SystemResearch & Discovery

Competitive Pipeline Watch

Competitive intelligence on every rival trial, readout and approval, in a sourced brief every Monday

Every rival trial, readout and approval lands in a sourced brief for each programme, every Monday.

See one case, screen by screen ↓
demo1,284sourceschecked in one Monday watch run, for 42 rival assets — 23 changes, 5 that matter
demo22minmedian analyst review per brief, over the last 12 weeks
target3daysfrom a rival’s move to the programme team knowing — same day for approvals
target100%of claims in every brief linked to their source passage
The problem

Why the competitive landscape is out of date by the time it is presented

A programme team needs to know when a rival’s trial changes phase, a pivotal study reads out or an agency approves a medicine in its indication. Today that knowledge is assembled by hand: an analyst checks trial registries record by record, reads newsrooms and pipeline pages, searches congress abstracts and agency databases, then rebuilds the landscape deck.

The same event turns up three times — in the registry, the press release and the abstract — and has to be merged before anyone can say whether it matters for our programme. By the time the deck reaches the team, the landscape has moved, and the numbers in it are hard to trace back to the page they came from.

demo4daysto refresh a landscape by hand
demo9daysmedian from a rival’s move to the programme team, by hand
Where the days go between a rival’s move and the team knowingestimated
By hand9 days
With the solution3.0 days
  • Waiting for the next sweep of the sources4 → 2.2 d
  • Reading registries, releases, abstracts and agency databases2 → 0.1 d
  • Merging duplicates and deciding what matters1 → 0.1 d
  • Writing the brief or the deck1.5 → 0.1 d
  • Analyst review and sending0.5 → 0.5 d

Estimated split of the 9-day median by hand (2025) and the 3-day target with the solution. Most of what is left is the wait for the Monday run — approvals and pivotal readouts reach the team the same day.

How it works

How a change moves

Six specialist agents watch, compare and write up every change; an analyst approves each brief.

What comes in
Sources in4 public sources · registries, news, congresses, agencies
Agents at work
Registry watcher
News watcher
Abstract reader
Approval watcher
Then
Change detectordedupe + significance
Then
Brief writerone page per team
A person decides
Analystapproves and sends
What comes out
Weekly brief
Landscape matrix
Sent to every team
One case, step by step

One Monday, from watch run to brief sent

Week 40 for the KRAS G12D programme team: a rival medicine is approved in pancreatic cancer, another enters Phase 3 in the indication we plan for, and a lung cancer combination reads out. Here is the morning, screen by screen, in the working solution.

  1. 01Monday morning

    The week’s changes, already found

    Lena Ortiz · Competitive intelligence lead

    Lena opens the solution to a finished watch run: 1,284 sources checked for 42 rival assets across four watch lists — 23 changes, 5 that matter. Two briefs wait for her approval, due to their teams by 12:00, and two changes about sites on the feasibility shortlist wait for her to confirm. Sam Patel has already sent the Claudin 18.2 brief to 11 people at 08:35.

    This morning’s watch run: “Change detector — 29 findings → 23 changes.”

  2. 02Next

    The KRAS landscape, by mechanism and phase

    Lena Ortiz · Competitive intelligence lead

    Oncology — KRAS holds 14 watched assets from 11 sponsors, set out by mechanism — G12C, G12D, pan-RAS, other alleles, pathway partners — and by phase. Our programme NX-4127 sits in Phase 2; two rivals are approved and two are in Phase 3 ahead of it. Assets that changed this week are marked, and the three significant changes are listed beside the matrix.

  3. 03Run watch now

    Four watchers, one change detector, one brief writer

    The agents

    The run is shown step by step. The Registry watcher checks 4 trial registries for the 14 assets; the News watcher reads 38 press releases and pipeline pages; the Abstract reader goes through 212 congress abstracts and matches the late-breaker LBA-71 to CORAL-301; the Approval watcher finds a new US approval for VNT-224. The Change detector merges 14 findings into 11 changes, and the Brief writer drafts the KRAS brief. The last step waits for the analyst.

    “11 changes found · 3 significant. VNT-224 approved in the US · CORAL-301 met its endpoint · SLN-0582 entered Phase 3. Every change is linked to its source.”

  4. 04Approval · Oct 2

    VNT-224 approved — before and after, with the passage

    Approval watcher

    The change feed shows what moved since the last capture on Mon 28 Sep: status from “Under review (priority)” to “Approved — United States”, indication 2L+ metastatic PDAC, KRAS G12-mutant, label test-selected with overall survival 12.4 vs 6.8 months. Beside it, the agency approval letter and Vantor Pharma’s press release, each with the exact sentence highlighted.

