EU CTR lay summaries drafted from the study report, every fact checked, in every language
Every trial’s lay summary drafted within days, each fact checked against the study report.
See one case, screen by screen ↓Every trial run in the EU must post a summary of its results for the public: 12 months after the end of the trial, 6 months for trials in children, 30 months for early studies in healthy volunteers. Each one needs the ten Annex V elements, written so a participant can follow them, and translated into the language of every country where the trial ran.
The hard part is not the writing. It is turning “treatment-emergent adverse events” and “PASI 90” into plain words without changing a single number, keeping the comments section neutral, and proving every sentence back to the CSR before the patient panel, medical, legal and a translator for every language each take their turn. An agency draft typically takes 2–4 weeks, translations add more on top, and the review window shrinks with each round.
Estimated split for a typical Phase 3 summary in nine languages (English and eight translations), by hand and with the solution.
Six specialist agents read, draft, check, reader-test and translate every summary; the writer and reviewers decide each flag and sign.
VLT-302 is a Phase 3 trial of Veltrimab in adults with moderate to severe plaque psoriasis. Its final study report arrived on October 2; its EU posting deadline is January 14, 2027. Here is what happens next, screen by screen, in the working solution.
Lena opens the solution to a posting runway: each summary is a bead at its EU posting deadline, with the 6-month deadlines for trials in children marked. 10 summaries are open, 3 are due in 60 days, none was late this year, and the average reading grade is 7.4. Five covers on her desk need her — VLT-302 among them, with 3 sentences waiting.
Agents at work: “Fact checker · VLT-302 · 29 statements checked against the study report · 3 need Lena.”
Early that morning the study report reader went through the synopsis and 14 results tables, and the Annex V drafter wrote all ten required sections as the questions a reader would ask — “Who took part in this study?”, “What medical problems did people have?”. Each factual sentence carries a numbered citation. The checklist on the left shows three sections flagged; the first flag is study-report jargon: “treatment-emergent adverse events” becomes “at least one medical problem”, with the line that not every problem was caused by the medicine.
“Drafted by agents · 29 statements cited · 3 flagged for you · grade 8.9”
The draft says about 7 out of 10 people on Veltrimab had almost clear skin after 16 weeks. The fact checker compares it with the report: 277 of 428 people, 64.7 %. Rounding to 7 out of 10 overstates the result by 5 points, so it proposes “about 65 out of 100” — the same number the picture of 100 dots in that section shows.
The fact checker cannot change the text. It proposes a fix; Lena accepts, edits or dismisses it.
“Open at this passage” takes Lena to the VLT-302 study report, Table 11.4, primary endpoint, page 52 of 1,284 — with the words the sentence relies on highlighted: PASI 90 at Week 16, veltrimab 277/428 (64.7 %), placebo 10/214 (4.7 %).
The draft’s “What do the results mean?” section says Veltrimab “worked very well and is a safe treatment for psoriasis”. The tone checker flags it as promotional: claims beyond this trial. It proposes a neutral line and shows the house-style rule from Good Lay Summary Practice it rests on.
Proposed: “In this study, more people who got Veltrimab had clearer skin than people who got the placebo. This is the result of one study. Other studies may find different results.”
Lena accepts the three fixes. The reading grade drops from 8.9 to 7.1 — inside the team’s limit of grade 8, about the level of a popular newspaper. Reading ease is 68, the average sentence 14.7 words against a target of under 15, every number is written as “x out of 100”, and 23 plain-word swaps came from the shared glossary.
Readability scores miss what people misunderstand, so the reader test answers eight key questions from the summary alone, then compares the answers with the study report. On the first draft it got three wrong — it read the main result as 7 out of 10, could not explain “treatment-emergent adverse events”, and took “a safe treatment” to mean one study proves safety. After the fixes: 8 of 8.
The reader test has no access to the study report while it answers.
Until every flag was resolved, the button read “3 to resolve before signing”. Now it opens the signature: 29 of 29 statements match the study report, reading grade 7.1, reader test 8 of 8. Lena signs with her password, and the meaning is recorded with it.
“I am the author and confirm every fact matches the final study report”
VLT-302 moves to patient read-through. The sign-off chain shows who is next: Priya Raman’s panel reads it, Dr. Maya Chen signs medical accuracy, Dana Okafor the non-promotional check, Sam Patel approves the translations and Omar Haddad the posting package. Translations into eight languages follow from the signed English, and every accept and signature is on the audit trail.
The dashboard answers the disclosure team’s two questions: are we posting on time, and are the summaries clear and correct? In the last six months, 9 of 9 summaries were posted on time, with a median of 236 days from end of trial to posting against a limit of 365. The fact check caught 61 flags, and a person decided every one.
The same summary, seen by the five people who carry it to posting — what their week looked like, and what it looks like now.
Six specialist agents read, draft, check, reader-test and translate every summary; the writer and reviewers decide each flag and sign.
Reads the study report synopsis and results tables and pulls out every fact a lay summary needs — identifiers, dates, sites, participants, medicines, results, adverse reactions, follow-up studies — each with its page and table.
Writes the ten required sections in plain words, as questions a reader would ask, using only the extracted facts and the plain-word glossary.
