PHARMACEUTICALS

From safety signal to submission: regulatory variation management for a global pharma company

A single safety-committee recommendation can ripple into label changes, variation filings across markets, country affiliates, packs and submission evidence. Wexa modelled the whole regulatory chain in a context graph and built an agent that computes the blast radius and routes every change to the right signer.

Download this case as PDF Book a demo

impacted objects found across 4 hops
packs in the supply chain reached by the change
submission artefacts ready; the blocker identified
human signature gates / autonomous executions

CLIENT: Global pharmaceutical company (regulatory and medical affairs)
REGION: EU / US
WHEN: 2026
ENGAGEMENT: Solution build
WEXA COMPONENTS: Wexa · regulatory ontology · context graph · routing agent · human signature gates

01Case 14

The challenge

Safety signals, labels, marketing authorisations and supply data sit in separate systems
Impact assessment is manual and slow
Every regulatory action needs a named human signature

What Wexa built

Ontology: product, marketing authorisation, country, health authority, core data sheet, label section, safety signal, case, change event, submission, label change
Blast-radius analysis from a safety signal to labels, filings, affiliates and packs in the supply chain
Evidence readiness check across the submission dossier
Routing agent proposing the variation route per market; it rejected a fast-track route when the wording went beyond the committee's core text

02Case 14

How it runs on Wexa

FigureHow this use case runs on Wexa: client systems connected, ontology discovered, context graph in the Wexa memory layer, agents and process flows, and governance. Highlighted components are central to this use case.

03Case 14

Results

Measure	Before / without Wexa	With Wexa
Impact-assessment hours per variation (150 variations a year)	120 hours	24 hours
Annual cost of the process	$2.2M	$432K
Time to implement the AI solution	6 months	6 weeks
AI build cost (one-time)	$480K	$44K
AI operating cost per year (models, infrastructure, upkeep, platform)	$312K	$166K
Total AI cost in year 1	$792K	$210K

With Wexa

net annual value to the client after Wexa's own operating cost

faster to implement

lower AI operating cost

less AI spend in year 1

Six change proposals, each routed to a named signer; nothing executed without a signature
Packs still in packaging within the change window identified for relabelling