    Why it matters for NX-4127: “It sits in the same second-line population as our NX-4127 Phase 2, and becomes the treatment our next trial is likely to be compared with.”

  5. 05To confirm

    SLN-0582 enters Phase 3 — the analyst decides

    Lena Ortiz · Competitive intelligence lead

    SOLSTICE-PANC is registered: SLN-0582 moves from Phase 2 to Phase 3 in first-line metastatic PDAC, NCT07214863, an estimated 540 patients, 210 planned sites in 19 countries. The Change detector explains why it matters — Solenne is about 12–18 months ahead, and 38 of its likely sites are on our feasibility shortlist — and proposes it as significant. Lena confirms it, or marks it not significant.

    Significant changes need an analyst’s confirmation; anything below 85 % confidence goes to review.

  6. 06Drafted at 06:14

    One page for the programme team, every claim linked

    Brief writer

    The KRAS landscape brief for week 40 opens with a three-point bottom line, then what changed — the VNT-224 approval, the SOLSTICE-PANC phase advance, the CORAL-301 readout — each with a “So what” for NX-4127. The sources panel shows 14 of 14 claims sourced, numbers matching the sources, one page. Click any number in the brief to open its passage.

  7. 07What it means

    Actions with an owner, and what to watch next

    Brief writer

    The brief turns the week into three actions: review the second-line Phase 3 design with VNT-224 as the likely comparator before the 2 Nov governance meeting (Priya Raman, clinical development); check the 38 shared sites before site selection closes on 30 Oct (Jonas Weber, clinical operations); update the target product profile with the 12.4-month benchmark (Dr. Maya Chen). Eight smaller changes follow under “Also this week”.

    The Brief writer’s rules: every claim needs a source passage; no forecasts stated as fact.

  8. 08Before it goes

    Approve and send to the KRAS G12D programme team

    Lena Ortiz · Competitive intelligence lead

    The approval screen shows 14 of 14 claims linked, the 14 readers, from the programme lead to regulatory and market access, and how it goes out — email with the brief as a PDF, and a post to the programme page. One change, PLH-12, is still unconfirmed; the screen says so, and that approving the brief confirms it as written.

    “Your approval is recorded with your name and the time, and the sent version is kept.”

  9. 09Sent

    Sent, with the trail behind it

    Lena Ortiz · Competitive intelligence lead

    The brief moves to Sent and the activity trail reads in order: the Change detector found 11 changes, the Brief writer drafted the brief with 14 claims, Lena confirmed 9 of 11 changes, confirmed PLH-12 with the brief, and approved and sent it to 14 people.

  10. 10Last 12 weeks

    How the watch is doing

    Lena Ortiz · Competitive intelligence lead

    280 changes detected across 42 watched assets, 45 significant — each confirmed by an analyst — and 46 briefs sent to 4 programme teams. Event to programme team: a median of 3.2 days, against 9 by hand. By watch list, briefs on time and analyst edits per brief; by source, where the changes came from.

Who it’s for

Built for the analysts who watch, and the teams who act.

The same Monday, seen by the people who write, approve and read the brief — what their week looked like, and what it looks like now.

LO
Lena OrtizCompetitive intelligence lead
Approver
Before
Spends the first days of every week sweeping registries and newsrooms before she can say what changed.
Now
Starts Monday from a finished run: confirms the significant changes, reads the brief against its sources, approves and sends.
SP
Sam PatelCompetitive intelligence analyst
Approver
Before
Rebuilds the landscape deck for each team by hand, chasing the same event across three sources.
Now
Approves the briefs for his watch lists — Claudin 18.2 and Lp(a) were sent before 08:40.
MC
Dr. Maya ChenProgramme lead — KRAS G12D
Reader
Before
Learns about a rival’s move days later, in a deck she cannot trace to the source.
Now
Reads one page every Monday, with every claim one click from its passage and the actions already owned.
PR
Priya RamanClinical development
Reader
Before
Hears about a new comparator when the trial design is already under review.
Now
Gets the action in the brief: review the second-line Phase 3 design before the 2 Nov governance meeting.
Built on the engine

6 agents. Each with one job, and hard limits.

Six specialist agents watch, compare and write up every change; an analyst approves each brief.

Registry watcher

Checks trial registries every week for each watched asset and records what changed in each trial record — status, phase, enrolment, dates, arms, sites — with the exact text before and after.

