Expectedness assessment against the right edition of the RSI, with the clause quoted and the causality evidence laid out
Every serious event judged against the label in force that day, with the clause quoted.
See one case, screen by screen ↓A serious case lands with its clock already running from Day 0. Before the safety physician can judge it, someone has to work out which reference applies: the investigator’s brochure edition in force on the onset date for a trial, the USPI for a US 15-day Alert decision, the SmPC for the EU — with the CCDS beside them, and a new edition approved but not yet in effect.
Then the term has to be found, and it is rarely there word for word. “Alanine aminotransferase increased” in the RSI does not make drug-induced liver injury expected; a listed reaction that was life-threatening this time may not be. The causality picture — dosing dates, dechallenge, other medicines, viral serology — sits across the narrative and the lab tables. Done by hand, every case is a search, and every search is a chance to judge the case on the wrong edition.
Estimated split for one serious case on a 15-day clock, by hand and with the solution.
Six specialist agents pick the right reference, search it, read the clause and lay out the causality evidence; a safety physician decides and signs.
A 52-year-old man in study VLT-302 is hospitalised with drug-induced liver injury on Veltrimab. The SAE form arrives on Oct 3. Here is what happens next, screen by screen, in the working solution.
Maya opens the solution to 8 serious cases open, 5 on an expedited clock and 7 waiting for her. The clock runway sets each case on its due date — clinical trials on the SUSAR clock, marketed products on the 15-day Alert and EU 15-day reports. The liver injury on Veltrimab, MR-26-1184, sits at Oct 18 with agents ready.
References in force today: “Veltrimab · IB Edition 10 takes effect Nov 1. Events with onset before Nov 1 stay on Edition 9.”
MR-26-1184: subject 114-0087 at Site 114, Chicago, Day 0 Oct 3, investigator’s causality “related”. The SAE report sits on the right with its passages highlighted. Five agents run in view: the reference picker, the MedDRA neighbour search, the listedness reader, the causality evidence agent and the reportability clock.
Reference picker: “Onset Sep 22 → IB Edition 9 in force · IB Edition 10 starts Nov 1 — not used”
The proposed verdict for drug-induced liver injury is Unexpected, high confidence. Onset was 22 Sep 2026, so Edition 9 applies. Edition 9 §7.4 lists alanine aminotransferase increased and transaminases increased — raised enzymes, not liver injury. The raised ALT reported in the same case is proposed as listed.
“Edition 9 lists only raised liver enzymes; liver injury is greater in severity, which makes it unexpected. Edition 10 lists it but is not in force on the onset date.”
The edition timeline shows the onset on Sep 22 inside Edition 9’s window (Mar 1 – Oct 31). Edition 10 was approved on Sep 30 but applies only from Nov 1. The search covers seven MedDRA terms under Hepatocellular damage and hepatitis NEC: two are found, both at lesser severity; five, including hepatotoxicity and hepatitis, are not found.
“The version in force at the moment of occurrence decides.”
Two doses of Veltrimab 150 mg on Aug 12 and Sep 9, acetaminophen from Sep 15 to 20, onset on Sep 22, in hospital from Sep 23 to 29, the drug stopped on Sep 24 — and ALT plotted against its upper limit of 41, from 24 at baseline to 412 and back down to 96 by Oct 2.
Four facts support a causal role: onset 41 days after the first dose, a positive dechallenge, viral hepatitis A to E negative with a normal ultrasound, and raised enzymes already in Edition 9. One is against: acetaminophen 2 g/day for 6 days, ending 2 days before onset. Hy’s Law is not met. The agents suggest reasonable possibility, RUCAM 7 · probable; Maya records Yes, and the comment for the case is drafted from the evidence.
The causality agent never records causality for the physician, and the investigator’s “related” is never downgraded for expedited reporting.
“Ask investigator” opens three numbered questions to the principal investigator at Site 114: liver tests in the 6 months before the first dose, any herbal or over-the-counter products besides acetaminophen, and the next ALT, AST and bilirubin results. The reply is requested by Oct 10, inside the clock.
Nothing is sent until the physician approves it.
The summary lists each event as proposed, the causality and the reportability: IND safety report, due Oct 18. Maya signs as Dr. Maya Chen · Safety physician with her password, and the meaning is recorded with the signature.
“Meaning: I have reviewed the listedness and causality of this case and approve this medical assessment.”
The reportability is now final, from the signed assessment: an IND safety report under 21 CFR 312.32 to FDA and a SUSAR report to EudraVigilance, both on the 15-day clock and due Oct 18, a SUSAR notice to investigators and ethics committees with the report — all three queued — and a DSUR line listing for the annual safety report.
“Seriousness: Hospitalisation · Listedness: Unexpected · Causality: reasonable possibility”
412 serious cases reviewed in the last 30 days, 37 of 37 expedited reports on time, a median of 14 minutes per case against 26 before agents. Every verdict the physician changed is listed with the reason and its second signature — interstitial lung disease, for one, moved to unexpected because the outcome was fatal and the label lists only non-fatal pneumonitis.
The same case, seen by the four people who handle it — what their week looked like, and what it looks like now.
Six specialist agents pick the right reference, search it, read the clause and lay out the causality evidence; a safety physician decides and signs.
Chooses the reference documents that apply to each event: the IB edition in force on the onset date for trials; USPI, CCDS, SmPC and local label by market for marketed products.
Expands each preferred term to its synonyms and the other terms in the same high-level term, searches the applicable documents and marks each hit as the same, lesser or greater severity.
Reads the hits in context and proposes listed, unexpected or “your judgement” for each event and document, quoting the clause it rests on.
