Freedom-to-operate patent analysis that charts every claim element by element — while the product can still change
Blocking claims found early, while a buffer change costs a study, not a launch.
See one case, screen by screen ↓It answers whether someone else's live patent claim covers your product in the markets where you will launch it. Every patent and application near the product is found, each independent claim is split into its elements, and each element is checked against the product — a claim reads only if every element does. Freedom-to-Operate Patent Analysis does that reading at scale and puts the result in the form counsel already uses, the claim chart.
How to choose AI solutions for life sciences →
Updated
A biologic does not sit under one patent. The molecule, its epitope, the formulation, the dosing regimen, the device and the manufacturing process each sit under different families, granted in different countries and expiring in different years. A freedom-to-operate analysis means finding every live patent and application near the product, reading each independent claim, splitting it into its elements and checking each one against the product. A claim reads only if every element does. By hand, an IP analyst spends weeks reading claims that were never close, and the answer changes every time a family grants or CMC changes the product.
The risk is the claim nobody charted. In the working solution's case, the last manual update read the US member of a formulation family and moved on: its claim 1 had been narrowed to IgG4 in prosecution and did not read. The European claim, granted in February 2024, was never narrowed — and it likely reads on the product's histidine formulation. Found before the formulation locks, a buffer change costs a development batch already on stability and one comparability study. Found after, it means new stability, comparability and possibly a bridging study — and a blocking claim found late can stall a launch.
Estimated split for one FTO update on a biologic in development, by hand and with the solution.
Five specialist agents search, read, match, check and rank every claim; patent counsel concludes and signs.
Halden Biotherapeutics is taking HLD-217 — a humanised IgG1 against TL1A, 150 mg/mL subcutaneous, for ulcerative colitis — into Phase 3. The formulation locks on 15 December 2026. On Tuesday 13 October, FTO update 3 runs. Here is what happens next, screen by screen, in the working solution.
Jonah's scope reads: anything that could block HLD-217 in its launch markets — TL1A antibodies, high-concentration antibody formulations with histidine or acetate buffers, dosing in ulcerative colitis, prefilled syringes and autoinjectors, CHO fed-batch processes. The search builder turns it into six search sets with keywords and classification codes, from C07K 16/24 for the antibody to A61M 5/20 for the device, plus the owners to watch. 1,860 documents come back in 412 families, duplicates removed.
“Say what to look for in plain words. The search builder turns it into search sets; the other agents read what comes back.”
The legal status checker keeps the 168 families granted or pending in at least one of eight launch markets. The claim reader splits 1,214 independent claims into 5,930 elements, ranges and units as written. The element matcher finds 61 claims that touch a feature of HLD-217; 38 fail on a decisive element and are recorded with it. The summariser ranks the other 23: 2 likely to read, 6 unclear, 15 do not read.
Elena opens the claim chart. At the top is EP 3 912 448 B1, claim 1 — Strathmore Biologics' stable high-concentration formulations of anti-TL1A antibodies, granted 14 Feb 2024, expiring 22 Sep 2038, 9.1 years after the US and EU launch. It is under opposition by Kerrow Pharma, with oral proceedings on 18 Nov 2026. Risk 40, first of 23.
“Show me the claim, the feature and where you read it — then I'll decide.”
Each element sits beside the HLD-217 feature it is matched to and the document behind it: a TL1A antibody, 150 mg/mL inside 120–200, sucrose 8% inside 4–10%, polysorbate 80 at 0.02%, and L-histidine 20 mM at pH 5.8 — all read, the formulation features from specification FS-217-04 §2.1. Element e asks for a viscosity below 20 cP at 25 °C. The development report has 18.6 cP at 20 °C only, so the matcher calls it unclear and likely to read. Re-checking the claim shows each agent's step.
“18.6 cP at 20 °C; viscosity falls as temperature rises, but 25 °C was never measured.”
The European claim is in force after launch in the UPC states, Spain, the UK and Switzerland, to 2038. Its Japanese member, JP 7 412 355 B2, is second on the list on the same reading. Its US member does not read: US 11,604,283 B2 claim 1 requires an IgG4 antibody, and HLD-217 is an IgG1. Each market carries its share of the launch forecast — the US 60%, the UPC states 17%, Japan 9%.
