How to attack weak patents, fortify your portfolio, and turn every validity challenge into a competitive lever — across the U.S., Europe, and the post-UPC landscape.
Part 1: Why Patent Invalidity Is a Business Discipline, Not a Legal Afterthought
A drug patent is not simply a legal certificate. It is the core asset on which pricing power, revenue projections, partnership valuations, and M&A premiums are built. When that asset faces an invalidity challenge, the downstream consequences touch every function in the organization: finance reprices exclusivity windows, business development recalibrates licensing terms, manufacturing revises capacity assumptions, and equity markets react in real time.
The scale of these stakes is difficult to overstate. A single blockbuster patent covering a drug with $5 billion in annual U.S. revenue can represent $40-50 billion in net present value of future cash flows. A successful Paragraph IV challenge, or a final written decision from the Patent Trial and Appeal Board (PTAB) invalidating the core composition-of-matter claim, can destroy that value inside 18 months.
The inverse is equally powerful. For a generic manufacturer, a winning invalidity challenge against the right patent unlocks 180 days of exclusive generic competition, a period that can generate $500 million to $1.5 billion in gross profit depending on the molecule. For a biosimilar company, clearing a biologic’s patent thicket through Inter Partes Review (IPR) or an EPO opposition can be the difference between a viable product launch and an indefinitely delayed one.
This guide does not treat invalidity as a reactive legal problem. It treats it as a proactive strategic tool — one that requires the same analytical rigor and resource commitment as clinical development or commercial launch planning.
The Patent Bargain and Its Enforcement Mechanism
The entire structure rests on what courts call the ‘patent bargain.’ An inventor receives a time-limited monopoly in exchange for teaching the public how to make and use a genuinely new and non-obvious invention. Invalidity proceedings are the system’s mechanism for enforcing that bargain. When a patent is granted on something that was already known, was obvious to skilled practitioners, or was inadequately disclosed, it represents a broken bargain — a rent-seeking monopoly that restricts competition without providing the offsetting public benefit of genuine technological teaching.
From a policy standpoint, this makes invalidity challenges legitimate and necessary. From a business standpoint, it means that any patent is only as durable as the quality of its underlying disclosure and the accuracy of its claims relative to what was known at the time of filing.
Key Takeaways: Part 1
Patent invalidity is a core business risk and a core business opportunity, depending on which side of the Orange Book you sit on. The financial consequences of a successful challenge — whether offensive or defensive — are measured in billions, not millions. Every business leader with exposure to pharmaceutical IP needs a working command of the grounds, forums, and strategic options covered in this guide.
Part 2: The Core Grounds for Invalidity — A Technical Dissection
Understanding which legal vulnerabilities apply to a given patent is the precondition for any strategic decision. The grounds for invalidity are not abstract legal categories; they are precision instruments, each one targeting a different type of flaw in the patent’s construction.
Lack of Novelty (Anticipation): Was the Invention Already Public?
The novelty requirement — codified in 35 U.S.C. § 102 in the U.S. and Article 54 of the European Patent Convention (EPC) — establishes the most fundamental condition for patentability: the claimed invention must not already exist in the public domain. In U.S. litigation, the invalidity argument based on this ground is called ‘anticipation.’
The technical standard is demanding for challengers in one sense but absolute in another. A single piece of prior art — a prior patent, a published journal article, a product that was publicly sold, a conference presentation, even a publicly visible manufacturing process — must disclose each and every element of the asserted patent claim. No gap is allowed. One missing claim element and the anticipation argument fails. But when a challenger finds a single reference that cleanly discloses the entire claimed invention, the patent has no defense. Under the America Invents Act (AIA), ‘prior art’ includes disclosures anywhere in the world, not just in the U.S., which substantially expands the search territory.
The Doctrine of Inherent Anticipation: Expanding the Prior Art Universe
Inherent anticipation is the doctrine with the most practical leverage in pharmaceutical invalidity campaigns, particularly against evergreening patents on formulations, polymorphs, and method-of-use claims.
The rule: even if a prior art reference does not explicitly disclose a particular property of a compound or process, the patent covering that property is anticipated if the property is the natural and inevitable result of what the prior art does describe.