Builds the exposure and event timeline and lists the evidence for and against a causal role — onset, dechallenge, rechallenge, other causes, known effects — with a WHO-UMC or RUCAM suggestion.
Drafts the question to the investigator or reporter when evidence is missing — baseline labs, other medicines, rechallenge, autopsy.
Works out what each authority needs and by when — SUSAR 7 or 15 days, US 15-day Alert, EU 15-day, periodic — from the signed verdicts and causality, and queues the report once the physician signs.
Decides and signs. The agents propose; a named person decides.
The physician can ask in plain words about any open case, reference document or review rule — or tell it what to change. Every answer points to the passage behind it.
Why is the liver injury on MR-26-1184 unexpected?
Because the onset was 22 Sep 2026, and the RSI in force on that date decides — IB Edition 9. Raised enzymes are listed, but liver injury is greater in severity, which makes it unexpected. Edition 10 lists it — but only from Nov 1. With the investigator’s “related”, it is a 15-day SUSAR due Oct 18.
Which open cases change when IB Edition 10 takes effect?
None of today’s open cases change — each is judged on the edition in force at onset, and every onset is before Nov 1. What changes is new events from Nov 1: drug-induced liver injury and life-threatening anaphylaxis become listed. MR-26-1184 and MR-26-1190 keep Edition 9.
Where do Corventa’s labels disagree?
One serious term: toxic epidermal necrolysis. CCDS v14 and the SmPC list it; the USPI of 06/2025 lists erythema multiforme only. Because a US 15-day Alert report is judged on the current labeling, MR-26-1179 is unexpected for the US and due Oct 16 — until the USPI update filed on Sep 2 is approved.
Flag every case where the USPI lags the CCDS
Done. New rule: flag a case when its event is in the CCDS but not yet in the USPI. Running it on the open cases finds one — MR-26-1179, toxic epidermal necrolysis on Corventa. The flag shows on the case and Lena Ortiz gets a weekly list. The rule is live in Settings and recorded in the audit trail.
13 screens from the working solution, on its sample data. Pick one to see it large.
Which document applies, how deep the search goes, the causality convention and who signs second are all settings.
| Setting | Default | Choose from |
|---|---|---|
| Use the reference in force on the event onset date | On | IB editions by onset date; labels by the date the case was received |
| Fatal or life-threatening reactions are unexpected unless the RSI says so | On | Listed terms without “life-threatening” go to the physician for judgement |
| Search MedDRA neighbours | Same high-level term | Preferred term only · Same high-level term · Same high-level group term |
| Judge US cases on the current USPI | On | The CCDS is shown beside it but never decides a 15-day Alert report |
| Causality convention | Reasonable possibility · yes or no | Reasonable possibility · yes or no · WHO-UMC categories |
| Suggest a RUCAM score for liver events | On | Shown beside the WHO-UMC suggestion |
| Second signature when fatal or life-threatening and unexpected | Dr. Omar Haddad | Dr. Omar Haddad · Dr. Priya Raman |
| Approves a new edition before agents use it | Lena Ortiz | Lena Ortiz · Dana Okafor |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample cases · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published rules (21 CFR 312.32 and 314.80, ICH E2A — 7- and 15-day expedited reporting). People, companies and products named on this page are fictional — characters and sample data in the working solution.
It is the decision whether a serious adverse reaction is listed in the applicable reference safety information — the IB’s RSI for a trial, the label for a marketed product — at the same severity and specificity. Together with seriousness and causality, it decides whether a case is reported on an expedited clock. The solution proposes the verdict with the clause quoted; the safety physician decides.
The reference picker reads the onset date of each event and chooses the investigator’s brochure edition in force on that date — never a newer approved edition that is not yet in effect. For marketed products it uses the USPI for US 15-day Alert decisions and the SmPC for the EU, and always shows the CCDS beside them.
Yes. The MedDRA neighbour search looks for the reported preferred term, its synonyms and the other terms in the same high-level term, and marks each hit as the same, lesser or greater severity. A lesser-severity listing such as “transaminases increased” does not make liver injury listed. The search depth is a setting.
The causality evidence agent builds the exposure and event timeline and lists what supports and what argues against a causal role — time to onset, dechallenge, rechallenge, other causes, known effects — with a WHO-UMC or RUCAM suggestion. It never records causality for the physician, and the investigator’s “related” is never downgraded for expedited reporting.
Yes. The agents only propose. A safety physician agrees with or changes each verdict, records causality and signs the assessment with an electronic signature. Fatal or life-threatening unexpected events and changed verdicts need a second physician’s signature before the report is queued.
Yes. Causality can be recorded as reasonable possibility yes or no, or as WHO-UMC categories, and a RUCAM score can be suggested for liver events. The second-signature approvers and who approves a new reference edition are settings too.
Your safety database for cases, MedDRA coding and E2B(R3) submissions, your labeling repository for CCDS, USPI, SmPC and IB editions with their effective dates, MedDRA, and the case follow-up mailbox for questions to investigators and reporters.
The Agentic Solution Engine builds and deploys it from your requirements — your medical review SOP, reference documents and a sample of past serious cases — and it goes live once every quality gate has passed. We will walk you through it on your own cases first.
Every side-effect report becomes a cited case in minutes, filed with days to spare.
1-min filmSafety LiteratureEvery journal article screened with a written reason, and reportable cases at intake in about a day.
1-min filmSignal ManagementEvery potential safety signal gets its evidence assembled in days, each number traced to its query.
1-min filmWe’ll run Expectedness Review on a sample of your own serious cases and reference documents.