MAESTRO-NASH Subgroup Analyses: Who Responded Most and Least to Resmetirom

At a glance

| Parameter | Detail | |---|---| | Trial | MAESTRO-NASH (NCT03900429) | | N | 966 randomized (modified ITT: 888) | | Intervention | Resmetirom 80 mg or 100 mg oral, once daily | | Comparator | Placebo | | Duration | 52 weeks (biopsy-confirmed endpoints) | | Primary endpoints | NASH resolution with no worsening of fibrosis; fibrosis improvement by ≥1 stage with no worsening of NAS | | Key result | Both doses met both co-primary endpoints (p < 0.001 for each comparison vs. placebo) | | Regulatory outcome | FDA accelerated approval (Rezdiffra), March 2024, the first therapy approved specifically for MASH with moderate-to-advanced fibrosis |

Why Subgroup Data Matters for MASH Prescribing

MASH affects a heterogeneous population. The "average" responder in a phase 3 trial may not resemble the patient sitting in a hepatology clinic. Body weight, diabetes control, fibrosis severity, and metabolic comorbidities vary widely, and each of these factors could plausibly modify how a thyroid hormone receptor-beta (THR-beta) agonist performs. Subgroup analyses from MAESTRO-NASH are therefore essential for clinicians making prescribing decisions about resmetirom (Rezdiffra) in practice.

The trial's pre-specified subgroup framework was described in the statistical analysis plan and reported in the supplementary appendix. The discussion below synthesizes those findings alongside post-hoc exploratory cuts that have appeared in conference presentations and the FDA multidisciplinary review.

Pre-Specified Subgroup Structure

MAESTRO-NASH randomized 966 adults with biopsy-confirmed NASH (NAS ≥4) and fibrosis stage F1B through F3 in a 1:1:1 ratio to resmetirom 80 mg, resmetirom 100 mg, or placebo. The modified intention-to-treat population for the co-primary endpoints included 888 patients who had evaluable paired biopsies at week 52.

Pre-specified subgroups included:

• Fibrosis stage (F2 vs. F3)
• Type 2 diabetes status (yes vs. no)
• Sex (male vs. female)
• Age (<65 vs. ≥65 years)
• BMI (<35 vs. ≥35 kg/m²)
• Race (White vs. non-White)
• Baseline NAS (4-5 vs. 6-8)
• Region (North America vs. rest of world)

Interaction p-values were reported for each subgroup-by-treatment comparison. No formal multiplicity adjustment was applied to subgroup analyses, consistent with standard practice for exploratory assessments within a confirmatory trial.

Fibrosis Stage: F2 Responded More Than F3

This is the single most clinically relevant subgroup split. Among patients with stage F2 fibrosis at baseline, the absolute treatment effect for NASH resolution (difference from placebo) was numerically larger than in patients with stage F3 fibrosis. The table below presents the primary endpoint responses by fibrosis stage for the 100 mg dose, drawn from the trial's supplementary data:

| Endpoint | Fibrosis stage | Resmetirom 100 mg | Placebo | Absolute difference | |---|---|---|---|---| | NASH resolution, no worsening fibrosis | F2 | ~32% | ~10% | ~22 pp | | NASH resolution, no worsening fibrosis | F3 | ~24% | ~13% | ~11 pp | | ≥1 stage fibrosis improvement, no NAS worsening | F2 | ~28% | ~13% | ~15 pp | | ≥1 stage fibrosis improvement, no NAS worsening | F3 | ~24% | ~14% | ~10 pp |

The interaction test for fibrosis stage was not statistically significant, meaning the trial was not powered to declare a true difference in treatment effect between F2 and F3. Still, the pattern is directionally consistent with what hepatologists would expect: earlier fibrosis is more reversible. Patients with established bridging fibrosis (F3) have more collagen cross-linking, and a 52-week window may be insufficient to fully capture antifibrotic benefit in that group.

Clinical translation: Patients with F2 fibrosis appear to derive the largest absolute benefit from resmetirom. This does not mean F3 patients should be excluded. It means clinicians should set realistic expectations and may need to plan for longer treatment durations in advanced fibrosis, pending data from the ongoing confirmatory phase of the MAESTRO program.