The tactical application is direct. A company seeking to attack a method-of-use patent on, say, the use of a known molecule to treat a neurological condition does not need to find a prior art paper that names that indication. It needs to find a reference that fully discloses the compound and its administration in a context where the therapeutic effect would necessarily follow. Expert testimony and new experimental data can then demonstrate that the result was always inherent in the prior disclosure, even if no one had characterized it at the time.
The Federal Circuit has consistently held that discovering a ‘previously unappreciated property’ of an existing compound does not create novelty in that compound. This doctrine is a structural threat to polymorph patents, salt-form patents, and metabolite patents — the workhorses of pharmaceutical evergreening — because those inventions often reclaim a compound that existed in the prior art in an essentially identical chemical state.
Obviousness: The Dominant Weapon and the KSR Effect
Obviousness under 35 U.S.C. § 103, or ‘lack of inventive step’ under EPC Article 56, is the ground on which most pharmaceutical patent challenges are ultimately won or lost. It is more complex than anticipation, more fact-intensive, and for that reason, more susceptible to a well-constructed challenge that weaves together multiple pieces of prior art into a coherent narrative.
The legal standard requires that the claimed invention be assessed from the perspective of a hypothetical ‘person having ordinary skill in the art’ (POSITA) — a skilled practitioner in the relevant field who is presumed to know all publicly available prior art as of the patent’s effective filing date. If that hypothetical POSITA would have been motivated to combine existing knowledge to arrive at the claimed invention, and would have had a reasonable expectation of success in doing so, the patent is invalid for obviousness.
The four-factor ‘Graham v. John Deere’ framework governs this analysis in the U.S.: (1) the scope and content of the prior art; (2) the differences between the prior art and the claimed invention; (3) the level of ordinary skill in the relevant art; and (4) objective indicia of non-obviousness (discussed below).
KSR International Co. v. Teleflex (2007): A Structural Shift
For decades before KSR, obviousness analysis required challengers to find an explicit ‘teaching, suggestion, or motivation’ (TSM) in the prior art to combine references. That requirement made the obviousness standard relatively predictable and, for innovators, relatively manageable.
The Supreme Court’s 2007 KSR decision dismantled the rigid TSM test. The Court held that innovation often arises not from flashes of genius but from ordinary skill, routine experimentation, and the predictable application of known techniques to known problems. If a technique improves one device in a known way, applying it similarly to a related device is likely obvious — no explicit suggestion required.
In pharmaceutical chemistry, where medicinal chemists routinely modify lead compounds through structural analogs, bioisosteric replacements, and stereochemical variations, KSR made it dramatically easier for challengers to argue that small molecular modifications were obvious design choices. A challenger no longer needs a paper that says ‘add a fluorine at position 4’; it can simply demonstrate that fluorination at that position was a known technique applied by practitioners in the field for predictable purposes.
The ‘Obvious to Try’ Doctrine in Drug Development
A specific variant of the KSR framework with high relevance to pharmaceutical cases is the ‘obvious to try’ doctrine. When the prior art identifies a problem, points toward a finite number of predictable solutions, and gives skilled practitioners reason to expect that trying those solutions would work, a court may find the resulting invention obvious.
The critical qualification is ‘reasonable expectation of success.’ The doctrine does not apply when the prior art presents an unpredictable field where success is genuinely uncertain, or where the number of possible approaches is too large to qualify as a ‘finite set.’ For high-profile biologics and first-in-class mechanisms, this qualification often provides a genuine defense. For follow-on formulation work, process chemistry, and analog series development, it frequently does not.
Secondary Considerations of Non-Obviousness: The Patent Holder’s Shield
Secondary considerations are objective, real-world evidence that the claimed invention was not, in fact, obvious. The Supreme Court in Graham explicitly identified these factors to guard against hindsight reconstruction — the temptation to stitch together prior art references after the fact to make an invention appear inevitable when it was not.
The most commonly deployed secondary considerations in pharmaceutical litigation include: commercial success of a product embodying the invention; long-felt but unresolved need in the field; documented failure of others who attempted to solve the same problem; unexpected results that would not have been predicted from the prior art; and copying by competitors or praise from technical experts in the field.
The burden is on the patent holder, and there is a critical procedural requirement: a clear ‘nexus’ must be established between the secondary consideration evidence and the specifically novel features of the claimed invention. Commercial success driven by marketing spend, prescriber relationships, formulary placement, or prior-art features of the product does not support non-obviousness. Only success attributable to the patented inventive feature itself qualifies. Courts scrutinize nexus arguments closely, and the failure to establish a tight nexus is one of the most common reasons secondary considerations fail as a defense.
