Patent clearance search for a small-molecule launch — landscape mapping, claim analysis, validity scoring, and clearance opinion
Every freedom-to-operate (FTO) analysis begins with a precise definition of what is actually being cleared. An FTO opinion is only as good as its scope: a search run against the wrong compound form, the wrong claim of use, or the wrong jurisdiction gives false comfort. Scoping sets the boundaries — chemical structure, formulation, method of treatment, target markets, and the date against which patent term and prior art will be measured.
Freedom-to-operate is not the same question as patentability. A company can hold its own patent on a compound and still infringe someone else's patent by making, using, selling, or importing it — patents are exclusionary rights, not permission slips. Under 35 U.S.C. § 271(a), "whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States, or imports into the United States any patented invention... infringes the patent."
An FTO opinion therefore asks a narrower, more practical question than "is this novel?" — it asks "will launching this specific product, in this specific form, in this specific market, on this specific date, infringe any third party's currently enforceable patent claims?" The answer is jurisdiction-specific: patents are territorial rights, so a product can be perfectly clear to launch in Japan while blocked in the United States by a patent with no EU or JP counterpart.
FTO searches are distinct from patentability (novelty) searches in one crucial respect: an FTO search must include expired, abandoned, and even invalid patents in its initial sweep, because determining enforceability is itself part of the analysis — you cannot exclude a patent from consideration just because you suspect it is weak.
A common and costly scoping error: clearing only the active pharmaceutical ingredient (API) while ignoring formulation and method-of-use claims. Roughly 40% of pharma FTO blocking patents identified in practice are formulation or dosing-regimen patents filed years after the original composition-of-matter patent — "life-cycle management" filings that extend effective exclusivity well past the base compound patent.
The scoping document that anchors an FTO search typically fixes:
• Active ingredient: exact chemical structure (Markush position, salt form, polymorph, stereochemistry) — a specific crystalline polymorph can itself be separately patented even when the base molecule is off-patent • Formulation: excipients, dosage form (tablet, capsule, injectable, extended-release), particle size, coating • Method of use: indication, dosing regimen (e.g., "once-daily," "loading dose followed by maintenance"), patient subpopulation (pediatric, renally impaired), combination therapy partners • Manufacturing process: synthetic route, if process patents are a concern for markets where the API will be manufactured or imported • Launch markets: each target country requires a separate national/regional search — US (USPTO), Europe (EPO + national validations), Japan (JPO), China (CNIPA) • Priority/launch date: patent term is 20 years from the earliest claimed priority filing date (with possible patent term extension/adjustment); the launch date determines which patents will still be in force and enforceable at time of market entry
FTO analysis is run by a cross-functional team: patent counsel (in-house or outside counsel) leads claim construction and legal risk assessment; a patent search analyst or professional searcher executes the database queries; R&D/CMC scientists confirm the technical scope (exact structure, formulation, synthetic route); and business development weighs in on launch timing and market priority.
A critical procedural point: FTO opinions prepared by counsel are typically covered by attorney-client privilege and work-product doctrine, and companies deliberately route the analysis through counsel (rather than having scientists free-search and write conclusions) to preserve that privilege — because a documented finding of a blocking patent that is later ignored can support a finding of willful infringement, which under 35 U.S.C. § 284 exposes the infringer to enhanced (up to treble) damages.
With scope fixed, the search phase casts a wide net across every major patent database, using both text-based and classification-based queries, then narrows systematically. The goal at this stage is deliberately over-inclusive: recall matters more than precision, because a missed blocking patent is far more costly than a false positive that gets screened out in claim mapping.
A thorough landscape search layers multiple databases, because no single source has complete global coverage or identical search syntax:
• USPTO Patent Public Search (replacing legacy PatFT/AppFT): full-text search of granted US patents and published applications, with CPC/USPC classification browsing • Espacenet (EPO): covers 100+ million documents from 100+ patent offices worldwide, strong for European and PCT family tracking • PatentScope (WIPO): authoritative source for PCT international applications, useful for catching applications not yet nationalized in the target market • CNIPA search platforms: essential for China coverage — many blocking patents in China are utility models or filed only in Chinese, requiring machine or professional translation • FDA Orange Book (for US small-molecule drugs): lists patents the branded manufacturer has certified as covering the approved drug, formulation, or method of use — a mandatory check for any generic or follow-on entrant
Query strategy combines: (1) keyword/full-text search on the compound class, mechanism of action, and chemical substructure; (2) CPC/IPC classification codes (e.g., A61K31/xx for medicinal chemistry preparations); (3) chemical structure and Markush searching via structure-search tools (e.g., against CAS REGISTRY or DWPI); and (4) assignee/applicant searching against known competitors in the therapeutic area.
