Medical Monitor Query Response: every site question answered, cited and consistent, in hours
Every site question answered, cited and consistent, in hours
See one case, screen by screen ↓A medical monitoring solution that answers site eligibility and dosing queries. Seven agents read each question, find the passages in the current protocol version, compare the case with every earlier signed answer and draft a cited reply; the medical monitor decides and signs, and the reply is sent, filed in the TMF and published to the study FAQ.
How to choose AI solutions for life sciences →
Updated
Investigators send the medical monitor a steady stream of questions: is this patient eligible, how long is the washout, do we hold the dose, what about this missed visit. They arrive through the site portal, by email and by phone, and are answered from memory or a search of the protocol — by two monitors and a CRO medical desk — and filed, sometimes, in a spreadsheet clarification log.
Nothing checks that a new answer agrees with an old one. When the protocol is amended, earlier answers are not re-checked, and sites keep quoting them to each other. A site told the wrong washout enrols the wrong patient, and a wrong eligibility answer becomes a possible important protocol deviation for every participant enrolled on it.
Estimated split for one eligibility question on a large Phase 3 study, answered by email and with the solution.
Seven agents read, find the protocol text, compare every earlier answer and draft the reply; the medical monitor decides and signs.
VLT-302 is a Phase 3 study of Veltrimab in plaque psoriasis at 142 sites, on protocol v3.0 since Amendment 2 of 12 May 2026. On Wednesday morning, Site 226 in Gothenburg asks about a biologic washout. Here is what happens next, screen by screen, in the working solution.
Maya's query inbox shows 7 open questions across VLT-302 and NXL-115, each with its site, its criterion and its clock: 1 h 48 m left on a Grade 3 neutropenia dose-hold question from Seoul, 23 h 06 m on an eligibility question from Gothenburg. Questions that differ from an earlier answer are sorted to the top and have their own tab.
“6 drafts ready · 2 differ from an earlier answer”
Participant 226-031 last received an IL-17A biologic on 4 Aug 2026, stopped for loss of response. Screening started on 1 Oct, day 7 of 28, and every other criterion is met. The Inquiry Reader highlights the participant, the last dose, the planned Day 1 — 10 weeks 2 days after the last dose — and the line that matters most.
“A colleague at another site understood that 8 weeks is enough for IL-17 agents — can we go ahead?”
The draft quotes VLT-302 Protocol v3.0, §5.2 exclusion criterion 7, p. 46, changed by Amendment 2. Day 1 cannot be on 15 Oct; the earliest is 27 Oct 2026. A Day 1 on 27 or 28 Oct keeps the participant inside the 28-day screening period, so no re-screening is needed. Every sentence that states a rule carries a numbered citation.
“For IL-17 and IL-23 biologics the washout is 12 weeks.” — Protocol v3.0 §5.2 E-7
The Consistency Checker finds MI-0384: on 15 Sep the CRO medical desk told Site 114 in Boston that the washout is 8 weeks, quoting protocol v2.0, superseded on 12 May. Both answers sit side by side, the 8 weeks struck through against 12. Participant 114-019 was randomised on 18 Sep, 8 weeks and 3 days after the last IL-17 dose.
“Participant 114-019 was randomised on 18 Sep … — a possible important eligibility deviation.” A conflict always stops the draft until a person looks.
Maya drafts the correction to Dr. James Whitaker at Site 114, copied to the site CRA and Sam Patel. It says the 8-week figure came from protocol v2.0, gives the 12-week rule, and asks the site to record 114-019 with its CRA as a possible eligibility deviation. She adds it to her signature with MI-0422 and flags 114-019 to the CRA.
“We apologise for the error.”
The signing screen shows the decision — not on 15 Oct, earliest Day 1 is 27 Oct — with the protocol version it rests on. Maya signs as medical monitor, with the meaning “Approved as medical monitor response” and her password. The correction to Site 114 goes out under the same signature, the decision is filed in the TMF, and she chooses to publish it to the study FAQ without participant numbers.
Only a physician with the medical monitor role can sign a medical decision.
An hour after the question arrived, the reply is signed and sent to Site 226, the correction is sent to Site 114, 114-019 is flagged to the CRA, and the decision is filed in the TMF and added to the study FAQ. The decision log now holds 1,986 signed decisions, each with its signer, time and protocol version.
Nothing is sent before a signature.
Dr. Henrik Lund at Site 207 Aarhus asks about participant 207-019: hepatitis C antibody positive, HCV RNA undetectable after curative treatment in March 2023. The agents classify it under E-12, start the 24-hour clock, quote §5.2 E-12(c), and draft: exclusion criterion 12 does not exclude the participant. No earlier answer covers this part of the criterion.
“No earlier answer on E-12 — once signed, this one becomes the reference.”
Site 114 asks for an exception: participant 114-022 has ALT 107 U/L, 2.6 × ULN, against a limit of 2 × ULN, and is anti-HBc positive. The full review shows the decision by criterion — excluded on E-11, allowed with conditions on E-12, re-screening allowed under §8.1.3 — with the Investigator's Brochure passage on hepatic findings among the sources. Proposed: not eligible at this screening, may re-screen from 4 Nov 2026.
“The protocol does not allow exceptions to eligibility criteria, so the reply says so plainly and offers the re-screening route.”