Practically, this means innovators must begin building the nexus record before litigation, not during it. Sales attribution analyses, market research linking clinical outcomes to novel structural features, and documentation of competitor efforts to replicate the invention should be assembled proactively, over years, not assembled reactively in the months before trial.
Insufficient Disclosure: The Amgen Earthquake and What Comes Next
The patent bargain requires the inventor to teach the public. 35 U.S.C. § 112 in the U.S., and Article 83 of the EPC in Europe, enforce that obligation through two related doctrines: written description and enablement.
Written description requires the specification to demonstrate that the inventor was actually in possession of the full scope of the claimed invention at the time of filing. It prevents a common prosecution tactic: filing early with narrow examples and then later claiming a vastly broader invention based on the original priority date.
Enablement requires the specification to teach a POSITA how to make and use the full scope of the claimed invention without ‘undue experimentation.’ The amount of experimentation that qualifies as ‘undue’ is context-dependent, based on factors including the complexity of the relevant science, the breadth of the claims relative to the disclosed examples, and whether a predictable methodology exists for practitioners to fill in gaps.
Amgen Inc. v. Sanofi (2023): Functional Genus Claims Under Siege
The Supreme Court’s unanimous 2023 decision in Amgen Inc. v. Sanofi is the most consequential patent ruling in the biopharmaceutical space since KSR. It did not create new doctrine; it applied established enablement principles with unusual clarity to a class of patents that had become a standard feature of large-molecule drug development.
The dispute centered on Amgen’s PCSK9 antibody patents, which protected Repatha (evolocumab). Rather than claiming only the specific antibody sequences Amgen had developed, the patents asserted claims over the entire functional genus of antibodies that could bind to a specific epitope on the PCSK9 protein and block its interaction with LDL receptors. This class potentially included millions of structurally distinct antibody sequences.
Amgen disclosed the structures of 26 specific antibodies and a general methodology for identifying additional antibodies through iterative screening and testing. Sanofi, which had independently developed Praluent (alirocumab), argued that Amgen’s disclosed examples plus screening methodology did not enable a POSITA to practice the full scope of the claims without exactly the kind of extensive trial-and-error research the claims purported to replace.
The Supreme Court agreed, articulating the governing principle in a sentence that now appears in virtually every biologic invalidity analysis: ‘the more you claim, the more you must enable.’ A patentee who claims an entire functional class of molecules must provide disclosure sufficient for a POSITA to make and use the full scope of that class. Providing a handful of examples plus a research roadmap to find more is not sufficient when the roadmap still requires extensive, unpredictable experimentation.
The strategic consequences split differently for challengers and innovators:
For challengers: Amgen elevated enablement from a secondary invalidity argument to a primary line of attack on any biologic patent with broad functional claims. The question ‘does the patent teach how to make everything it claims?’ is now a threshold inquiry in every biologic IPR or litigation strategy.
For innovators: The old standard practice of filing early with functional genus claims and a small number of exemplified species is now structurally vulnerable. The new imperative involves providing substantially more working examples — not a handful of representative antibodies, but dozens or hundreds. It also requires more careful calibration of claim breadth to the actual experimental support in the specification, accepting narrower initial claims in exchange for defensibility. Continuation strategy becomes critical: filing continuations as the development program generates additional data allows the company to build out a claim portfolio with each layer grounded in a richer experimental record.
Other Fatal Flaws: §101, Inequitable Conduct, Inventorship, and Double Patenting
Beyond the primary grounds, several other vulnerabilities can invalidate or render unenforceable pharmaceutical and biologic patents.
Patent-ineligible subject matter under 35 U.S.C. § 101 prohibits patents on laws of nature, natural phenomena, and abstract ideas. For pharmaceutical purposes, this ground has its highest practical relevance in diagnostic method patents and personalized medicine claims. A patent claiming a correlation between a specific genetic biomarker and a disease risk, or between a metabolite level and a drug dosage requirement, is vulnerable to a §101 challenge if the claim does not add a sufficiently ‘inventive concept’ beyond the natural correlation itself. Mayo Collaborative Services v. Prometheus Laboratories (2012) and Association for Molecular Pathology v. Myriad Genetics (2013) established the legal framework that makes companion diagnostic patents particularly difficult to secure and maintain.
