CRO toxicology report review in hours, every finding and margin cited to its page
Every CRO toxicology report checked against its own tables in hours, with one cited safety table for the programme.
See one case, screen by screen ↓A draft 13-week report arrives from the CRO with 300 to 700 pages and a comment window of 10 to 20 days. The study monitor has to check that every statement in the summary is in the tables, that incidences match the individual animal listings, that the exposure at the NOAEL is the right day and sex from the TK report, and that every out-of-range formulation result has a recorded deviation.
Then the same numbers have to be carried into the programme’s cross-study table and recomputed as margins against the clinical dose. Done by hand, this is days of page-turning per draft. A Day 1 AUC quoted as Day 91, or one testis finding where the listing shows two, slips through — and moves the NOAEL discussion, the margin and the Investigator’s Brochure with it.
Estimated split for one 13-week repeat-dose draft, in hours, by hand and with the solution.
Six specialist agents read, extract, compute margins, check and draft the CRO letter; the toxicologist decides each issue and signs.
Draft v1 of CRV-T-013 — a 13-week oral toxicity study of Corventa in Wistar Han rats with a 4-week recovery, 412 pages from Ashgrove Preclinical — is waiting for the study monitor. Here is the review, screen by screen, in the working solution.
Priya opens the safety map. Across her programmes, 4 drafts are read and checked by agents, 14 discrepancies are open, and the next comments are due to Corrin BioServices in 7 days. Corventa has 8 studies on the map. The target-organ grid shows liver findings in 4 studies, and two cells of the 13-week rat study flagged — thyroid and testis. The margins at the NOAEL sit on one scale against 200 mg once daily.
“The signed GLP report stays the official record. Every value here is a cited working copy — open any cell to see its page.”
The report reader split 412 pages into 15 sections and 59 tables and matched the draft to its protocol and amendments. The findings extractor pulled 214 findings, 7 of them test article-related, with SEND terms. The margin calculator read the TK tables; the consistency checker ran 7 checks and found 6 discrepancies; the cross-study check compared the draft with 7 other Corventa studies and flagged thyroid.
The last step reads “Study monitor decides — waiting for Dr. Priya Raman”.
Centrilobular hepatocellular hypertrophy: 0/10, 0/10, 6/10 and 10/10 males across 0, 10, 30 and 100 mg/kg/day, with severity, the adverse or adaptive call as the report states it, and recovery. The CRO’s own words sit beside the SEND term, and each value links to the page it came from. Priya confirms the findings one by one or all at once.
No value without a page citation; low-confidence values go to the reviewer.
ALT at 30 mg/kg/day missing from the summary. Testis degeneration in 1/10 males in the summary, 2/10 in Table 22. The Day 1 AUC quoted as the Day 91 exposure at the NOAEL. A Week 6 high-dose formulation at 87.0% of nominal with no deviation recorded. Thyroid missing from the target-organ list. And a minor one: an organ weight header that says “% body weight” over values in grams.
For D1 the summary on page 6 says there were no test article-related clinical chemistry changes at 10 or 30 mg/kg/day. Table 14 on page 188 shows male ALT at 54 ± 13 U/L against 37 in controls — +46%, p<0.05. The checker gives its reason: the mid-dose value is above the historical control range of 24–52 U/L and comes with hypertrophy and higher liver weights at the same dose.
The checker never decides — it proposes to the study monitor.
The summary quotes 48,200 ng·h/mL at the NOAEL — the Day 1 value — which gives 7.9×. TK Table 3 gives 42,800 ng·h/mL in males on Day 91, so against 6,100 ng·h/mL at 200 mg the margin is 7.0×, set by the lower-exposure sex. The human-equivalent dose is 290 mg (30 mg/kg ÷ 6.2 × 60 kg). Switch to 400 mg and the rat margin falls to 3.3×.
“13-week rat study supports clinical trials up to 3 months (ICH M3(R2)) — planned: 200 mg once daily · 12 weeks.”