Protocol hot spots map 106 questions in 30 days onto VLT-302's criteria. Tuberculosis screening (E-9) reached 14 questions and liver tests (E-11) reached 11, so a clarification letter is suggested for each. An answer from February, MI-0279, is flagged: Amendment 2 now allows anti-HBc positive with undetectable HBV DNA, and the old answer is still in the FAQ.
“A clarification letter would answer them all at once.”
The same question, seen by the people who ask it, answer it and live with the answer — what their week looked like, and what it looks like now.
Seven agents read, find the protocol text, compare every earlier answer and draft the reply; the medical monitor decides and signs.
Reads each site question and its attachments; pulls out the participant number, lab values with units and × ULN, dates and medicines, and flags any request for an exception.
Sorts the question by topic and protocol section (I-3, E-11, 6.4 …), sets urgency and starts the right clock: 4 hours for urgent dosing and toxicity, 24 for eligibility, 72 for everything else.
Finds the exact passages in the current protocol, amendments, clarification letters, Investigator's Brochure and manuals, each with document, version, section and page.
Compares the case with every signed earlier answer and FAQ entry on the same criterion and marks each consistent, conflicting or out of date, explaining any difference in one sentence.
Decides each criterion as met, not met, met with conditions or needs information, and writes a plain reply to the investigator; offers the route the protocol allows, such as re-screening or a repeat test.
Checks the draft: no eligibility exceptions, no advice beyond the protocol, and an eCRF reminder when dosing was interrupted for surgery or an illness.
After the signature, sends the reply, writes the decision log with signer, time, meaning and protocol version, files it in the TMF and publishes a de-identified FAQ entry when the signer chose to.
Decides and signs. The agents propose; a named person decides.
12 screens from the working solution, on its sample data. Pick one to see it large.
Clocks, who signs, and the rules the agents follow are settings, and every change goes to the audit trail.
| Setting | Default | Choose from |
|---|---|---|
| Urgent dosing and toxicity clock | 4 h | 1, 2, 4, 6 or 8 h |
| Eligibility clock | 24 h | 8, 12, 24 or 48 h — keeps the participant inside the 28-day screening window |
| Everything else | 72 h | 24, 48, 72 or 120 h — medicines, visits, laboratory and general questions |
| Signer for VLT-302 | Dr. Maya Chen | Dr. Maya Chen · Dr. Anil Mehta |
| Backup when the signer is away | Dr. Anil Mehta, after 2 hours without a reply | Dr. Anil Mehta · Dr. Omar Haddad |
| Suggest a clarification letter when a criterion reaches | 10 questions in 30 days | 5, 6, 8, 10, 12, 15 or 20 questions in 30 days |
| Who can read the FAQ | All sites in the study | All sites in the study · Sites in the same country · Study team only |
| After a protocol amendment | Flag answers that cite changed text and ask Lena Ortiz to confirm | Flag them and ask Lena Ortiz to confirm · Retire them automatically |
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 study data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (Tufts CSDD, 2024: 76% of Phase II/III protocols amended, 3.3 amendments each; context from ICH E6(R3) (FDA, Sept 2025) and FDA draft guidance on protocol deviations (Dec 2024)). People, studies, sites and participants named on this page are fictional — characters and sample data in the working solution.
A medical monitoring solution that answers site eligibility and dosing queries. Seven agents read each question, find the passages in the current protocol version, compare the case with every earlier signed answer and draft a cited reply; the medical monitor decides and signs, and the reply is sent, filed in the TMF and published to the study FAQ.
Before a reply goes out, the Consistency Checker compares it with every signed answer and FAQ entry on the same criterion, including answers given under a superseded protocol version. If another site was told something different, both answers sit side by side with the protocol version each relied on and the participants affected, and one click drafts the correction.
Answers that cite changed text are flagged for the clinical scientist to confirm, or retired automatically if you choose that setting. In the working solution, the check after Amendment 2 found an answer from February on hepatitis B that the amendment had made out of date.
No. The Exception Guard rejects any wording that could be read as an exception or waiver, and the rule cannot be turned off. When a site asks for one, the reply says plainly that the protocol does not allow it and offers the route the protocol does allow, such as re-screening.
On-drug dosing and toxicity questions are always marked urgent with a 4-hour clock, and move to the backup signer after 2 hours without a reply. When dosing was interrupted for surgery or an illness, the reply asks the site to record the interruption and its reason in the eCRF, and that reminder cannot be removed.
Yes. The agents draft; only a physician with the medical monitor role signs a medical decision, with the meaning of the signature, and nothing is sent before that signature. The monitor can edit or shorten the draft, ask the site for more, or reassign the question.
Yes. Signed answers can be published to a study FAQ that sites search through the portal, without participant numbers or dates. You choose who reads it: all sites in the study, sites in the same country or the study team only.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your protocol and amendments, clarification letters, Investigator's Brochure and a sample of past site questions — and it goes live once every quality gate has passed. We will walk you through it on your own protocol first.
A checked first-draft protocol in days, in your approved language.
1-min demoeTMF Quality CheckAn inspection-ready trial master file every day, each site scored on what is due today.
1-min demoClinical Study Report AuthoringResults drafted within days of database lock, linked to tables.
1-min demoWe'll run Medical Monitor Query Response on your protocol, its amendments and a sample of your past site questions.