Matches every factual sentence back to the study report and flags any number, group or time point that differs, any rounding that moves a number by more than 2 points, and any fact without a source.
Scores reading grade and reading ease per section, finds study-report terms that have a glossary entry, and flags promotional or overreaching wording with a neutral alternative.
Answers eight key questions using only the summary, the way a lay reader would, and marks an answer wrong if it differs from the study report or the summary does not let a reader answer it.
Translates the approved English into each country’s language with glossary terms locked, then translates it back into English and compares every number, group and time point.
Decide each flag, sign to post. The agents propose; a named person decides.
Anyone on the team can ask about a summary, a deadline or a check — or tell it what to change.
Why was the results sentence in VLT-302 flagged?
The draft said “about 7 out of 10” people on Veltrimab had almost clear skin. The study report says 277 of 428 — 64.7 %. Rounding to 7 out of 10 overstates the result by 5 points, so the fact checker proposes “about 65 out of 100”, which matches the picture of 100 dots in the same section. The reader test also got this question wrong from the draft.
Which summaries are closest to their deadline?
NXL-205 is ready to post, with 21 days to its deadline — Omar Haddad approved the package, so it can be posted today. CRV-210 is in translation with 44 days left. VLT-305 is a trial in children, so its deadline is 6 months after the end of the trial, not 12: 56 days left.
How do we explain a study that stopped early?
CRV-220 stopped early, so the summary must say why and whether safety was involved. The draft said “terminated prematurely for strategic reasons”; the report says it was stopped for business reasons, not because of a safety concern. It also said “about half” took part — the report says 96 of 240 planned, about 4 out of 10.
Require a patient read-through for every summary of a trial in children
Done. New rule: summaries of trials in children always go to Priya Raman’s panel, with the young readers version (ages 8–12), before medical sign-off. Two open summaries match today — VLT-305 and NXL-310. 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.
Who signs what, how plain is plain enough, and the rules every summary follows — all settings.
| Setting | Default | Choose from |
|---|---|---|
| Highest reading grade | Grade 8 | Grade 6 to 10 |
| How numbers are written | About 65 out of 100 people | About 65 out of 100 people · 65% of people · Both — 65 out of 100 (65%) |
| Block posting while any fact is unchecked | On | On · Off |
| Flag promotional words | “safe”, “effective”, “breakthrough”, “worked very well” and 40 more | On · Off |
| Reader test before patient read-through | At least 7 of 8 key questions | On · Off |
| Version for young readers | Ages 8–12, for trials in children | On · Off |
| Back-check numbers in every translation | On | On · Off |
| Who signs each step | Patient read-through: Priya Raman · Medical accuracy: Dr. Maya Chen · Non-promotional check: Dana Okafor · Translations: Sam Patel · Posting package: Omar Haddad | Any of the six people on the team |
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 trial data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published or industry figures (agency drafts typically take 2–4 weeks) · Sources: EU Clinical Trials Regulation 536/2014, Art. 37 and Annex V · EC Good Lay Summary Practice (2021). People, medicines and studies named on this page are characters and sample data in the working solution.
A summary of a clinical trial’s results written for the public, which the EU trials regulation requires for every trial: 12 months after the end of the trial, 6 months for trials in children, and 30 months for early studies in healthy volunteers. Annex V lists the ten elements it must contain, from trial identification and who took part to the results, the medical problems people had and where to learn more.
The study report reader pulls every fact the summary needs from the CSR synopsis and results tables, with page and table. The Annex V drafter writes the ten sections as questions a reader would ask, using only those facts and your plain-word glossary, with numbers as natural frequencies such as “about 65 out of 100 people”.
The fact checker matches every factual sentence back to the study report and compares each number, unit, group and time point. It flags any mismatch, any rounding that moves a number by more than 2 points, and any fact without a source. Posting stays blocked while a fact is unchecked.
Yes. The tone checker flags words such as “safe”, “effective”, “breakthrough” and “worked very well”, and any conclusion beyond the trial, then proposes a neutral line with the reason. Dana Okafor’s non-promotional check is a separate signature.
Besides the reading grade, the reader test answers eight key questions from the summary alone and compares the answers with the study report. A wrong answer points to the sentence that misled it. The patient panel then reads it, with a young readers version for trials in children.
Into the language of every country where the trial ran, starting only from the signed English. Glossary terms are locked per language, each translation is read back into English and every number compared, and a term with no locked translation goes to the translation coordinator as a query.
Yes. Agents propose and people decide: every flag is accepted, edited or dismissed with a reason, and every step — author, patient read-through, medical accuracy, non-promotional check, translations and posting package — is signed by a named person, with the meaning of the signature recorded.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your Annex V template, plain-word glossary and a sample of final study reports — and it goes live once every quality gate has passed. We will walk you through it on one of your own study reports first.
A trial synopsis becomes a checked first-draft protocol in days, in your own approved language.
1-min filmTMF CheckEvery trial document is read, filed in the right place and checked, so the file is inspection-ready every day.
1-min filmCSR WriterResults sections and patient narratives are drafted within days of database lock, every number linked to its table.
1-min filmWe’ll run the Plain-Language Results Summary Writer on one of your own final study reports.