Clinical protocol review that finds the avoidable amendments before the committee meets
Every draft protocol checked against past amendments, so avoidable changes are fixed before the committee meets.
See one case, screen by screen ↓Most amendments that cost a study months were visible in the draft: an inclusion criterion stricter than the target label, a screening window too short for a centrally read colonoscopy, a primary estimand that leaves intercurrent events to the analysis plan, a Schedule of Activities that disagrees with the synopsis. The sponsor has usually made the same mistake before — the summary of changes is sitting in the TMF.
But nobody on a protocol team has time to read 148 summaries of changes, four design standards and the ICH guidelines against a 148-page draft before the committee date. So review comes down to whoever remembers what went wrong on the last study, and the lesson is relearned after first participant in — with re-consent, ethics resubmissions and months added to the timeline.
Estimated split for one Phase 3 draft before the review committee, by hand and with the solution.
Six reviewer agents read every section and rank the risks; protocol owners decide each finding and the lead signs.
VLT-305, a Phase 3 induction and maintenance study of Veltrimab in Crohn’s disease — draft v0.9, 148 pages, 640 participants at about 180 sites in 22 countries — goes to the Protocol Review Committee on Oct 14. Seven days out, here is how the protocol lead takes it from crew review to signed dispositions, screen by screen, in the working solution.
Maya opens the solution to 8 draft protocols in review, 4 of them reaching the committee in the next two weeks, and 26 findings waiting for an owner — 6 of them hers. Each draft shows where it is likely to need an amendment, section by section. VLT-305 stands out: 73% amendment risk, committee in 7 days.
“Expected cost of avoidable amendments: $1.46M and 9.3 months of delay across drafts.”
Maya presses Re-run crew. The draft is split into its sections, the synopsis and the Schedule of Activities, and five reviewers read in parallel — eligibility, schedule and burden, endpoints and estimands, safety monitoring and operational feasibility. Then the lead reviewer compares the synopsis, body text and Table 2, merges 4 duplicate findings, estimates each finding’s likelihood from the 148 past amendments and ranks 12 findings.
Owners assigned by area: “eligibility → protocol lead · estimands → biostatistics · safety → safety physician”.
At the top, 22% likely to force an amendment: Inclusion 3 requires failure of at least two prior biologic classes. In VLT-204 the same criterion held enrolment at 41% of plan after six months. It also clashes with the stratum “prior biologic exposure (yes or no)” — every participant would be “yes”. The finding quotes the protocol line, the VLT-204 amendment, standard NWB-STD-014 §3.2 and ICH E8(R1) §3.3, and proposes the redline.
Suggested change: “conventional therapy (corticosteroids or immunomodulators) or at least one biologic class”.
Maya opens the precedent: VLT-204 Amendment 2, March 2024, $162K direct cost and 4.1 months. Inclusion 3 was broadened after 58% of screen failures were traced to it, and investigators at 31 of 52 sites reported eligible patients being referred elsewhere. 64 participants re-consented; 52 ethics committee approvals followed.
“The original criterion was not required by the target label and was not supported by the Phase 1b data.”
She accepts. The new wording goes into the draft as a tracked change, and the risk meter moves from 73% to 67% — expected cost of avoidable amendments from $656K to $556K, expected delay from 3.7 to 3.1 months. The decision, her name and the time go straight onto the record.
“Change applied to the draft · amendment risk 73% → 67%”
Maya accepts a 35-day screening window (the median wait for a colonoscopy slot in VLT-204 was 19 days, plus 6 days of central reading) and a washout of 5 half-lives or 12 weeks instead of 12 months. Risk is now 54%. The primary estimand says intercurrent events “will be handled as described in the statistical analysis plan”; ICH E9(R1) expects each strategy in the protocol, and CRV-301 was amended for exactly this — $610K and 4 months. She raises it for the committee.
The co-primary ileocolonoscopy is at Week 12 in the synopsis and §8.2, but Table 2 places it at Week 10 — and sites follow the table. NXL-115 had the same mismatch and recorded 18 deviations before its amendment. Five more gaps sit beside it: calprotectin at Week 8, ECG at Week 12, PK at Week 2, antibodies at Week 24 and the annual TB test. The Week 4 visit runs 6.5 hours against a 4-hour standard. One click fixes all six in Table 2.
“38.0 hours in total · burden index 1.38× the Crohn’s study median”
Generate pack collects the decisions and reasons, draws the risk profile by section, writes the summary and the questions for the committee, and checks that every citation resolves. The pack lists the five accepted changes with their precedents, two questions for the committee — the estimand, and the placebo endoscopic response assumption Priya Raman raised — and every finding still open.
“Committee pack ready · 5 sections, every statement linked”
Send to committee only opens once every high finding has a decision. Maya signs 7 dispositions for VLT-305 draft v0.9 with her password, choosing the meaning of her signature: “I confirm these dispositions as protocol lead”.
VLT-305 moves to Committee for Oct 14. The decisions log shows every finding, its likelihood, who decided it, when and why — the record the committee and inspectors see. Beneath it, the audit trail lists each step by people and agents, from the eligibility reviewer’s first citation to Maya’s signature. Medium findings still open stay listed for their owners.
“Every finding, who decided it, when and why — the record the committee and inspectors see.”
The same draft, seen by the people who own its findings and the chair who receives it — what review looked like, and what it looks like now.
Six reviewer agents read every section and rank the risks; protocol owners decide each finding and the lead signs.
Reads inclusion and exclusion criteria, washouts and population definitions, and compares each with the target label, the eligibility design standard and every past amendment that changed a similar criterion.
Totals participant hours and invasive procedures for each visit in the Schedule of Activities, compares burden with the therapy-area median and flags visits over 4 hours.