  • Public sources only
  • Never infers a status the record does not state
News watcher

Reads company press releases and pipeline pages and captures pipeline moves, filings and topline results for watched sponsors.

  • Public sources only
  • Quotes the exact sentence for every claim
Abstract reader

Reads congress abstracts and pulls design, population, endpoints (PFS, OS, ORR, DoR) with confidence intervals and grade ≥ 3 adverse events into a comparable table.

  • Numbers copied, never computed
  • Flags interim versus final data
Approval watcher

Checks agency approval databases in the US, EU and Japan for approvals, filing acceptances and label changes.

  • Public sources only
  • Indication copied word for word
Change detector

Merges what the watchers found — one real-world event is one change, even when seen in three sources — scores significance with the watch list’s rules and explains in two sentences why it matters to our programme.

  • Significant changes need analyst confirmation
  • Confidence below 85 % goes to review
Brief writer

Writes the one-page weekly brief for each programme team — bottom line, what changed, what it means with a suggested owner, also this week, watch next — with every claim linked to its source passage.

  • Every claim needs a source passage
  • No forecasts stated as fact
  • An analyst approves before anything is sent
Analyst

Approves before anything is sent. The agents propose; a named person decides.

Ask in plain words

Ask about any rival, in plain words

Ask about any rival asset, change or brief — or tell it what to watch. Every answer links to its sources.

What changed for KRAS this week?

This morning’s run found 11 changes across the 14 KRAS assets — 3 matter for NX-4127: the VNT-224 approval in second-line pancreatic cancer, SLN-0582 as the first G12D-selective medicine in Phase 3 in first-line PDAC, and the KST-907 Phase 3 readout in first-line lung cancer (HR 0.64). The other 8 are registry and pipeline updates, all in this week’s brief.

How does the VNT-224 approval affect NX-4127?

Three things change. Comparator: a second-line Phase 3 would likely be compared with VNT-224, not chemotherapy. Benchmark: median OS to beat in second line is now 12.4 months (HR 0.46 vs chemotherapy). Timing: first line is still open — VNT-224 is approved only after prior therapy, and SLN-0582’s first-line trial reads out around mid-2029.

Which of our sites overlap with competitor trials?

41 sites on the NX-4127 feasibility shortlist also appear in competitor KRAS trials: 38 in SOLSTICE-PANC (SLN-0582) and 3 in PELL-12-101 (PLH-12). SOLSTICE-PANC is the one to watch — it is not yet recruiting, so our site selection, closing 30 Oct, can still move first.

Alert me the same day when a rival KRAS trial is stopped

Done. A new rule under “What counts as significant”: a watched trial is suspended, terminated or withdrawn — with a same-day alert to you and Sam Patel, outside the weekly brief. It would have fired once in the last 12 weeks, when MRL-209 was suspended on 23 Sep. The change is saved as a new version of the watch settings and is on the audit trail.

Every screen

The working solution, as it ships.

12 screens from the working solution, on its sample data. Pick one to see it large.

HomeMonday’s watch run in numbers, the briefs and changes waiting for the analyst, and what was significant this week.
The landscapeEvery watched asset by mechanism and phase, our programme marked, this week’s changes flagged.
The watch runFour watchers, the change detector and the brief writer, step by step — then it waits for the analyst.
A change, before and afterThe fields that moved since the last capture, why it matters for our programme, and the source passages.
Confirm what is significantThe change detector proposes; the analyst confirms or marks it not significant.
The weekly briefOne page per programme team: bottom line, what changed and so what — every claim linked to its source.
What it means for usActions with a named owner, the smaller changes of the week, and what to watch next.
Approve and sendClaims checked, readers listed, unconfirmed changes called out — then sent by email and to the programme page.
The activity trailEvery agent step and every human decision on the brief, in order, with the time.
Every briefEach team’s brief by week, with significant changes, claims sourced, stage and who sent it.
The dashboardChanges per week, event-to-team time against the by-hand baseline, sources and change types.
Your rulesWatch lists and their approvers, the watch run schedule, brief delivery, same-day alerts and what counts as significant.
Governance

Built to be trusted: public, sourced, approved.

