Clinical trial protocol writing — from synopsis to a checked first draft, in your own approved language
A trial synopsis becomes a checked first-draft protocol in days, in your own approved language.
See one case, screen by screen ↓The synopsis arrives from Clinical Development and the medical writer starts again from a blank sponsor template. Eligibility criteria, safety definitions, the estimand, the schedule of activities: most of it has been written and approved before, in an earlier protocol for the same drug or the same population. So the writer opens the old protocols and copies by hand, section by section, and changes the words that must change.
That is where the trouble starts. A schedule adapted from a Phase 2 study still assesses the endpoint at the old week. A contraception period changes in Section 5 but not in Section 9. Nobody sees it until team review — or until the protocol is with sites and the fix becomes a substantial amendment.
Estimated split for one Phase 3 protocol, from synopsis to first draft, by hand and with the solution. The by-hand total is the working solution’s own baseline of 38 days.
Five specialist agents read, reuse, write and check every section; a medical writer approves each one.
VLT-302 is a Phase 3 trial of Veltrimab in moderate-to-severe plaque psoriasis. Its synopsis arrives at 8:12. Here is what happens next, screen by screen, in the working solution.
Lena opens the solution: 10 protocols in progress, 31 sections to approve across 7 drafts, 8 consistency checks to fix (2 marked high) and 8 open questions waiting on 3 people. At the top of her list, VLT-302: a new synopsis from Clinical Development, ready to draft, due Oct 23.
Agents at work: “Synopsis reader — Read VLT-302 synopsis v1.0 · 31 fields from 9 pages.”
The synopsis reader has already gone through all 9 pages: co-primary endpoints PASI 90 and sPGA 0/1 at Week 16, adults with PASI ≥12, sPGA ≥3 and BSA ≥10%, 150 mg SC at Weeks 0 and 4 then every 12 weeks, about 660 participants randomized 2:1 at about 120 sites in 14 countries. Next to it, the closest approved protocols: VLT-201 at 92% match — same drug and population, Phase 2.
Its rule: quote endpoint wording, never paraphrase — and mark a timepoint as missing rather than guess it.
Lena starts the draft. The agents map the synopsis to Sponsor template v3.2, search 38 final protocols and find 112 passages that fit, then write all 14 sections in parallel. The overview shows where the text came from: 58% approved language with its source, 29% new from the synopsis and cited to the page — with 4 checks to fix and 4 open questions for 3 people.
“First draft ready — 14 sections in your template. Written by 14 section writers in 3.2 s · v0.1 · every reused passage keeps its source.”
Trial Population: 10 criteria reused word for word, 2 adapted with the changed words marked. IGA ≥3 became sPGA ≥3 to match the synopsis; contraception until 16 weeks became 20 weeks after the last dose — five half-lives, per the synopsis. The IL-23 class exclusion comes from VLT-115, because VLT-201 predates it. Every criterion carries its chip: VLT-201 · p. 31.
Any chip opens the source document at the exact page: here VLT-201 Protocol, Amendment 1 · v3.0 · Final, page 31, with the reused passage highlighted and a note on what changed in this draft. Reused, adapted, new from the synopsis or standard template text — each passage says which it is.
“Adapted in this draft. IGA ≥3 became sPGA ≥3 to match the synopsis.”
One high check: Section 3 sets both co-primary endpoints at Week 16, but the schedule of activities was adapted from VLT-201, whose primary endpoint was at Week 12 — it assesses PASI, sPGA and BSA at Weeks 12 and 20, not at Week 16. Three more: Section 9.4 still says 16 weeks of pregnancy reporting, visit windows disagree (±3 days against ±7), and sPGA is used before it is defined. Each finding shows both passages and one suggested fix.
The checker proposes — it never applies a fix itself.
Lena accepts the fix. PASI, sPGA and BSA are added to the Week 16 visit in Table 1, the draft is saved as v0.2, and the check moves to Fixed with her name on it. The medium and low checks wait for her in the same list.
“Schedule updated — PASI, sPGA and BSA added at Week 16 · saved as v0.2.”
Four gaps become four short questions. Dr. Maya Chen: should prior biologic use be capped — VLT-201 excluded more than two, the synopsis says nothing, the draft keeps two — and is the Week 16 switch blinded? Sam Patel: which testing order for the key secondary endpoints? Dr. Priya Raman: is a data monitoring committee needed? Each comes with its options and an “Ask” button that sends the passage along.
One question per decision; every answer is recorded with name and time and written back into the section.
Lena reads Objectives and Estimands — the co-primary endpoints at Week 16, the composite estimand adapted from VLT-201 with Week 12 changed to Week 16 — and approves it. Sam Patel, its reviewer, is notified. Approving a section with a check still open asks her first, and records that she saw it.
“Section 3 approved by Lena Ortiz · Sam Patel notified.”
VLT-212, the palmoplantar pustulosis protocol, shows where VLT-302 is heading: all 14 sections approved. Sending it runs a full consistency re-check of the synopsis against every section, confirms the clinical, statistics, operations and safety reviewers have approved, and sends v0.9 to Dana Okafor. She signs with a reason, and the version is locked.
“VLT-212 v0.9 sent to Dana Okafor for e-signature.”
The same protocol, seen by the four people who carry it — what their weeks looked like, and what they look like now.
