MLR review of promotional materials, with every claim checked against its reference page and the label before a reviewer opens the piece
Every claim checked against its reference page and the label before a reviewer opens it.
See one case, screen by screen ↓A draft lands from the agency with a reference pack and an annotation on every claim. Before the medical, regulatory and legal reviewers can judge anything, someone has to open each cited publication at the anchored page, check the number, the population, the endpoint and the timepoint, compare the claim with the Prescribing Information, and read the safety information word for word against the current label.
Most of that is checking, not judgement — and it is where pieces stall. A wrong anchor, an exploratory endpoint dressed as a claim or an ISI from the superseded label goes back to the agency for another round. Committee time goes on pieces that only needed a signature. And when the label changes, nobody can say quickly which approved, live pieces still carry the old safety information.
Estimated split for a typical branded HCP piece in working days, by hand and with the solution.
Six specialist agents extract, verify and check every claim, then route the piece; medical, regulatory and legal reviewers decide.
P-2417, a Veltrimab email to US gastroenterologists, is on Thursday’s committee agenda with ten findings against it. Here is what happens to it, screen by screen, in the working solution.
Maya opens the solution to 10 pieces in review across Veltrimab, Corventa and Talmirex: 6 waiting for her signature, 18 findings to decide, and the Thursday committee at 10:00. The routing wall shows where the risk router placed every piece — fast track, standard or full committee.
“Findings to decide: 18 — every one cites its page or label section.”
The agency’s Talmirex HCP email, P-2437, comes in with two references attached. One press of Run pre-review: the extractor finds 6 claims, 6 anchors and 3 statistics; the reference verifier opens TLX-115 and sees the month-1 result is a secondary endpoint; the router scores it 54 and sends it to standard review, three reviewers.
“Six agents check every claim before a reviewer opens the piece.”
P-2417, “Remission at week 12 — and at one year”, version 2, risk score 72, full committee. Every claim in the email is marked, and the side panel lists 10 findings — 4 high, 2 medium, 4 low — with 2 claims fully supported. The first one: the week-2 rectal-bleeding line rests on an exploratory endpoint.
Maya opens the evidence. VLT-301, page 7, Exploratory endpoints: no rectal bleeding at week 2 in 22% versus 14% — nominal p=0.03, not controlled for multiplicity. The label describes no timepoint before week 12. The verdict is “Not supported as a claim”, at 96% confidence.
“This endpoint was exploratory and not controlled for multiplicity; no conclusions on early onset should be drawn.”
The Claims tab is the claim-to-reference verification table: each claim, the reference and page it rests on, the verdict and the confidence. “2 out of 3 patients stay in remission at 1 year” is overstated — it comes from a descriptive subgroup of 141 patients. “Similar to placebo” is misleading. The ISI is outdated. “#1 prescribed biologic” is broader than the data on file.
A balance reading of 41 against the team’s line of 70. Risk carries 14% of the words, is set in 8 pt against a 22 pt headline at 45% contrast, and sits in the footer below the sign-off. The safety text is from label rev 03/2025, and only the Prescribing Information link is there — no Medication Guide.
“A reviewer makes the call — the reading is evidence, not a verdict.”
For the week-2 line Maya uses approved claim CL-031 — clinical response at week 12, 58% vs 33%. She reads the remaining nine suggestions, including the current safety information with 5.4 Hepatotoxicity, and applies them to draft v3. Then she signs the medical review with her password and the meaning of her signature.
“I reviewed this piece for medical accuracy and confirm each claim is supported by the cited evidence and consistent with the current label.”
Omar signs the regulatory review with label rev 09/2026 confirmed; Lena signs the legal review of comparative wording and trademarks. Only then does the 2253 packager build the package: Form FDA 2253 fields completed, the final piece clean and annotated, the current Prescribing Information and 7 annotated references.
The package carries the application, material type, material ID VLT-US-2417 v3 10/2026, the date of initial dissemination, the current labeling enclosed and all three signature dates. Omar downloads it or schedules the filing for first use.
Veltrimab rev 09/2026 added Warning 5.4 Hepatotoxicity on Sep 18 and retired claim CL-019. The label checker re-reads all 7 approved, live pieces against it and proposes revising five and withdrawing two — a speaker deck that also says “similar to placebo” and a leave-behind with the retired claim. Dana confirms; the requests go to the agency.
The same piece, seen by the five people who carry it — what their week looked like, and what it looks like now.
Six specialist agents extract, verify and check every claim, then route the piece; medical, regulatory and legal reviewers decide.
Finds every claim, statistic and reference anchor in a piece — email, PDF, deck or TV script — and classifies it as efficacy, safety, dosing, comparative or market.
Opens each cited reference at the anchored page, finds the supporting passage and checks the value, comparator, population, endpoint type and timepoint; re-anchors a wrong page.
Compares each claim with the current Prescribing Information, and the safety information word for word with the current label version, naming any missing warning.
Measures how prominent risk is next to benefit — share of words, type size, contrast, placement and, for audio, words per second — and checks required statements and links for the channel.