Once a seed set of relevant patents is found, citation mapping expands the landscape in both directions:
Backward citations (prior art cited by the patent): reveal what the patent examiner and applicant considered relevant — useful for understanding the technology's lineage and for later validity analysis (was material prior art missed during prosecution?).
Forward citations (later patents that cite this one): reveal the "patent family tree" growing outward — competitors building on the same technology, and critically, continuation and divisional applications from the same original assignee that may claim narrower or differently-scoped subject matter around the same core invention.
Continuation practice and "submarine" risk: because a continuation application can be filed at any time before the parent issues (and historically, under pre-1995 rules, patent term ran from issuance rather than filing), some patent families contain applications that stay pending for years, only to issue later with claims drafted specifically to cover products that entered the market in the interim. Modern US patents (post-1995, 20-year-from-filing term) have largely eliminated the extreme "submarine patent" scenario, but continuation practice still allows an applicant to keep a pending application alive and adjust claim scope in response to a competitor's marketed product — a live risk that requires monitoring pending applications, not just issued patents, throughout a long clinical development timeline.
In the search phase for this compound class, backward/forward citation mapping expanded the initial 62-patent keyword hit set to 247 documents across 89 patent families — including 3 pending continuation applications from the same assignee that had not yet issued, each potentially adjustable in scope before grant.
Raw search output is triaged in two passes:
Pass 1 — Family consolidation: 247 individual patent documents (patents + published applications across jurisdictions) collapse to 89 distinct patent families, since the same invention is often filed as a US patent, an EP validation, a JP national phase, and a CN national phase from one PCT application.
Pass 2 — Relevance and status triage: each family is screened for (a) whether it is still in force (not expired, not lapsed for non-payment of maintenance/annuity fees, not finally invalidated), (b) whether its claims are directed to subject matter that plausibly reads on the product's composition, formulation, or method of use, and (c) whether its jurisdictional coverage overlaps a target launch market.
Of 89 families, 14 survive triage as candidate blockers requiring full claim-by-claim mapping — patents whose claims are broad enough, current enough, and jurisdictionally relevant enough to warrant the detailed infringement analysis performed in Stage 3.
Claim mapping is the technical and legal heart of an FTO analysis. Patent claims — not the specification, not the abstract, not the drawings — define the legal boundary of what is protected. Each of the 14 candidate blocking patents is broken into its independent claims, each independent claim is broken into its individual limitations ("elements"), and each element is tested against the corresponding feature of the actual product.
Patent claims are read as a set of limitations joined by "comprising," "consisting of," or "consisting essentially of" — the choice of transitional phrase matters enormously:
• "Comprising" is open-ended: a product infringes if it contains every claimed element, even if it also contains additional, unclaimed elements. This is the default and most common phrasing in composition and method claims. • "Consisting of" is closed: the product must contain exactly the claimed elements and nothing more — any additional component defeats infringement of that claim. • "Consisting essentially of" is intermediate: additional elements are permitted only if they do not materially affect the basic and novel characteristics of the claimed invention.
The all-elements rule (or "all-limitations rule") governs literal infringement: every single element recited in the claim must be found in the accused product, exactly as construed. Missing even one element defeats a literal infringement finding for that claim — a rule sometimes summarized as "if you don't infringe the last element, you don't infringe."
Before mapping can even begin, disputed claim terms must be construed — given their proper legal meaning — following the framework from Phillips v. AWH Corp. (Fed. Cir. 2005): claim language is read in light of the specification and prosecution history from the perspective of a person having ordinary skill in the art (PHOSITA), with dictionaries and expert testimony as secondary aids.
The core work product of this stage is the claim chart: a table with the claim language broken into numbered elements in the left column, and the corresponding product feature (with supporting evidence — structural data, formulation specs, package insert language) in the right column, for every independent claim of every candidate blocker.