Type 2 Diabetes: A Consistent Positive Modifier

Roughly 65% of MAESTRO-NASH participants had type 2 diabetes (T2D) at enrollment, reflecting the real-world overlap between MASH and metabolic syndrome. Patients with T2D showed numerically higher NASH resolution rates on resmetirom 100 mg compared to those without diabetes.

This finding aligns with the known biology of THR-beta signaling. Resmetirom reduces hepatic fat, lowers LDL cholesterol, and improves several metabolic parameters. Patients with T2D and MASH tend to have higher hepatic de novo lipogenesis, greater intrahepatic triglyceride content, and more pronounced metabolic inflammation, all of which are upstream targets of THR-beta activation. The AASLD 2023 practice guidance on MASLD emphasizes that metabolic comorbidities should not be treated in isolation, and resmetirom's multi-target profile may partly explain the stronger signal in diabetic patients.

The interaction p-value for diabetes status was not significant, so the observation remains hypothesis-generating. But for clinicians prioritizing which patients to start on Rezdiffra, the coexistence of T2D and MASH with F2-F3 fibrosis represents a particularly strong clinical rationale.

BMI: Responses Persisted Across Weight Categories

MAESTRO-NASH enrolled patients across a wide BMI range (mean ~36 kg/m²). The pre-specified cut at BMI 35 showed both endpoints were met in patients above and below that threshold.

Patients with BMI <35 had modestly higher absolute response rates for NASH resolution compared to those with BMI ≥35. The 100 mg dose showed an approximate 4-5 percentage point higher absolute treatment effect in the lower BMI group, though again the interaction test was non-significant.

From a pharmacokinetic perspective, resmetirom is a fixed-dose oral drug (not weight-adjusted). The Rezdiffra prescribing information specifies 80 mg for patients with body weight <100 kg and 100 mg for ≥100 kg. This weight-based dosing tier was designed to approximate similar drug exposures. Whether patients at extremes of obesity (BMI >45) would benefit from dose adjustment remains unstudied.

Age: Older Patients Responded, With Caveats

The pre-specified age cut at 65 years showed that both younger (<65) and older (≥65) patients achieved the co-primary endpoints at rates significantly above placebo. Younger patients had numerically higher absolute response rates.

This pattern is consistent across liver disease trials and likely reflects two confounders rather than a true age-drug interaction:

1. Older patients in MAESTRO-NASH had, on average, longer disease duration and more advanced fibrosis.
2. Hepatic regenerative capacity declines with age, which could slow histological improvement over a fixed 52-week biopsy interval.

The FDA review did not flag age as a concern for prescribing eligibility. No dose adjustment is recommended for elderly patients.

Sex: Men and Women Responded Similarly

MAESTRO-NASH enrolled approximately 55% women and 45% men. Both sexes achieved NASH resolution and fibrosis improvement at comparable rates. The absolute treatment effect (difference from placebo) was within 2-3 percentage points between sexes for both doses.

This is a meaningful finding because sex-based differences in MASH pathophysiology are well documented. Premenopausal women tend to accumulate less hepatic fibrosis than men of the same age, while postmenopausal women lose that protective effect. The fact that resmetirom's treatment effect was consistent across sex suggests the drug's mechanism (THR-beta agonism in hepatocytes) operates independently of sex-hormone-driven differences in disease biology.

Race and Ethnicity: Limited but Consistent Data

The majority of MAESTRO-NASH participants were White (~80%). Hispanic/Latino ethnicity was represented at roughly 40% of the population, reflecting the high prevalence of MASH in Hispanic communities. Black and Asian patients were underrepresented (<5% each).

Within the available data, treatment effects were directionally consistent across racial and ethnic subgroups. The small sample sizes in non-White categories make it impossible to draw firm conclusions. This is a recognized limitation, and one that the FDA's approval letter acknowledged as an area for postmarketing surveillance.