Dr. Aisha Bello has a development batch, DB-217-A12, in sodium acetate 20 mM at pH 5.3, already six months on stability. Elena sets the buffer to acetate in the design-around probe. Element f flips to does not read, and every chart is read again: two claims stop reading and one starts. EP 3 912 448 drops to 9th of 23, and the new top claim is Calder Immunology's US 11,739,152 B2 — unclear, because the epitope has not been mapped residue by residue.
“Element f does not read: Acetate buffer, no histidine. A claim reads only if every element does.”
Acetate brings Kerrow Pharma's pending US 2026/0311842 A1 into view: its claim 1 reads on an acetate formulation, though its claims may narrow before grant. Adding an autoinjector brings in Ensley Devices' US 11,452,880 B2 at risk 65. The cost of each change comes from CMC: for acetate, one comparability study, with the lock still on 15 Dec 2026 if decided by 6 Nov.
“Scenarios are what-ifs. The product profile changes only when CMC changes it.”
Six changes in the week of 5 October. Kerrow's acetate application published on 8 Oct — it does not read on HLD-217 today, only on the design-around. The EPO summoned the parties in the opposition to EP 3 912 448, with a preliminary opinion that claim 1 may lack an inventive step. Pellmont narrowed a European dosing claim to every 2 to 6 weeks, which no longer reads on HLD-217's 8-week maintenance dose.
“The office's preliminary view, not a decision. Claim 1 stays at the top of the list until the outcome.”
Three gaps decide eight claims: viscosity at 25 °C for the European and Japanese formulation claims, residue-level epitope mapping and a competition assay with antibody C-1D7 for four Calder claims, and the Feed C composition from the supplier for two Brightwater process claims. The request to Dr. Aisha Bello is drafted in counsel notes, marked privileged, and waits for Elena.
Nothing is sent until counsel approves it.
Elena records her conclusions: design around for the European and Japanese formulation claims, further data needed for Calder's US method-of-treatment claim. She signs them for FTO update 3; the five claims without a conclusion stay open for the next update. The signed analysis is locked and in the record — 34 months before the planned Q3 2029 launch.
“The agents' analysis is attached as the basis; it is not a legal opinion.”
The same FTO update, seen by the four people who carry it — what their month looked like, and what it looks like now.
Five specialist agents search, read, match, check and rank every claim; patent counsel concludes and signs.
Turns a plain-words scope and the product profile into a structured patent search — keywords, classification codes, owners and dates — with search sets that together cover the antibody, its formulation, its dosing, its device and its process.
Reads each independent claim and splits it into its elements in the order written, keeping ranges, units and “selected from” lists exactly, and marks the claim type: antibody, formulation, method of treatment, process or device.
Matches each element to a product feature and gives a verdict — reads, does not read or unclear — with the source document and section behind it, and says what would decide an unclear element.
Records status, expiry, term adjustment and supplementary protection, oppositions and litigation, country by country, and runs the watch every Monday at 06:00.
Ranks the claims by verdict, launch markets, years in force after launch and status, with the weights in Settings, and drafts the analysis for counsel — scope, search, results, the charts and the design-around options with their cost from CMC.
Draws every conclusion and signs. The agents propose; a named person decides.
Ask about any claim, feature or country in the landscape — or ask it to try a design-around. Every answer cites the document it read.
Why is EP 3 912 448 claim 1 at the top?
Five of its six elements read on HLD-217 — TL1A antibody, 150 mg/mL, 8% sucrose, polysorbate 80 and a 20 mM histidine buffer at pH 5.8. The sixth, viscosity below 20 cP at 25 °C, is unclear: CMC measured 18.6 cP at 20 °C only. It is in force in the UPC states, Spain, the UK and Switzerland until 22 Sep 2038. Whether that is a risk to act on is your call.
What happens if we switch to an acetate buffer?
With the buffer set to acetate 20 mM, pH 5.3 — the formulation of development batch DB-217-A12 — the buffer element of EP 3 912 448 claim 1 no longer reads, and its Japanese member drops too. One pending application, Kerrow's US 2026/0311842 A1, starts to read; its claims may still narrow. This is a what-if; the product profile is unchanged.
What is new on the watch this week?
Two things matter. Kerrow published US 2026/0311842 A1 on 8 Oct, which reads only on the acetate design-around; and the EPO set oral proceedings in the opposition to EP 3 912 448 for 18 Nov 2026. Pellmont also narrowed its European dosing claim to every 2–6 weeks, which takes it off HLD-217's 8-week regimen.