Example structure for a representative method-of-use claim:
[1a] "A method of treating [indication] in a patient in need thereof" → Product's labeled/intended indication matches [1b] "comprising administering a therapeutically effective amount of Compound X" → Product's active ingredient reads directly on claimed compound [1c] "wherein the compound is administered once daily" → Product's dosing regimen is twice daily — literal element NOT met [1d] "in a dosage of 50-200 mg" → Product dosage (75 mg) falls within claimed range
In this example, element [1c] is not literally met (once-daily vs. twice-daily), so literal infringement fails for claim 1 as a whole — but this triggers doctrine-of-equivalents analysis before the claim can be cleared.
Dependent claims are mapped only after their parent independent claim is found not to be a literal or equivalents match — a dependent claim narrows its parent, so if the broader independent claim is cleared, all claims depending from it are also cleared (a dependent claim cannot be infringed unless its parent is also infringed).
Even when literal infringement fails on one element, the doctrine of equivalents (DOE) can still find infringement if the accused product's corresponding feature performs substantially the same function, in substantially the same way, to achieve substantially the same result as the claimed element (the "function-way-result" or "triple identity" test from Graver Tank & Mfg. Co. v. Linde Air Products, 1950) — or, under the more modern "insubstantial differences" test, if a PHOSITA would consider the difference between the claimed and accused element insubstantial.
DOE is constrained by several important limiting doctrines that FTO analysis must apply:
• Prosecution history estoppel: if the applicant narrowed a claim during prosecution to overcome a prior-art rejection (e.g., narrowing "once daily" from a broader "administered periodically"), DOE cannot be used to recapture the surrendered scope (Festo Corp. v. Shoketsu Kinzoku, 2002) • All-elements rule applies to DOE too: DOE is applied element-by-element, not to the invention as a whole — you cannot argue the overall product is equivalent while ignoring that a specific claimed element is entirely absent • Dedication-disclosure rule: subject matter disclosed in the specification but not claimed is dedicated to the public and cannot later be captured via DOE • Vitiation: DOE cannot be used to effectively read a claim element out of the claim entirely — if applying DOE would render a specific claim limitation meaningless, equivalents will be rejected
In the once-daily vs. twice-daily dosing example above, a court would likely find the difference non-insubstantial if the specification and prosecution history show the applicant specifically claimed once-daily dosing to secure a patient-compliance advantage over prior twice-daily regimens — making that dosing frequency the very point of novelty, not an insubstantial variation.
Of the 14 candidate blockers, element-by-element mapping found 6 patents where every claim element literally reads on the product (high infringement risk), and 4 additional patents where one non-matching element carries meaningful DOE exposure — most often a formulation or dosing-regimen limitation with a close but not identical match. The remaining 4 were cleared: at least one claim element is absent with no plausible DOE argument.
A patent whose claims read on your product is only a real commercial threat if it is also valid and enforceable. Stage 4 scores each of the 10 infringement-risk patents (6 literal + 4 DOE) on four independent axes — validity strength, remaining term, prosecution history exposure, and the patent owner's litigation posture — to convert a binary "infringes / does not infringe" finding into a graded commercial risk.
Validity analysis asks: if challenged, would this patent survive? Under 35 U.S.C. § 102, a claim is invalid if it was anticipated — every element disclosed in a single prior-art reference before the effective filing date. Under 35 U.S.C. § 103, a claim is invalid for obviousness if the differences between the claimed invention and the prior art would have been obvious to a PHOSITA, evaluated under the Graham v. John Deere (1966) factors: scope and content of the prior art, differences between the prior art and the claims, the level of ordinary skill in the art, and objective indicia of non-obviousness ("secondary considerations") such as unexpected results, commercial success, long-felt but unsolved need, and failure of others.
For pharmaceutical claims specifically, obviousness analysis frequently turns on "obvious to try" reasoning post-KSR International Co. v. Teleflex (2007): if there were a finite number of identified, predictable solutions and a PHOSITA would have had good reason to pursue the known options, a resulting invention may be obvious even without an explicit teaching, suggestion, or motivation in the prior art. This has made formulation, salt-selection, and dosing-regimen patents more vulnerable to obviousness attack than they were pre-KSR, because the "obvious to try" pathway is well developed in pharma prior art (routine optimization of known variables like dose, polymorph, or release profile).
A validity score for each blocking patent aggregates: strength of prior art uncovered in the backward-citation search (Stage 2) that was NOT cited during prosecution (undisclosed material prior art is a strong invalidity argument and also raises inequitable conduct considerations), the presence of objective indicia the patentee could assert in defense, and whether analogous claims in the same family have already been challenged and survived (or failed) at the PTAB or in EPO opposition.