Public sources onlyThe agents read public sources only — trial registries, company newsrooms and pipeline pages, congress abstracts and agency approval databases — and compare them with your own watch lists and feasibility shortlist.
Every claim opens its passageEach statement in a brief links to the exact passage it came from. Numbers are copied from the source, never computed.
An analyst confirms what is significantThe Change detector proposes; a named analyst confirms or marks a change not significant. Anything below 85 % confidence goes to review.
Nothing is sent before approvalA brief goes to a programme team only after a named analyst approves it. The approval is recorded with the name and the time, and the sent version is kept.
What the page said that dayEvery version of every source is kept, so the record shows what a registry entry or pipeline page said on the day it was captured.
Every step on the recordAgent steps, confirmations, approvals and changes to the settings are on the audit trail. Run history and briefs are kept for 7 years.
Configuration

Your watch, your rules

What to watch, when, who approves and what counts as significant are settings — each change saved as a new version and recorded in the audit trail.

SettingDefaultChoose from
Watch runEvery Monday, 06:00 CETEvery Monday, 06:00 CET · Every Monday and Thursday, 06:00 CET · Every day, 06:00 CET
Brief deliveryAs soon as it is approvedAs soon as it is approved · Monday 12:00, if approved
Same-day alerts for approvals and pivotal readoutsOnOn · Off
Approver for each watch listLena Ortiz (KRAS, TL1A) · Sam Patel (Claudin 18.2, Lp(a))Lena Ortiz · Sam Patel
Significant: approval, filing acceptance or label change; pivotal readout or topline result; phase advanceOnOn · Off, per rule
Significant: new trial in our indication; sites added in our shortlist countriesOnOn · Off, per rule
Significant: trial stopped, suspended or terminated; primary completion moves more than 3 monthsOff — a stopped trial is reported in the weekly briefOn · Off, per rule
Watch lists4 lists · 42 assetsAdd a watch list or an asset
Connections

Works with the sources your analysts already read

Trial registries4 registries — US, EU, Japan, China
Press releases & pipeline pages126 companies, each page version kept
Congress abstracts9 oncology, immunology and cardiology congresses
Agency approval databasesUS, EU and Japan
Your feasibility shortlistsites compared with rival trials
Email and the programme pagethe approved brief, as a PDF and a post
What it changes

The difference, in numbers.

Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.

target
36min
to refresh the whole landscape, review included
By hand≈ 4 days
With agents≈ 36 min
target
3days
from a rival’s move to the team knowing, same day for approvals
Told on day 3
median · was 9 days by hand
target
100%
of claims in every brief linked to their source
every claim opens its source passage

“demo” = seen in the working solution, on its sample landscape · “target” = the design goal, measured in the live solution · “estimated” = our estimate · by-hand figures are the 2025 analyst workload baseline shown in the solution. People, companies, assets and trials named on this page are characters and sample data in the working solution.

Questions

What competitive intelligence teams ask us.

What is Competitive Pipeline Watch?

A competitive intelligence solution for drug development programmes. Six agents watch trial registries, company press releases and pipeline pages, congress abstracts and agency approval databases for every rival asset on your watch lists, and write a one-page brief for each programme team every Monday. An analyst approves each brief before it is sent.

How does it decide what is significant?

The Change detector merges what the watchers found into one change per real-world event, then applies your watch list’s rules — approvals and label changes, pivotal readouts, phase advances, new trials in your indication, sites in your shortlist countries — and explains why it matters for your programme. An analyst confirms each significant change; anything below 85 % confidence goes to review.

Which sources does it read?

Public sources only: 4 trial registries (US, EU, Japan, China), press releases and pipeline pages of 126 companies, abstracts from 9 oncology, immunology and cardiology congresses, and agency approval databases in the US, EU and Japan. Each source version is kept, so the record shows what a page said on the day it was captured.

Can we trace every number in a brief?

Yes. Every claim links to the exact passage it came from — click a number in the brief to open the registry record, press release, abstract or approval letter at that passage. Numbers are copied from the source, never computed.

Does anything reach the programme team without review?

No. A brief is sent only after a named analyst approves it, and the approval is recorded with the name and the time. Same-day alerts for approvals and pivotal readouts are confirmed by the analyst too.

Can we set our own watch lists, schedule and rules?

Yes. Watch lists, assets, approvers, the watch run schedule, brief delivery, same-day alerts and what counts as significant are all settings. You can also ask in plain words — “alert me the same day when a rival trial is stopped” — and every change is saved as a new version on the audit trail.

How long does it take to go live?

The Agentic Solution Engine builds and deploys it from your requirements — your programmes, the rival assets you watch, your brief template and your significance rules — and it goes live once every quality gate has passed. We will walk you through it on one of your own landscapes first.

See it on
your rivals.

We’ll run Competitive Pipeline Watch on the rivals of one of your own programmes.