Priya adds D1 to D5 to the comment letter. She dismisses the minor table-label item with a reason from the list — not a discrepancy, already raised with the CRO, or to be fixed by the CRO in the final report. The reason stays with the discrepancy and goes into the audit trail.
To Dr. Elin Varga, Study Director at Ashgrove Preclinical: five comments, each with its location, what is wrong, what the sponsor asks for, and its class. It asks for a reply within 10 working days, by Oct 21. It can be downloaded as .docx, and leaves through the CRO portal only after Priya signs.
“The next draft is re-checked against every item.”
The signature dialog shows 5 comments to Ashgrove and states that signing confirms the remaining 7 findings as extracted. Priya chooses the meaning — “Reviewed — comments sent to the CRO” — and enters her password. Her signature, its meaning, the time and the report version go into the audit trail; the letter goes to the CRO and the cross-study table is refreshed.
Seven Corventa studies — 5 final, 2 draft labelled “may change” — with species, route, doses, target organs, NOAEL, exposure, margin and HED, and 43 values that each open their source page. The lowest margin is 4.9× in the 4-week dog study. Liver is the only target seen in both species; thyroid is rodent-only. It exports as a CTD 2.6.7 table, a brochure section 5 table or a governance slide.
“Working copy · QC against the signed reports before submission.”
The same draft, seen by the five people who carry it — the monitor, the safety lead, the pathologist, the writer and QA.
Six specialist agents read, extract, compute margins, check and draft the CRO letter; the toxicologist decides each issue and signs.
Splits each CRO report into summary, methods, results, tables, listings, pathology, TK and appendices, keeps page numbers, and matches it to its protocol, amendments and earlier versions.
Extracts dose groups, findings by organ with SEND terms, incidence by sex and dose, severity, reversibility and the stated NOAEL — each value with its page.
Reads the TK tables, picks the right day, sex and dose, and computes exposure margins and human-equivalent doses against the clinical exposure reference.
Checks the summary against the tables, the tables against the individual listings, the TK report against the summary, the report against protocol and amendments, and the study against the programme.
Turns the accepted discrepancies into a numbered comment table for the CRO study director.
Keeps the programme table of target organs, NOAELs, reversibility, exposure and margins up to date, every cell cited.
Decides each issue and signs. The agents propose; a named person decides.
The study team can ask about any CRO report, NOAEL or exposure margin across the programmes — or tell it what to check. Answers point to the report page.
What is our Corventa margin at 400 mg?
At 400 mg once daily (human AUC 12,900 ng·h/mL) the dog sets the lowest margins — 2.3× in the 4-week study and 2.4× in the 13-week study. The 13-week rat falls to 3.3× using the Day 91 value from TK Table 3. The hERG margin stays above 20× against free Cmax. Liver is the target in both species, so 400 mg would need liver monitoring.
Why did the margin change from 7.9× to 7.0×?
The summary gives the Day 91 male AUC at the NOAEL as 48,200 ng·h/mL, but that is the Day 1 value in TK Table 3. The Day 91 value is 42,800. Against 6,100 ng·h/mL at 200 mg, the margin is 7.0×, not 7.9×. Females are higher (61,500), so males set the margin.
Which target organs are consistent across species?
For Corventa, liver is the only target organ in both species — hypertrophy with ALT up to 2.1-fold in the 13-week rat and ALT up to 1.8-fold in the 13-week dog. Thyroid follicular hypertrophy is rat-only and adaptive (liver enzyme induction). Testis degeneration is rat-only, at 100 mg/kg/day, and not fully reversible. GI intolerance is dog-only.
Add a check: flag NOAELs set at the top dose with no findings
Done. The new check, “NOAEL at the highest dose tested”, asks whether exposure was adequate (for example ADA or saturation) and whether a higher dose or a limit dose was justified under ICH M3(R2). It would have flagged VLT-T-031 today, where three monkeys at the low dose lost exposure. It runs from the next report.