Inequitable conduct is an unenforceability defense specific to misconduct during USPTO prosecution. It requires proving by clear and convincing evidence both that the applicant withheld material prior art or submitted false information, and that they did so with a specific intent to deceive the examiner. The high bar for proving intent — established by the Federal Circuit’s en banc decision in Therasense v. Becton Dickinson (2011) — makes successful inequitable conduct defenses relatively rare. But when the evidence of deceptive intent is clear, the consequence is severe: permanent unenforceability of the entire patent, not just the affected claims.
Incorrect inventorship — either adding non-inventors as courtesy co-inventors or omitting actual contributors — can also ground an invalidity challenge. In collaborative research environments and academic-industry partnerships, inventorship disputes are a persistent hazard. Honest errors can generally be corrected before the patent office, but intentional misrepresentation creates inequitable conduct risk.
Double patenting prevents a company from obtaining multiple patents on the same invention as a mechanism to serially extend monopoly protection. Obviousness-type double patenting, which bars a later patent on an invention that is only an obvious variation of an earlier one, is the more commonly encountered form. Applicants file ‘terminal disclaimers’ to overcome double patenting rejections during prosecution, but this tool links the term of the later patent to the earlier one and can complicate later litigation.
Part 3: IP Valuation — How Invalidity Risk Reprices Patent Assets
For portfolio managers, M&A teams, and licensing executives, the financial value of a drug’s IP position is not binary. It is a probability-weighted calculation that must account for the likelihood, timing, and outcome of invalidity challenges. A patent covering a $3 billion-per-year drug with a nominal expiry in 2031 does not have a clean six-year runway if there is a credible Paragraph IV ANDA on file or a pending IPR petition with strong prior art.
Valuing Composition-of-Matter Patents vs. Secondary Patents
The pharmaceutical industry’s IP protection structure typically layers two distinct categories of patents. Composition-of-matter (COM) patents claim the active pharmaceutical ingredient itself — its chemical structure, its salt forms, its polymorphs. These are the foundational assets, the hardest to challenge (because they must anticipate or render obvious the specific molecule), and the most valuable per unit of expected remaining exclusivity.
Secondary patents cover formulations, manufacturing processes, dosing regimens, methods of treatment, and patient selection methods. These are the instruments of evergreening — the mechanism by which companies extend effective market exclusivity beyond the COM expiry. They are also substantially more vulnerable to invalidity challenges, because the prior art burden for a formulation or method-of-use patent is almost always lower than for a novel chemical entity. When a generic’s invalidity team conducts a nullity search, secondary patents are often the first to fall.
The IP valuation consequence is direct: a portfolio consisting exclusively of secondary patents, or a drug whose COM patent has expired and whose market exclusivity depends entirely on secondary protection, carries meaningfully higher invalidity risk and must be discounted accordingly. Financial models that assign full exclusivity value to Orange Book-listed secondary patents, without accounting for Paragraph IV risk, systematically overvalue these assets.
The Evergreening Premium and Its Discount
Evergreening — the strategic practice of filing secondary patents to extend market exclusivity beyond the COM patent’s expiry — is legal, commonplace, and commercially rational. AbbVie’s Humira (adalimumab) is the most cited example in recent pharma history, with more than 165 U.S. patents covering the drug in various forms, many of which extended effective exclusivity years beyond the base composition patent. The resulting biosimilar entry delay has been the subject of significant policy and litigation attention.
But evergreening strategies carry a compounding invalidity risk premium. Each secondary patent filed on a drug that has been on the market for years faces a growing body of scientific literature, clinical experience, and competitor patent filings that all constitute prior art. The longer the drug has been studied and the more widely its properties are known, the harder it becomes to patent ‘new’ aspects of those properties. Courts have grown progressively more skeptical of evergreening patents, and the post-KSR obviousness standard makes formulation and method-of-use patents filed five or more years after initial approval particularly vulnerable.
Detailed evergreening technology roadmaps are addressed in Part 6.
Key Takeaways: Part 3
IP valuation in pharma requires distinguishing between the durability of COM patents and the elevated invalidity risk of secondary and evergreening patents. Portfolio managers pricing in full exclusivity value from secondary patents without Paragraph IV probability adjustments are carrying material modeling error. Every Orange Book-listed patent covering a drug with more than $500 million in annual revenue should be assessed for challenge likelihood using the factors outlined in this guide.