Every issued patent carries a "file wrapper" — the complete prosecution history including office actions, examiner rejections, and applicant responses/amendments — that is public record and essential reading for both DOE analysis (Stage 3) and validity scoring. Key items reviewed:
• Claim amendments made to overcome prior-art rejections, and the arguments made to distinguish the prior art — these create prosecution history estoppel limiting DOE scope and can also be used as admissions in a later validity challenge • Restriction requirements and resulting divisional applications — reveal the full scope of what the applicant considered patentably distinct inventions from one original disclosure • Terminal disclaimers — filed to overcome obviousness-type double patenting rejections against a related family member; a terminal disclaimer ties the challenged patent's expiry to the earlier patent's expiry and can be a signal of a crowded, related-claims family • Examiner interview summaries and any IDS (Information Disclosure Statement) filings — show what prior art was actually considered
For Orange Book-listed patents specifically, the Hatch-Waxman framework (21 U.S.C. § 355) adds a further wrinkle: a generic ANDA filer submitting a Paragraph IV certification asserting non-infringement or invalidity of an Orange Book patent triggers a potential 30-month stay of FDA approval if the brand sues within 45 days — a procedural risk layer distinct from the underlying infringement/validity merits, and one that must be factored into launch-timing risk even for a technically strong non-infringement position.
Remaining term: US utility patent term is 20 years from the earliest non-provisional filing date, subject to Patent Term Adjustment (PTA, compensating for USPTO delay) and, for the first-issued patent covering an FDA-approved active ingredient, Patent Term Extension (PTE) of up to 5 additional years under 35 U.S.C. § 156 to compensate for regulatory review time (capped so total post-approval exclusivity does not exceed 14 years). A blocking patent expiring before or shortly after target launch carries low commercial risk regardless of infringement strength; one with 10+ years remaining is a durable barrier.
Owner litigation posture is scored from: the assignee's history of asserting this patent family (or sibling families) against competitors, whether the assignee is a practicing entity (a competitor pharma company, likely to seek an injunction/exclusivity) versus a non-practicing entity (more likely to seek licensing revenue), and whether the patent has already survived a validity challenge (IPR final written decision upholding claims, or EPO opposition division maintaining the patent) — a "battle-tested" patent that has already survived a validity attack is a materially stronger blocker than an unchallenged one of similar claim scope, because the population of easily invalidated patents has already been filtered out by the earlier challenge.
Combining all four axes (validity strength, remaining term, prosecution estoppel exposure, owner posture) into a composite 0-10 risk score, 3 of the 10 infringement-risk patents score ≥7 — the "high-risk" set requiring active mitigation before launch; 5 score 4-6 (moderate, monitor and consider design-around); 2 score ≤3 (low risk despite technical infringement, likely to fail a validity challenge if asserted).
The single highest-risk patent in this landscape (risk score 8.7/10) is a formulation patent held by a practicing competitor with 11.2 years of remaining term (including a granted Patent Term Extension), no undisclosed material prior art found in backward citation review, and a prior successful assertion against a different generic entrant in the same drug class two years earlier — a combination of strong validity, long duration, and demonstrated willingness to litigate that moves this patent to the top of the mitigation queue in Stage 5.
The FTO analysis culminates in a written legal opinion, jurisdiction by jurisdiction, classifying the product's launch risk and recommending a mitigation path for each patent that cannot simply be cleared. This is a business decision informed by legal analysis, not a pure legal question — the same set of facts can support different recommended actions depending on the company's risk tolerance, launch timeline, and the commercial value of the market.
FTO opinions are conventionally expressed in one of three tiers, per jurisdiction and often per patent:
Clear: no valid, enforceable, in-force patent claim reads on the product as scoped. Launch may proceed without further action on this patent. A "clear" opinion is never absolute certainty — it is counsel's reasoned judgment based on available information, and it should be revisited if new patents publish or the product design changes.
Clear-with-risk (sometimes "yellow" or "cautionary"): one or more patents present some infringement exposure, but counsel assesses the risk as manageable — typically because validity is questionable, remaining term is short, the equivalents argument is weak, or the patent owner has shown no history of assertion. Launch may proceed, often paired with active monitoring, a inter partes review filing held in reserve, or budget reserved for a potential license negotiation.