Draft the request to CMC for the missing data
Drafted in your counsel notes for Dr. Aisha Bello: viscosity of the 150 mg/mL formulation at 25 °C, and residue-level epitope mapping of HLD-217 against TL1A residues 140–160. Both turn an “unclear” into a verdict. Nothing is sent until you approve it.
13 screens from the working solution, on its sample data. Pick one to see it large.
What counts as a launch market, how claims are ranked, who is told and who signs are settings, on the record.
| Setting | Default | Choose from |
|---|---|---|
| Launch markets, share of the launch forecast | US 60% · UPC states 17% · Japan 9% · UK 5% · Spain 3% · Canada 3% · Australia 2% · Switzerland 1% | Share and launch date per market |
| Planned launch | US and Europe Q3 2029 · Canada Q4 2029 · Australia Q1 2030 · Japan Q3 2030 | Per market |
| Ranking weight — likely reads / unclear | 0.85 / 0.45 | Step up or down |
| Ranking weight — pending application / granted and opposed | 0.40 / 0.85 | Step up or down |
| Weekly run | Mondays 06:00 · Elena Marsh, Jonah Pereira | On · Off |
| Tell me at once | A grant in the Strathmore, Calder or Kerrow families | On · Off |
| Read new claims against every scenario | On — not only the product profile | On · Off |
| Who signs conclusions | Patent counsel only: Elena Marsh, Ravi Anand, Sofia Lindgren | Counsel only |
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 data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (AIPLA 2021 Report of the Economic Survey, cited in AIPLA's 2022 comment to ACUS — median cost to litigate a patent suit with $1–10m at stake; 35 U.S.C. § 284 — damages increased up to three times; I-MAK, Overpatented, Overpriced, 2018 — patent applications on one top-selling antibody). An analysis for patent counsel, not a legal opinion. People, companies, products, patents and documents named on this page are fictional — characters and sample data in the working solution.
It answers whether someone else's live patent claim covers your product in the markets where you will launch it. Every patent and application near the product is found, each independent claim is split into its elements, and each element is checked against the product — a claim reads only if every element does. Freedom-to-Operate Patent Analysis does that reading at scale and puts the result in the form counsel already uses, the claim chart.
No. It produces an analysis for patent counsel, and every page says it is not a legal opinion. The agents never say a claim is clear or that the product is free to operate; counsel draws every conclusion and signs it.
The claim reader splits each independent claim into its elements, keeping ranges, units and alternatives exactly as granted. The element matcher reads each element against the product profile and gives a verdict — reads, does not read or unclear — citing the product document and section. Anything not measured stays unclear, with the data that would decide it.
Change a feature in the design-around probe — buffer, maintenance dose, concentration, surfactant or device — and every chart is read again and the claims are re-ranked. You see which claims stop reading and which start, and you can save the change as a scenario, which goes to the CMC lead for a cost estimate. A scenario is a what-if; the product profile changes only when CMC changes it.
Every Monday at 06:00 the legal status checker reads what changed in the watched families — new publications and grants, oppositions, amended claims, continuations, annuities and lapses — and says what each means for the product. A grant in a family you name can be reported at once, and new claims can be read against every saved scenario.
The launch markets you set, each with its share of the launch forecast — in the working solution the US, the UPC states, Spain, the UK, Switzerland, Japan, Canada and Australia. It reads patent full text, national and regional registers including term adjustment and supplementary protection, the EPO opposition register, UPC and national court dockets, US district courts and the PTAB, and your own product documents.
Yes. The agents search, read, match, check and rank; only patent counsel records a conclusion or signs. The agents never judge validity, contact a patent owner, file anything with an office or change the product profile, and nothing they draft is sent until counsel approves it.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your launch plan, the product's source documents and your last FTO — and it goes live once every quality gate has passed. We will walk you through it on one of your own products first.
A cited generic entry date per market within hours of any change.
1-min demoParagraph IV Notice ManagementThe 45-day clock set and a cited response pack ready the same day.
1-min demoContract Review & RedliningCounterparty contracts redlined and back in under two days.
1-min demoWe'll run Freedom-to-Operate Patent Analysis on one of your own products and its launch markets.