Vigila · Stress-test
Find the rare adverse event before a patient does
Vigila oversamples the subgroups where safety signals hide, by genotype, comorbidity, age or ancestry, and maps where predicted risk rises with exposure.
SZ-0412 · hepatic adverse events
- Patients simulated
- 250,000
- Overall grade 3+ rate
- 0.4% (0.3–0.5%)
- At-risk cluster
- CYP2C9 poor metaboliser + CKD
- Cluster rate
- 3.1% (2.2–4.2%)
- Exposure threshold
- AUC above 2,100 ng·h/mL
Capabilities
What Vigila does
Rare subgroups
Oversampled at scale
Simulate hundreds of thousands of patients so a one-in-a-thousand event has enough cases to study.
Clusters
At-risk combinations
Find genotype, comorbidity and exposure combinations where predicted risk concentrates.
Thresholds
Exposure and dose limits
Map predicted risk against exposure to support dose caps and monitoring.
Constraints
Safety into design
Send safety limits to Protova so unsafe designs are ruled out early.
Monitoring
Continuous pharmacovigilance
Re-run as real-world outcomes arrive. Planned for marketed products.
Evidence
Signals on the record
Each signal and its later outcome is logged in Evidra.
How it works
Inside Vigila
Amplify
Oversample rare subgroups in Cohorta.
Simulate
Run exposure and safety models in Insilon.
Cluster
Find where predicted risk concentrates.
Report
Write a signal map for safety review.
Specification
Specification
Items marked planned are in design with pilot partners.
| Item | Detail |
|---|---|
| Scale (target) | 105 to 107 patient scenarios per study |
| Outputs | Adverse-event distributions, at-risk clusters, exposure thresholds |
| Statistics | Calibrated tail-risk models with intervals |
| Label support | Warning and monitoring suggestions (planned) |
| Causal attribution | Mechanism attribution of signals (planned) |
| Status | In development |
Place in the loop
Vigila in the Sanenza loop
- 01PopulateCohorta
Generates calibrated virtual-patient cohorts for an indication, region or rare subgroup.
- 02SimulateInsilon
Runs the compound across the whole cohort with PK-PD surrogates and biomolecular response.
- 03DesignProtova
Searches dose, cohort, endpoint and sample size, and scores every candidate protocol.
- 04Stress-testVigila
Oversamples rare subgroups to find adverse-event signals before a patient is exposed.
- 05LocalisePhysova
Renders organ-scale twins for tissue exposure and drug-device questions.
- 06ValidateEvidra
Links each prediction to the real outcome and sends recalibration back to every product.
Questions
About Vigila
Is a predicted signal a real signal?
No. It is a hypothesis to investigate. Vigila helps decide where to look and what to monitor; clinical data decides.
Which adverse events can it model?
Events with a mechanistic or exposure link in the models, starting with hepatic, renal and cardiac events in selected areas.
Does it replace pharmacovigilance systems?
No. It adds simulation to your existing safety process and systems.
See Vigila in a briefing
We walk your team through Vigila on a public example, then discuss where a pilot could test it against outcomes you already know.