Drugs lost in translation in CKD: from redundant designs to learning systems

Area of Science:

• Nephrology
• Translational Medicine
• Drug Development

Background:

• Chronic kidney disease (CKD) is a major cause of death, with many drugs failing in late-stage development.
• Current preclinical models and clinical trial designs inadequately represent human CKD complexity, leading to translational failures.

Purpose of the Study:

• To analyze the reasons for high failure rates in CKD drug development.
• To propose a re-engineered translational ecosystem for more effective CKD therapies.

Main Methods:

• Review of current preclinical models and early/late-phase clinical trial strategies in CKD.
• Analysis of factors contributing to drug development failures, including endpoint selection and population heterogeneity.
• Examination of the ocedurenone trial as a case study.

Main Results:

• Preclinical models fail to capture the chronic, multifactorial nature of human CKD and its complications.
• Early trials often use surrogate endpoints in heterogeneous populations, masking potential benefits.
• Late-phase trials face termination due to neutral outcomes or safety concerns, with limited learning from negative data.

Conclusions:

• CKD drug development needs a paradigm shift towards learning health systems.
• Key improvements include human-relevant models, biology-driven patient enrichment, patient-centered endpoints, and adaptive trial platforms.
• Systematic reporting of trial failures is crucial to prevent repeated errors and advance CKD therapeutics.