For medical affairs & VP R&D
Every claim you make has a source behind it.
Standard of care, comparator choice, comparable-trial outcomes, regulatory precedent and safety signals — assembled from public evidence and linked line by line to the record, so the position you take internally is the one you can defend externally.
Every claim traced to the public record it came from.
The situation
The evidence exists. Assembling it is the job.
Nobody in medical affairs is short of literature. What is scarce is a current, complete, citation-linked view of the evidence around a single asset — one that can be re-derived next quarter when someone asks whether it still holds.
- Live public sources, continuously refreshed
- ~50
- Analyses, each cited to its record
- 30+
- The same question, re-derived on today’s evidence
- Re-runnable
What happens today
The evidence base is rebuilt every time someone asks.
A slide deck is assembled for a specific meeting, then goes stale. Six weeks later a different team asks a version of the same question and starts again from the search bar.
Every question restarts the search
The landscape review that answered a question in March is not reusable in June, because nobody recorded which sources it drew on or when they were read.
The sources do not talk to each other
Registry data, literature, label precedent, safety reports and pricing each live in a separate tool with a separate export. Reconciling them is manual, and it is where the errors enter.
Claims are supported, but not traceably
A reference list attached to a deck is not the same as a figure that resolves to the record it came from. Under challenge, the difference matters.
Nothing tells you when it changed
A new comparable readout, a label update or a safety signal changes the position you took. Today you find out when a reviewer happens to notice.
What you get instead
One evidence base, and it stays current.
Standard of care, established from evidence
What is actually used in this indication today, drawn from approved labels, guidelines and trial precedent — the baseline every comparator argument starts from.
The comparable-trial record
Every comparable trial in the indication with its outcome and, where stated, its reason for stopping — including the failures that quietly never reported.
Regulatory precedent
How comparable programs progressed through FDA and EMA review, and what the precedent timeline implies for yours.
Adverse-event signal
Safety signals for the molecule and its class read from public pharmacovigilance data, presented with the caveats that public spontaneous-report data requires.
Citations, not a bibliography
Every figure and statement resolves to the specific record and snapshot date it was read from — so a reviewer can check it rather than take it on trust.
Signed, and flagged when it moves
Positions are approved with a 21 CFR Part 11 e-signature. If the underlying evidence later changes, the signed decision is flagged for re-review — nothing stays approved on stale data.
What runs, and who drafts
The scope for this job
Medical affairs work concentrates in the evidence pillars — the science, the design precedent and the standard-of-care/commercial context. Cohort and operations run too, and become relevant the moment the question turns into a study.
How it runs
Ask once, re-derive whenever.
Frame the asset and indication
Name the molecule or class and the indication, or upload the protocol or synopsis you are working against.
The evidence base assembles itself
Registries, literature, labels, guidelines, safety data, patents and pricing are pulled together into one cited view — with the gaps stated rather than smoothed over.
Take a position, and keep it honest
Your team signs off the position under Part 11. When the underlying evidence changes, the signed decision is flagged so it gets re-reviewed rather than quietly aging.
What lands on your desk
An evidence dossier with the receipts attached.
Structured for a reviewer who intends to check it, not for an audience that will take it on trust.
What lands on your desk
Cited evidence dossier + signed position record
- Standard of care in the indication, with the labels and guidelines it was established from.
- The full comparable-trial record — outcomes, and stated reasons for stopping where they exist.
- Regulatory precedent for comparable programs, with the review timeline it implies.
- Adverse-event signal for the molecule and its class, with the limits of public spontaneous-report data stated.
- Design and endpoint norms for the indication, so a departure from them is a deliberate choice rather than an oversight.
- A hash-chained audit record of who approved what, when, and on which evidence snapshot.
Honest limits
Where this stops
What this does not do for you. Naming it here is cheaper for both of us than finding it in a procurement review.
It is not a promotional-review or MLR system, and it does not clear material for external use.
Adverse-event signals come from public spontaneous-report data, which is subject to reporting bias and cannot establish causation. The output says so wherever it appears.
It reads public evidence only — it does not see your internal safety database, your medical information queries, or unpublished study data.
Where the evidence is thin, the assessment reports the gap. It does not fill it with an inference dressed as a finding.
Drafted positions are drafts for a qualified reviewer. The Part 11 signature records your team’s judgment — it does not substitute for it.
Questions
The ones you would ask first
Is this a literature-search tool?
No. Literature is one of around 50 sources it reads. The output is a structured assessment across six pillars — design norms, comparable outcomes, regulatory precedent, safety signal, standard of care — with each element cited, rather than a list of papers to read.
How do we know a claim is still current?
Every figure carries the record and snapshot date it came from, and runs are reproducible against that snapshot. Re-running the same question on today’s evidence shows you exactly what moved.
Can we use it for regulatory or promotional submissions?
It is decision-support. It assembles and cites the evidence base your regulatory and medical-review processes then work from — it does not replace those processes and is not itself a submission.
What happens when the evidence changes after we sign off?
The signed decision is flagged for re-review. That is deliberate: a position approved on evidence that has since moved is exactly the thing that should not stay quietly approved.
Build the position on evidence you can show.
Name an asset and an indication. We’ll return a cited evidence dossier — standard of care, comparable outcomes, regulatory precedent and safety signal — every line traceable to its public record.