Blocked: a valid, enforceable, in-force claim is very likely infringed with no credible invalidity or non-infringement position. Launch as currently scoped is not recommended without one of the mitigation actions below — the product must be redesigned, the launch delayed past expiry, a license obtained, or the patent invalidated before commercialization.
The most direct resolution: negotiate a license from the patent owner, converting an infringement risk into a contractual royalty or lump-sum obligation. Licensing is typically pursued when: the blocking patent is strong (high validity score, long remaining term) and design-around is not commercially or technically feasible without compromising product performance, the patent owner is a non-practicing entity or a company with no competing product (more open to licensing than a direct competitor protecting market share), or when cross-licensing is possible (the launching company holds patents useful to the blocking patent's owner).
License negotiation leverage is directly informed by the validity score from Stage 4 — a company walks into negotiations very differently when it is holding a well-documented invalidity position (undisclosed prior art, strong obviousness argument) versus when the patent has already survived a challenge. Royalty rates in pharma licensing for a blocking formulation or method patent typically range from low-to-mid single-digit percentages of net sales, though this varies enormously with remaining term, market size, and negotiating leverage.
Where technically feasible, redesigning the product to fall outside the claim scope is often the most durable solution, because it eliminates the infringement risk entirely rather than converting it into an ongoing royalty obligation or leaving validity untested. Common pharma design-arounds identified in claim mapping:
• Reformulation: substituting an excipient, changing particle size distribution, or altering the release profile to fall outside a formulation claim's specific limitations • Dosing regimen change: adjusting frequency or dose range to move outside a claimed regimen (as in the once-daily vs. twice-daily example from Stage 3) — provided the change does not compromise clinical efficacy or safety, which itself requires new or bridging clinical data • Alternative salt form or polymorph: selecting a different, non-claimed crystalline form or salt of the same active molecule, if pharmacokinetically equivalent • Process design-around: adopting a synthetic route that avoids a claimed manufacturing process, relevant when process patents (rather than composition patents) are the blocker
A design-around must itself be validated by renewed claim mapping — a naive reformulation intended to avoid one claim element can inadvertently fall within a different, previously-uninfringed claim of the same or a related patent, so the redesigned product should be re-run through Stage 3 claim mapping before being finalized.
Inter partes review (IPR) at the Patent Trial and Appeal Board (PTAB), created by the America Invents Act (2011), allows a challenger to petition for cancellation of issued US patent claims based on prior art patents or printed publications (§ 102/§ 103 grounds only — no § 112 or inequitable conduct grounds). PTAB must decide whether to institute review within 6 months of the petition, and if instituted, must issue a final written decision within 12 months (extendable by 6). IPR is faster and cheaper than district court litigation and uses the broadest reasonable interpretation historically, though post-2018 rules align claim construction more closely with the Phillips district-court standard. A strong invalidity position uncovered in Stage 4 (undisclosed material prior art) is exactly the fact pattern IPR is designed for.
EPO opposition provides an analogous but distinct mechanism in Europe: any party may file opposition within 9 months of a European patent's grant, requesting the EPO Opposition Division revoke or amend the patent on grounds including lack of novelty, lack of inventive step, or insufficient disclosure — centrally affecting all national validations of the European patent in one proceeding, which is far more efficient than challenging each national validation separately.
Launch-at-risk: in some cases a company proceeds to commercialize despite an unresolved blocking patent, accepting exposure to injunction and damages (including potential treble damages for willful infringement under § 284 if the infringement is found deliberate) in exchange for early market entry — a strategy generally reserved for cases with a strong invalidity or non-infringement position, a large first-mover commercial advantage, and organizational risk tolerance for litigation. Launch-at-risk decisions are made at the executive level, not by patent counsel alone, precisely because they trade a quantifiable legal risk for a quantifiable commercial upside.
Final recommendation for this program: of 3 high-risk patents, the top-ranked formulation patent (risk score 8.7/10, practicing-entity owner with prior litigation history) is routed to design-around via reformulation, since the clinical team confirmed an alternative excipient system achieves bioequivalence without new efficacy trials. The remaining 2 high-risk patents — both process patents with weaker validity scores due to undisclosed prior art found in citation mapping — are routed to IPR petitions filed 14 months ahead of target launch, leaving time for a final written decision before commercialization. Overall opinion: Clear-with-risk in all 4 markets, contingent on successful completion of the reformulation and IPR filings.