13 screens from the working solution, on its sample data. Pick one to see it large.
What the reviewer checks, the clinical exposure it compares against, and who signs are all settings.
| Setting | Default | Choose from |
|---|---|---|
| Checks the reviewer runs | 7 on | Summary vs tables · incidence vs listings · TK · protocol · formulation · programme · SEND terms |
| Summary-vs-table mismatch that changes a NOAEL, margin or target organ | Major | Major · Minor |
| Clinical exposure reference | Corventa 200 mg · AUC 6,100 ng·h/mL | AUC and Cmax for each planned dose |
| Signs Corventa reviews | Dr. Priya Raman | Dr. Priya Raman · Dr. Omar Haddad · Dr. Sam Patel · Lena Ortiz · Dana Okafor |
| Adverse vs adaptive calls | Dr. Sam Patel, pathologist | Dr. Priya Raman · Dr. Omar Haddad · Dr. Sam Patel · Lena Ortiz · Dana Okafor |
| CRO comment window | 10 days · Ashgrove Preclinical | 10 · 15 · 20 days, per CRO |
| SEND controlled terminology | 2026-06 release | 2026-06 · 2026-03 release |
| Severity grading scale | 5-grade (minimal–severe) | 5-grade · 4-grade |
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 programme data · “target” = the design goal, measured in the live solution · “estimated” = our estimate. Context: ICH M3(R2) · 21 CFR 58 (GLP). People, companies and products named on this page are fictional — characters and sample data in the working solution.
It is a solution that reviews CRO toxicology and safety-pharmacology reports for the sponsor. Six agents read each draft, extract findings, NOAELs and TK exposures with their pages, compute exposure margins and check the report against itself, its protocol and the rest of the programme. The study monitor decides each issue and signs the review.
Seven checks by default: the summary against the tables, incidence against the individual animal listings, the TK report against the summary, the report against the protocol and amendments, dose formulation results against recorded deviations, target organs across the programme, and finding terms against SEND. You can turn each one off or add your own.
The exposure and margin calculator reads the TK tables at the last sampling day, uses the lower-exposure sex, and divides the AUC (or Cmax, or IC50 against free Cmax) at the NOAEL by the clinical exposure in your reference. It states the basis of every margin, shows the human-equivalent dose by body-surface area, and recomputes when you change the clinical dose.
No. The source PDF is never altered and the signed GLP report stays the official record. What the solution holds is a cited working copy; your comments go to the CRO, and the CRO issues the next version.
Yes. The agents propose; the study monitor adds each discrepancy to the letter or dismisses it with a reason, a pathologist confirms adverse or adaptive calls, and nothing goes to the CRO until the review is signed with a password and the meaning of the signature.
Yes. The cross-study table builder keeps target organs, NOAELs, exposure, margins and HED for every study in the programme, from confirmed findings only, with draft values labelled. It exports as a CTD 2.6.7 table, a brochure section 5 table or a governance slide, with source references for QC against the signed reports.
GLP and non-GLP repeat-dose, safety pharmacology and genetic toxicology reports, with their TK and pathology reports and protocols — in the working solution, rat, dog and cynomolgus monkey studies, a hERG assay, dog telemetry and a micronucleus test, for small molecules and an antibody.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your checks, severity rules, clinical exposure reference, CRO contacts and a sample of past CRO reports — and it goes live once every quality gate has passed. We will walk you through it on your own reports first.
Ask the literature in plain words and get a sourced answer in seconds.
1-min filmPipeline WatchEvery rival trial, readout and approval lands in a sourced brief for each programme, every Monday.
1-min filmPreclinical ReportsRaw study data becomes a draft report in hours, with every number traced to its source.
1-min filmWe’ll run the Nonclinical Study Report Reviewer on a sample of your own CRO draft reports.