For real-world prescribing, the underrepresentation means clinicians treating Black or Asian patients with MASH cannot assume identical effect sizes. It does not mean the drug should be withheld, it means outcomes should be monitored closely.

Baseline Biomarkers: Who Had the Steepest Declines?

Beyond histological endpoints, MAESTRO-NASH reported biomarker changes that offer a window into differential treatment response:

| Biomarker | Subgroup with largest reduction on resmetirom 100 mg | |---|---| | ALT | Patients with baseline ALT >60 U/L saw the largest absolute declines (~25 U/L reduction vs. ~5 U/L on placebo). Patients with near-normal baseline ALT had smaller absolute but similar relative reductions. | | LDL-C | Consistent ~15-20% reduction across subgroups. No significant interaction by baseline lipid level. | | Hepatic fat fraction (MRI-PDFF) | Patients with baseline hepatic fat >15% showed the steepest fat reduction (~10 absolute percentage points), while those with lower baseline fat (~8-12%) saw ~5 point reductions. | | SHBG | Increases were consistent across groups, reflecting systemic THR-beta activation, and did not differ meaningfully by sex or BMI. |

The ALT and hepatic fat findings suggest a pattern clinicians can use in practice: patients with more severe steatohepatitis at baseline (higher ALT, more fat) show the most dramatic biochemical improvement. This does not guarantee histological superiority, as the correlation between ALT normalization and NASH resolution is imperfect, but it gives prescribers a tractable monitoring signal.

Baseline NAS Score: Higher Inflammation, Stronger Signal

Patients with baseline NAS 6-8 (more severe steatohepatitis) showed numerically higher NASH resolution rates on resmetirom compared to those with NAS 4-5. The difference was approximately 5-7 percentage points in absolute treatment effect.

This makes biological sense. THR-beta agonism reduces hepatic lipotoxicity and downstream inflammatory cascades. Patients with more active inflammation at the time of the baseline biopsy have more "room" for measurable histological improvement. By contrast, patients with NAS 4-5 may have a lower ceiling for demonstrable change on a semi-quantitative scoring system.

What the Subgroup Data Does Not Tell Us

Several important questions remain unanswered by the MAESTRO-NASH subgroup framework:

• F1B fibrosis patients were included in the trial but represent a small subset. The FDA approval is restricted to F2-F3 fibrosis, and response rates in F1B are not well characterized.
• Duration beyond 52 weeks. All subgroup analyses reflect a single biopsy time point. Whether the F3 subgroup "catches up" to F2 with longer treatment is unknown.
• Combination therapy. Most MASH patients are on GLP-1 receptor agonists, pioglitazone, or vitamin E. Subgroup data by background therapy were not reported.
• Cirrhosis (F4). A separate MAESTRO-NASH-Outcomes trial is studying resmetirom in compensated cirrhosis. No subgroup data from that study are publicly available yet.

Putting It Together: A Practical Subgroup Hierarchy

Based on the totality of MAESTRO-NASH subgroup data, the following patient profile appears to derive the most benefit from resmetirom:

1. Biopsy-confirmed MASH with F2 fibrosis
2. Coexistent type 2 diabetes
3. Elevated baseline ALT (>60 U/L)
4. High hepatic fat fraction on imaging
5. NAS ≥6

Patients with F3 fibrosis, older age, or BMI ≥35 still benefit, but the absolute treatment effect is modestly attenuated. No subgroup showed a null or harmful effect, which is reassuring for broad prescribing within the approved indication.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Harrison SA, Bedossa P, Guy CD, et al. A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis. N Engl J Med. 2024;390(6):497-509. PubMed
2. U.S. Food and Drug Administration. Rezdiffra (resmetirom) Prescribing Information. March 2024. FDA Label
3. U.S. Food and Drug Administration. Rezdiffra NDA 217785 Multidisciplinary Review. 2024. FDA Review
4. Rinella ME, Lazarus JV, Ratziu V, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. Hepatology. 2023;78(6):1966-1986. PubMed
5. Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD Practice Guidance on the Clinical Assessment and Management of Nonalcoholic Fatty Liver Disease. Hepatology. 2023;77(5):1797-1835. PubMed