Checks approved phrasing, trademarks, absolute words, comparative wording and channel rules.
Scores each piece and routes it to fast track, standard review or full committee, then builds the Form FDA 2253 package on approval.
Medical, regulatory and legal sign. The agents propose; a named person decides.
Reviewers and coordinators ask in plain words — or tell it what to change. Answers point to the page they rest on.
What is blocking P-2417?
10 findings are open, 4 of them high: the week-2 relief rests on an exploratory endpoint; “2 out of 3” comes from a descriptive subgroup; “similar to placebo” minimises the safety profile; the safety information is from the superseded label. Plus medium and low findings on wording, links and the ® mark. Each high finding has an approved claim or the label wording ready.
Which live pieces still carry the old Veltrimab safety information?
Seven approved, live pieces still carry the rev 03/2025 safety information, which has no hepatotoxicity warning — five to revise and two to withdraw. Two also use retired wording. The sweep is on your desk: one click re-checks them and sends the requests.
Flag “well tolerated” unless the adverse-event rates are beside it
Added. The rule “well tolerated” needs the rates beside it is now a check in Settings, and the fair balance agent applied it to the pieces in review: P-2402 slide 22 says “well tolerated” with no rates — discontinuations were 4.2% vs 3.9%. It is flagged for Thursday.
Prepare Thursday’s committee agenda
Thursday Oct 8, 10:00–11:00 — four pieces, only the open judgement calls: P-2417 at 10:00, P-2421 at 10:15, P-2402 at 10:35, P-2419 at 10:50. The agenda pack carries each piece with its findings and evidence; the other pieces in review are decided without the meeting.
12 screens from the working solution, on its sample data. Pick one to see it large.
Routing, signers and the checks the agents run are settings your MLR lead owns.
| Setting | Default | Choose from |
|---|---|---|
| Fast track below | Score 30 | Score 20 · 30 · 40 |
| Full committee from | Score 60 | Score 50 · 60 · 70 |
| Always full committee | On — broadcast and first pieces of a launch | On · Off |
| Fair balance threshold, email and digital | 70 | 60 · 70 · 80 |
| Who signs, per brand | Veltrimab: Dr. Maya Chen, Omar Haddad, Lena Ortiz | Medical: Dr. Maya Chen or Dr. Sana Iqbal · regulatory: Omar Haddad or Dana Okafor · legal: Lena Ortiz or Ben Adler |
| Checks the agents run | 7 of 8 on | Each check on or off — accessibility is off by default |
| Your own rules | Add in a sentence | Turned into a check for your approval |
| Prescribing Information | Current versions, checked nightly | From your regulatory label store |
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 pieces · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published MLR cycle-time benchmarks (≈ 29 business days). Regulatory context: 21 CFR 202.1 and Form FDA 2253 at first use. People named on this page are characters in the working solution; its brands, studies and companies are fictional.
MLR pre-review checks a promotional piece before the medical, legal and regulatory reviewers open it. The Promotional Material Pre-Review Agent extracts every claim and its reference anchor, checks the cited page supports it, compares it with the current label, measures fair balance and checks the safety information and required statements — then annotates the piece so reviewers start with the evidence.
The reference verifier opens each cited reference at the anchored page, finds the supporting passage and checks the value, comparator, population, endpoint type and timepoint. Verdicts are supported, overstated, not supported or wrong anchor, each with the page and passage. Exploratory and subgroup results are never marked supported as efficacy claims.
Yes. The fair balance checker measures share of words, type size, contrast and placement — and words per second for audio — against your threshold for the channel. The label checker compares the safety information word for word with the current label version and names any missing warning.
No. Agents propose; reviewers decide. Every finding is accepted, sent to the agency or dismissed with a reason, and named reviewers sign every piece — one on fast track, three on standard and committee — with the meaning of each signature recorded. Every step is on the piece’s record.
When a new label version is approved, the label checker re-checks every approved, live piece against it and proposes revise or withdraw for each. A person confirms, and the change requests go to the agency with the new text.
It builds the package once every required signature is in: the completed form fields, the final piece clean and annotated, the current Prescribing Information and the annotated references. Your regulatory reviewer downloads it or schedules the filing for first use.
Yes. The fast-track and full-committee score thresholds, which pieces always go to committee, the fair balance threshold, who signs per brand and which checks run are all settings. You can add a rule by describing it in a sentence, and it becomes a check after your approval.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your reference library, label versions, claims library, brand guides and routing rules — and it goes live once every quality gate has passed. We will walk you through it on your own past pieces first.
Every question from a healthcare professional answered in a signed letter where each sentence links to its source.
1-min filmInvestigator Study ReviewEvery investigator's study concept checked against our priorities, ongoing studies and its budget, and decided in half the time.
1-min filmMedical InsightsEvery field insight becomes a cited theme within days, ready for the quarterly report.
1-min filmWe’ll run Promo Pre-Review on a sample of your own past pieces and their references.