Checks objectives and estimands against ICH E9(R1), flags intercurrent events without a stated strategy and sample-size assumptions without a benchmark source.
Matches stopping rules, monitoring and reporting to the risks seen in earlier studies of the product and the investigator brochure.
Compares screening windows, sample shipping, vendor turnaround and country plans with site feasibility history.
Compares every endpoint and procedure across the synopsis, the body and Table 2, merges duplicates, estimates each finding’s likelihood of forcing an amendment, ranks them, assigns owners and writes the committee pack.
Decide each finding; the lead signs. The agents propose; a named person decides.
Ask about any draft, finding or past amendment — or say what to change. Every answer cites the protocol line, the amendment or the standard.
Has this biologic criterion forced an amendment before?
Yes. VLT-204 Amendment 2 (Mar 2024) broadened the same criterion — failure of at least two biologic classes — to conventional therapy or one biologic. Enrolment was 41% of plan at month 6, and 58% of screen failures came from that criterion. It cost $162K directly and 4.1 months. VLT-305 inclusion 3 repeats the wording; the eligibility reviewer estimates a 22% chance it forces an amendment.
What is wrong with the VLT-305 estimand?
Section 3.2 says intercurrent events “will be handled as described in the statistical analysis plan”. ICH E9(R1) expects the strategy for each intercurrent event to be stated in the protocol. CRV-301 did the same and the agency asked for the strategies after first participant in — Amendment 2, $610K and 4 months. Suggested: a composite strategy for rescue medication or stopping for lack of efficacy, and treatment policy for other reasons. Owner: Priya Raman.
Do the synopsis and the schedule agree?
Not yet. The lead reviewer found 6 gaps between the body text and Table 2. The one that matters most: the co-primary endoscopy is at Week 12 in the synopsis and §8.2, but Table 2 places it at Week 10. The others are missing time points: calprotectin at Week 8, ECG at Week 12, PK at Week 2, antibodies at Week 24 and the annual TB test.
Add a rule: flag screening windows under 35 days when an endoscopy is needed
Done. A review rule for the Feasibility reviewer: flag any screening window under 35 days when screening needs an endoscopy with central reading. It is based on VLT-204, where the median wait for a colonoscopy slot was 19 days plus 6 days of central reading. It applies from the next crew run, already matches VLT-305, and you can switch it off in Settings.
13 screens from the working solution, on its sample data. Pick one to see it large.
How the crew works and who decides are settings, changed by the protocol team — not a project.
| Setting | Default | Choose from |
|---|---|---|
| Minimum likelihood of an amendment | 3% | 3 · 5 · 10 % — below it, a finding goes to a watch list |
| Block “Send to committee” while high findings are open | On | On · Off |
| Re-run the crew on every new draft version | On | On · Off — decisions carry over when the line is unchanged |
| Reviewers on the crew | All six live | Pause any reviewer |
| Who decides eligibility and schedule findings | Dr. Maya Chen | Dr. Maya Chen · Priya Raman · Dr. Omar Haddad · Lena Ortiz · Dana Okafor — the protocol lead can reassign |
| Who decides endpoint and estimand findings | Priya Raman | Dr. Maya Chen · Priya Raman · Dr. Omar Haddad · Lena Ortiz · Dana Okafor |
| Who decides safety monitoring findings | Dr. Omar Haddad | Dr. Maya Chen · Priya Raman · Dr. Omar Haddad · Lena Ortiz · Dana Okafor |
| Review rules | e.g. “Flag visits longer than 4 hours” | Add 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 protocol data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (Tufts CSDD protocol amendment studies, 2016 and 2024 — a Phase 3 amendment has a median direct cost of $535K and adds about 3 months). ICH E6(R3), E8(R1) and E9(R1) are named as the guidelines the reviewers cite. People and studies named on this page are characters and examples in the working solution.
A review of a draft clinical protocol that looks for the design choices most likely to force a substantial amendment later — eligibility criteria, visit burden, estimands, safety monitoring, operational feasibility and internal consistency. This solution does it with six reviewer agents that cite the sponsor’s own past amendments for every flag, and gives each finding a likelihood.
The lead reviewer estimates each finding’s likelihood of forcing a substantial amendment from the amendment archive — 148 summaries of changes in the working solution — and combines the open findings into one amendment risk for the draft. Expected cost and delay add up each open finding’s likelihood times the median direct cost of an amendment in that phase ($535K for Phase 3) and about 3 months of extra study time.
ICH E6(R3) for quality by design and feasibility, ICH E8(R1) for critical-to-quality factors and ICH E9(R1) for estimands, plus your own design standards. Drafts are split into ICH M11 sections.
No. The reviewers propose changes and never edit the controlled document. An owner accepts a change, and only then is it applied to the draft as a tracked change, under their name.
Yes. A named owner decides every finding — accept, reject with a written reason, or raise for the committee. The lead reviewer ranks but never decides, high findings must be decided before the draft can go to the committee, and the protocol lead signs the disposition log.
Yes. The lead reviewer compares every endpoint and procedure across the synopsis, the body text and Table 2, and shows each mismatch side by side — for example an endoscopy at Week 12 in the body and Week 10 in the table — with a one-click fix in Table 2.
Yes, in plain words — for example “Flag visits longer than 4 hours” or “Flag primary estimands without an intercurrent-event strategy”. The likelihood threshold for raising a finding, who decides each area and whether the committee is blocked while high findings are open are settings too.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your design standards, SOPs and past summaries of changes — and it goes live once every quality gate has passed. We will run it on one of your own drafts 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 review crew on one of your drafts, against your own past amendments.