Five specialist agents read, reuse, write and check every section; a medical writer approves each one.
Reads the trial synopsis and pulls out design, population, objectives, endpoints, dosing, visits and sample size as 31 structured fields, each with the exact wording and its page.
Searches the library for approved passages that fit each template section, preferring the same drug and population, and returns each with its document, version, page and the differences the writer must see.
Writes one template section from the synopsis fields and the retrieved passages — 14 writers in parallel. Reuses a passage verbatim when it fits, marks every changed word when it adapts one, and writes new text only from the synopsis, cited.
Checks the draft against the synopsis and against itself: objectives, endpoints and estimands, eligibility, dose and regimen, visit windows and the schedule of activities. Shows both passages for every mismatch and proposes one fix.
Turns gaps in the synopsis into short questions for the right reviewer and writes the answers back into the section.
Approves every section. The agents propose; a named person decides.
Ask about any protocol, section or source — or tell it what to change. Answers point to the passage they rest on.
Where did the eligibility text come from?
All 12 criteria in Section 5 come from approved, final protocols. 10 are word for word — 9 from VLT-201 §5.1–5.2 and the IL-23 class exclusion from VLT-115. 2 are adapted, with the changes marked: sPGA ≥3 instead of IGA ≥3, and contraception until 20 weeks after the last dose. The cap on prior biologics was kept from VLT-201 and has an open question for Dr. Maya Chen.
What is wrong with the schedule?
The co-primary endpoints are at Week 16, but the schedule was adapted from VLT-201, whose primary endpoint was at Week 12 — it assesses PASI, sPGA and BSA at Weeks 12 and 20 only. The suggested fix adds them to the Week 16 visit (Visit 6, Day 113). It is the one high check in the draft.
Add a check: the contraception period must match in every section
Added to Settings → Consistency checks. It runs on every draft and every edit from now on. In VLT-302 it already matches one finding: Section 5 says 20 weeks after the last dose, Section 9.4 still says 16 weeks.
Who approves and signs?
You approve each section as the medical writer; the named reviewer is notified. When all 14 are approved, the draft gets a full re-check and team review, then Dana Okafor signs with a reason and the version is locked.
13 screens from the working solution, on its sample data. Pick one to see it large.
What differs between sponsors and therapeutic areas is a setting, not a project.
| Setting | Default | Choose from |
|---|---|---|
| Protocol template | Sponsor template v3.2 (ICH M11) | Sponsor template v3.2 (ICH M11) · v3.1 (legacy) |
| Style guide | Medical writing style guide 2026 | Style guide 2026 · 2025 |
| What the Word export carries | Sources as Word comments | Sources as Word comments · Clean copy, sources in an appendix |
| Who reviews each section | Clinical lead, biostatistician, clinical operations, safety physician | Any named reviewer, after the writer approves |
| Who signs the final version | Head of medical writing | Any named signer |
| Reuse only from final protocols | On | Drafts and superseded versions are never a source |
| Oldest language allowed | 5 years | 1 to 15 years since the source protocol was finalized |
| Consistency checks | 6 of 7 on | Each on or off · add your own in plain words |
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 protocols · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (Tufts CSDD protocol amendment studies, 2016 and 2024 — 45% of substantial amendments avoidable; median $535K per substantial amendment in Phase 3). People, companies and products named on this page are fictional — characters and sample data in the working solution.
Software that helps medical writers turn a trial synopsis into a full clinical trial protocol in the sponsor’s template. Protocol Author does it with five agents: one reads the synopsis, one finds approved language in your final protocols, fourteen section writers draft in parallel, one checks the draft for consistency and one collects the open questions. A medical writer approves every section.
It searches your library of approved, final protocols for passages that fit each template section, preferring the same drug and population. A passage is reused word for word when it fits; when it has to change, every changed word is marked. Every reused passage keeps its document, version and page.
It drafts into your protocol template — in the working solution, a 14-section sponsor template in ICH M11 style, from Protocol Summary to References. The template and your style guide are settings, so the draft follows your sections, headings and standard text.
The consistency checker compares the draft with the synopsis and with itself: endpoint timepoints against the schedule of activities, objectives against endpoints and estimands, eligibility against the synopsis population, dose and regimen, visit windows and abbreviations. Each finding shows both passages and a suggested fix; the writer accepts it or dismisses it with a reason.
No. New text is written only from the synopsis and cited to its page, and endpoint wording is quoted, never paraphrased. If the synopsis does not state something the template needs — a testing order, a cap on prior therapies — it becomes a question for a named person.
Yes. Agents draft and propose; the medical writer approves each section, named reviewers review their sections after that, and the head of medical writing signs the final version with a reason before it is locked. Every step is in the version history.
Yes. The draft exports as a Word document with its sources as Word comments — or as a clean copy with the sources in an appendix, if you set it that way.
The Agentic Solution Engine builds and deploys it from your requirements — your protocol template, style guide and library of final protocols — and it goes live once every quality gate has passed. We will walk you through it on one of your own synopses first.
Every 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 filmProtocol Risk ReviewEvery draft protocol checked against past amendments, so avoidable changes are fixed before the committee meets.
1-min filmWe’ll draft a protocol from one of your synopses, in your template, from your own final protocols.