Article 2

Challenges in Designing Clinical Trials for Liver Disease Drugs: The Rezdiffra Experience

Liver diseases, including non-alcoholic steatohepatitis (NASH), hepatitis, cirrhosis, and hepatocellular carcinoma (HCC), present a formidable global health challenge. Developing effective drugs requires well-designed clinical trials, yet researchers encounter significant hurdles stemming from disease complexity, patient variability, diagnostic limitations, endpoint dilemmas, regulatory demands, and logistical constraints. The journey of Rezdiffra (resmetirom), approved by the FDA in March 2024 as the first NASH-specific therapy, exemplifies these challenges. Its pivotal Phase 3 MAESTRO-NASH trial, conducted by Madrigal Pharmaceuticals, highlights the intricate balance researchers must strike to bring liver disease treatments to market.

Disease Complexity and Trial Design

Liver diseases involve intertwined pathological processes—steatosis, inflammation, fibrosis, and malignancy—varying by condition and stage. The MAESTRO-NASH trial tackled NASH’s multifactorial nature, targeting both fat reduction and fibrosis improvement with resmetirom, a thyroid hormone receptor-beta agonist. Designing a trial to capture these dual effects was complex; researchers opted for co-primary histological endpoints—NASH resolution (no ballooning, minimal inflammation) and fibrosis improvement (≥1-stage reduction)—measured via biopsy at baseline and week 52. This approach ensured comprehensive efficacy assessment but increased trial intricacy, requiring large cohorts (over 900 patients) and multiple arms (80 mg, 100 mg, placebo), inflating costs and duration.

Patient Heterogeneity and Recruitment Struggles

Patient variability complicates trial enrollment, as seen in MAESTRO-NASH. NASH patients range from early-stage asymptomatic individuals to those with advanced fibrosis (F2-F3), influenced by genetics (e.g., PNPLA3), obesity, and diabetes. The trial required biopsy-confirmed NASH with fibrosis (F1B-F3) and a NAFLD Activity Score ≥4, narrowing the eligible pool to those with histological evidence, a process slowed by NASH’s often-silent progression. Recruiting such patients across multiple sites took years, reflecting the broader challenge of finding representative yet specific cohorts. Researchers mitigated this by allowing some flexibility (e.g., F1B inclusion), but slow enrollment remains a persistent liver disease trial hurdle.

Diagnostic Challenges: Biopsy vs. NITs

Accurate diagnosis and monitoring are critical, yet liver disease trials grapple with diagnostic tools. MAESTRO-NASH exemplifies this tension: biopsies were integral for enrollment and primary endpoints, confirming NASH and fibrosis stages at baseline and assessing changes at week 52 (e.g., fibrosis improvement in 25.9% of 100 mg patients vs. 14.2% placebo). However, biopsies are invasive, costly, and variable—sampling errors can skew results by up to 30%. Recognizing this, the trial also employed an array of NITs—MRI-PDFF, FibroScan VCTE, ELF test, and liver enzymes—as secondary measures. These showed robust responses (e.g., >70% of 100 mg patients had ≥30% fat reduction via MRI-PDFF), correlating with biopsy outcomes and supporting real-world applicability, as the FDA label for Rezdiffra omits a biopsy requirement.

This hybrid approach highlights a key challenge: balancing biopsy’s precision with NITs’ practicality. While biopsies met regulatory demands for histological proof, NITs addressed the need for scalable, less invasive monitoring, though their lack of full validation as standalone endpoints forced reliance on paired biopsies, doubling the diagnostic burden.

Endpoint Selection and Disease Progression

Choosing endpoints is a perennial issue, amplified in MAESTRO-NASH. NASH’s slow progression—years from F1 to F3—meant short-term trials risked missing meaningful changes. Researchers selected 52-week histological endpoints, feasible yet debated as surrogates for hard outcomes like cirrhosis prevention. The trial’s success (e.g., NASH resolution in 29.9% of 100 mg patients vs. 9.7% placebo) satisfied the FDA’s accelerated approval pathway, but confirmatory trials for long-term benefits (e.g., survival) are ongoing, illustrating the trade-off between speed and clinical relevance. High placebo responses (14.2% fibrosis improvement) further complicated efficacy signals, necessitating larger samples and statistical finesse—challenges mirrored across liver disease trials.

Placebo and Control Arm Ethics

Control arm design poses ethical and practical dilemmas, evident in MAESTRO-NASH. With no prior NASH drugs, a placebo arm was justifiable, but lifestyle improvements in placebo patients (e.g., diet changes) mimicked resmetirom’s effects, inflating responses and reducing power. In diseases with existing treatments (e.g., hepatitis antivirals), active comparators are ethically preferable but costlier, a tension Rezdiffra avoided yet highlights for future trials. Blinding resmetirom’s metabolic effects (e.g., fat loss) was also tricky, risking unmasking—a common liver trial pitfall.

Regulatory and Safety Demands

Regulatory scrutiny shapes trial design, as MAESTRO-NASH demonstrates. The FDA required histological endpoints for approval, met by biopsies showing fibrosis and NASH improvements, but mandated post-approval studies to confirm clinical benefits, doubling research efforts. Safety monitoring was rigorous—resmetirom’s liver-metabolizing target raised hepatotoxicity concerns, though well-tolerated (e.g., mild diarrhea as a common side effect). Long-term safety data, critical for chronic NASH, extended timelines, a challenge compounded in multinational trials where EMA and FDA endpoint preferences (fibrosis vs. NASH resolution) diverge, necessitating flexible protocols.

Logistical and Economic Barriers

MAESTRO-NASH’s scale—multicenter, biopsy-driven, with NIT integration—underscored logistical woes. Site variability in biopsy reading and NIT equipment (e.g., FibroScan availability) risked inconsistency, mitigated by centralized pathology and AI tools (e.g., qFibrosis). Costs soared with imaging, patient travel, and retention efforts, a burden small biotechs face acutely, though Madrigal’s backing enabled completion. Economic constraints often stall liver trials in late stages, unlike oncology’s richer funding, delaying therapies despite need.

Lessons from Rezdiffra and Broader Implications

The Rezdiffra experience encapsulates liver trial challenges: complex pathology demanded dual endpoints, patient variability slowed recruitment, diagnostics balanced biopsy’s rigor with NITs’ promise, and regulatory/safety needs extended timelines. Yet, it offers solutions—NITs like MRI-PDFF validated against biopsies, adaptive designs could refine future studies, and collaboration (e.g., with AI firms) improved consistency. Other trials, like those for FGF21 analogs (89bio’s pegozafermin, Akero’s efruxifermin), face similar issues—heterogeneity, endpoint debates—but build on Rezdiffra’s path, integrating NITs earlier.

Conclusion

Designing clinical trials for liver disease drugs, as exemplified by MAESTRO-NASH, is a high-stakes endeavor fraught with scientific, ethical, and practical challenges. Rezdiffra’s success—approved for NASH without biopsy mandates—shows progress, yet underscores persistent gaps: NIT validation, endpoint relevance, and resource demands. As of March 2025, overcoming these requires innovation—AI, biomarkers, partnerships—to transform liver disease treatment, building on lessons from resmetirom’s journey.