Case Study: Sanofi — Rare Disease Benefit-Risk Analysis at Scale

Company

Sanofi (Top 10 Global Pharmaceutical)

Therapeutic Areas

Immunology (Dupixent/dupilumab) & Rare Disease (Cerdelga/eliglustat)

Focus

Unified BRA for complex multi-indication products

Challenge

Rare disease data scarcity and multi-indication BRA complexity

ArcaScience Modules Used

Data Intelligence, Decision Intelligence, Automated Outputs

Timeline

16-week implementation

The Challenge

Data Scarcity and Multi-Indication Complexity in Rare Disease

Sanofi's rare disease and specialty portfolio presented unique BRA challenges. Cerdelga (Gaucher disease type 1) had a limited patient population making traditional statistical approaches unreliable. Dupixent's rapid expansion across 6+ indications (atopic dermatitis, asthma, CRSwNP, prurigo nodularis, EoE, COPD) created a complex cross-indication safety profile that was difficult to characterize holistically. Manual BRA processes could not keep pace with the expanding indication portfolio. Real-world evidence was scattered across patient registries, natural history studies, and rare disease databases with inconsistent formats.

The Solution

ArcaScience Platform Deployment

ArcaScience deployed a specialized rare disease BRA framework combining small-population statistical methods with AI-powered real-world evidence integration, enabling Sanofi to create rigorous benefit-risk assessments despite data limitations.

Implementation Process

Weeks 1–3

Data Landscape Assessment

Mapped all available data sources across Dupixent and Cerdelga programs. Integrated 34M+ real-world records from specialty pharmacy, registry, and claims data. Established rare disease-specific data quality frameworks.

Weeks 4–7

Rare Disease Analytics Build

Deployed Bayesian methods optimized for small populations (Cerdelga). Built cross-indication safety analysis for Dupixent's 6+ indications. Created unified signal detection across rare and common disease populations.

Weeks 8–11

BRA Framework & Modeling

MCDA frameworks tailored for rare disease regulatory expectations. Patient preference integration using adaptive conjoint analysis. Sensitivity analyses accounting for data sparsity and uncertainty.

Weeks 12–14

Regulatory Document Generation

Automated PSUR/PBRER generation for both products. Created indication-specific and unified BRA summaries. Prepared RMP updates incorporating new signal findings.

Weeks 15–16

Validation & Knowledge Transfer

Cross-validated results with Sanofi's medical experts. Trained rare disease and immunology PV teams. Established ongoing monitoring dashboards.

Results

52%

Faster regulatory
submissions

85%

Signal detection
improvement

34M+

Real-world records
integrated

Additional results: 6 products unified on single platform

Quantitative Outcomes

Metric	Before ArcaScience	After ArcaScience
Regulatory submission prep	18 weeks average	8.6 weeks (52% faster)
Signal detection sensitivity	Baseline	85% improvement
RW data sources integrated	4 (manually curated)	12 sources, 34M+ records
Cross-indication analysis	Annual manual review	Continuous automated
Products on unified platform	0	6 (Dupixent x5 indications + Cerdelga)
BRA update frequency	Semi-annual	Continuous (monthly reports)

"Rare disease BRA has always been our biggest challenge—limited data, unique patient populations, and high regulatory scrutiny. ArcaScience gave us the analytical framework to produce rigorous benefit-risk assessments even with small populations, and the real-world evidence integration added a dimension we'd been missing entirely."

— Global Head of Rare Disease Safety, Sanofi

"Managing Dupixent across six indications was becoming unmanageable with manual processes. ArcaScience's cross-indication analysis not only saved us enormous time but identified safety patterns that only become visible when you look across the entire patient exposure. The 85% improvement in signal detection speaks for itself."

— VP, Global Pharmacovigilance, Sanofi

Key Takeaways

Rare disease BRA requires specialized statistical approaches—Bayesian methods and AI-powered evidence synthesis can overcome traditional sample size limitations.
Cross-indication safety analysis reveals patterns invisible to indication-specific reviews, critical for multi-indication products like Dupixent.
Real-world evidence integration is especially valuable in rare disease, where clinical trial data alone cannot characterize the full safety profile.
52% faster regulatory submissions demonstrate that rigor and speed are not mutually exclusive when supported by the right platform.
Unified BRA platforms scale efficiently—adding new indications to an existing framework is dramatically faster than building from scratch.

Ready to Transform Your Benefit-Risk Strategy?

Learn how ArcaScience can support your next regulatory submission with AI-driven benefit-risk analysis.

Schedule a Consultation | info@arcascience.ai

Sanofi: Rare Disease Benefit-Risk Analysis at Scale