Rescuing a Failed Trial: PhaseV Causal-ML Identifies Key Responders to Investigational Oral Insulin Treatment
Key Takeaway
PhaseV’s Causal-ML approach uncovered responsive patient subgroups, enabling a more targeted and informed clinical trial strategy
Challenge
Oramed faced a significant hurdle when its Phase III trial for an oral insulin candidate in Type II diabetes failed to meet the primary endpoint: a reduction in baseline A1C levels compared to placebo after 26 weeks of treatment. Without a clear understanding of which patients might benefit most, the path forward for the candidate was uncertain, putting the program's viability at risk
Solutions
- PhaseV applied its Causal-ML HTE Pipeline to analyze Oramed’s Phase III trial data
- PhaseV enriched the analysis with data from Oramed’s Phase II study and validated the results using multiple external Phase III studies
- Rigorous sensitivity testing and false discovery analysis were conducted to ensure robustness and reliability
Proven Results
The analysis identified patient subgroups—characterized by body mass index (BMI), baseline hemoglobin A1C (HbA1c), and age—that responded significantly to oral insulin
Real-World Impact
Based on PhaseV’s analysis, Oramed re-initiated a Phase III trial of its oral insulin candidate in the U.S. under an amended protocol