT carry GRADE low-to-moderate quality for common adverse events (adequate N, limited duration) and GRADE very low for rare adverse events (underpowered for events with incidence <1%).

Common Adverse Events

In MAESTRO-NASH, the most frequent adverse events were gastrointestinal: nausea occurred in 17% (80 mg), 19% (100 mg), and 9% placebo; diarrhea in 17%, 22%, and 10%, respectively. [1] These events were mostly mild-to-moderate and transient, concentrated in the first 4-8 weeks of treatment.

Hepatotoxicity Signal

Resmetirom carries a labeling warning for drug-induced liver injury (DILI). ALT and AST elevations >5x ULN occurred in approximately 3% of treated patients in MAESTRO-NASH. [1] Monitoring liver enzymes at baseline, 3 months, and 6 months is specified in the FDA label. [3] This monitoring requirement reflects genuine uncertainty about hepatotoxicity rates in broader real-world populations where biopsy-grade disease severity may differ from the trial population.

Drug Interactions

Resmetirom is a P-glycoprotein inhibitor and may increase exposure of P-gp substrates including certain statins. The FDA label specifically notes that rosuvastatin dose should not exceed 20 mg daily during resmetirom therapy. [3] This interaction is relevant given that dyslipidemia often co-exists with MASH.

Guideline Positions and Expert Consensus

The American Association for the Study of Liver Diseases (AASLD) 2023 guidance on MASH states: "Pharmacologic therapy for NASH should be considered in patients with NASH and significant fibrosis (stage F2-F3) or those at high risk for disease progression." [6] Resmetirom received accelerated FDA approval under this framework before the AASLD guidance was updated to incorporate the 2024 MAESTRO-NASH results.

The AASLD practice guidance also specifies: "The presence of NASH with fibrosis stage ≥2 is generally considered the threshold at which pharmacotherapy provides sufficient benefit to justify treatment risk." [6] MAESTRO-NASH enrolled patients with F1B-F3, meaning the key trial population slightly exceeds the AASLD minimum threshold, and the F1B cohort represents a pre-guideline expansion of the treated population.

The European Association for the Study of the Liver (EASL) has not yet issued a formal position statement incorporating resmetirom post-approval, though the 2024 EASL clinical practice guidelines note that liver biopsy remains the gold standard for staging in clinical trials and high-stakes treatment decisions. [7]

How MAESTRO-NASH Data Compare to Other MASH Trials

Obeticholic acid (OCA), a farnesoid X receptor agonist, reached a fibrosis improvement rate of 23% vs. 12% placebo in the REGENERATE trial at 18 months, but failed its co-primary endpoint of NASH resolution (11.7% vs. 8.0%, P=0.18). [8] OCA did not receive FDA approval for MASH. Resmetirom's dual co-primary success distinguishes its evidence package.

Lanifibranor, a pan-PPAR agonist, showed 49% NASH resolution vs. 22% placebo in the NATIVE trial (N=247) at 24 weeks. [9] That trial was smaller and shorter, and the drug has not yet completed phase 3 in the U.S., so the evidence quality under GRADE remains lower than resmetirom's at present.

Semaglutide 2.4 mg showed NASH resolution in 59% vs. 17% placebo in a phase 2 trial (N=320) at 72 weeks. [10] The ESSENCE phase 3 MASH trial is ongoing. Until ESSENCE reports, semaglutide's MASH evidence sits at GRADE low (single phase 2 RCT, no biopsy-based fibrosis endpoint met).

This comparative snapshot positions resmetirom as the only agent with GRADE moderate evidence for dual histological endpoints in a phase 3 MASH RCT as of early 2025.

Patient Selection: Translating GRADE Evidence to Practice

Resmetirom 80 mg or 100 mg once daily is indicated for adults with MASH and liver fibrosis stages F1B, F2, or F3. Stage F4 (cirrhosis) is excluded because MAESTRO-NASH excluded patients with decompensated cirrhosis, and safety in compensated F4 has not been formally established.

Body Weight and Dosing

The FDA label recommends starting at 80 mg in patients with moderate hepatic impairment (Child-Pugh B is a contraindication). Weight alone does not drive dose selection; the 80 mg vs. 100 mg difference was studied as parallel arms rather than a titration protocol, and both doses met their co-primary endpoints. [3]

Concomitant Diabetes and Obesity

Approximately 56% of MAESTRO-NASH participants had type 2 diabetes, and mean BMI was 35.6 kg/m². [1] This baseline profile is broadly representative of the MASH population seen in hepatology and endocrinology clinics. Patients already on GLP-1 receptor agonists were not excluded from MAESTRO-NASH, though combination pharmacology data with semaglutide or tirzepatide remain limited to post-hoc subgroups.

Monitoring Requirements Per FDA Label

Baseline liver enzymes, thyroid function, and lipid panel are recommended before initiating resmetirom. [3] If ALT or AST rises to >5x ULN and is confirmed on repeat testing, the FDA label advises discontinuation. Pregnancy must be excluded and reliable contraception confirmed before prescribing, given the Category X equivalent labeling status.

Ongoing Evidence Gaps and What Will Change the GRADE Rating

Four specific evidence gaps, if resolved by future trials, could shift resmetirom's GRADE ratings upward.

First, the MAESTRO-NASH OUTCOMES trial, targeting cirrhosis-free survival as its primary endpoint, could move liver-related hard outcomes from GRADE very low to GRADE moderate or high if results are positive and the trial is adequately powered. [4]

Second, a dedicated cardiovascular outcomes trial using the TR-β mechanism would resolve the current GRADE very low rating for CV endpoints. No such trial has been registered publicly as of January 2025.

Third, comparative effectiveness data against semaglutide and lanifibranor in head-to-head RCTs do not yet exist. Without them, indirect comparisons carry GRADE low quality at best.

Fourth, long-term safety data beyond 52 weeks in a broad real-world population, particularly for hepatotoxicity and bone mineral density in post-menopausal women, could either confirm or downgrade the current safety profile. A post-marketing requirement for a 5-year safety registry was included in the FDA approval. [3]

Clinicians should access the full MAESTRO-NASH publication directly at PubMed PMID 38324483 and review the FDA prescribing information before initiating therapy. The approved prescribing information specifies liver enzyme monitoring at 12 weeks after starting resmetirom as the first post-baseline safety check.