Part 4: The U.S. Battlefield — Hatch-Waxman Litigation and the PTAB
The U.S. is simultaneously the largest pharmaceutical market in the world and the most litigated. The legal architecture governing pharmaceutical patent disputes in the U.S. is purpose-built, with a set of incentives and procedures that do not exist in any other jurisdiction. Mastery of this architecture is non-negotiable for any company competing in the U.S. market.
Paragraph IV: The Mechanics of the Gauntlet
The Drug Price Competition and Patent Term Restoration Act of 1984 — the Hatch-Waxman Act — created the Abbreviated New Drug Application (ANDA) pathway for generic drugs. Under this pathway, a generic manufacturer seeking approval for a copy of a brand-name drug files an ANDA with the FDA, demonstrating bioequivalence rather than conducting independent clinical trials.
The act of patent engagement is built directly into the ANDA process. Every ANDA applicant must make a certification for each patent listed in the FDA’s Orange Book that covers the reference listed drug (RLD). The four possible certifications range from ‘the patent has expired’ (Paragraph III) to the most consequential certification: Paragraph IV, in which the ANDA filer attests that the listed patent is either invalid, unenforceable, or will not be infringed by the generic product.
A Paragraph IV certification is a statutory act of patent infringement. Congress deliberately structured it this way to allow the validity dispute to be adjudicated in federal court before the generic enters the market. Once the patent holder receives notice of the Paragraph IV filing, it has 45 days to file suit. If it files within that window, an automatic 30-month stay of ANDA approval takes effect, preserving the brand manufacturer’s market exclusivity while the litigation proceeds.
The economic incentive driving this system is the 180-day first-filer exclusivity reward. The first generic company to file a substantially complete ANDA with a Paragraph IV certification against each listed patent earns six months of exclusive generic sales before any other generic can enter the market. For a drug with $3 billion in annual U.S. revenue, this period can generate $400 million to $800 million in gross profit for the first-filer, even at the typical 80% price discount to brand.
The result is a system that actively incentivizes patent challenges. Data consistently shows that drugs with higher annual revenues face higher rates of Paragraph IV challenge, with drugs in the top decile of U.S. revenue facing challenge rates of 70-90%. The pharmaceutical markets of New Jersey and Delaware have become the principal federal court venues for these disputes, producing a sophisticated cadre of district court judges with deep familiarity in pharmaceutical patent law.
The strategic dynamic also drives increasingly aggressive timing. First-filer status rewards speed, so generic manufacturers file Paragraph IV certifications as early as possible after brand approval — often within 12 to 24 months of the brand’s NDA approval date — which has compressed the period of undisturbed exclusivity that innovators can practically expect to achieve.
Inter Partes Review: Statistics, Strategy, and Estoppel
The America Invents Act of 2011 created Inter Partes Review (IPR) before the USPTO’s Patent Trial and Appeal Board (PTAB), a parallel administrative forum for challenging issued patents on grounds of novelty and obviousness based on prior patents and printed publications. IPR cannot address other invalidity grounds — no enablement challenges, no inequitable conduct, no §101 arguments, no prior use or prior sale evidence.
Despite that scope limitation, IPR has become a dominant instrument of pharmaceutical patent litigation, and the statistics explain why. Patents that reach a final written decision at the PTAB are invalidated in whole or in part at rates that would be unrecognizable in district court. Across all IPR proceedings, approximately 78% of challenged claims that reach a final written decision are found unpatentable. In approximately 70% of final decisions, all challenged claims fall. These are not cherry-picked numbers from a favorable year; they have been roughly stable since the PTAB’s establishment and have trended upward since 2019.
Several structural features of IPR create these outcomes. The standard of proof is lower: a challenger in IPR must prove unpatentability by a ‘preponderance of the evidence’ (‘more likely than not’), while a district court challenger must meet the higher ‘clear and convincing evidence’ standard. There is no presumption of validity at the PTAB. The decision-makers are technically trained Administrative Patent Judges (APJs) — patent attorneys with advanced scientific degrees — rather than generalist judges and lay juries who may struggle with complex chemistry or molecular biology. And the rules of claim construction, while now aligned with the district court Phillips standard, are applied by judges with deep familiarity in how patent claim language operates.
The timeline advantage is real but must be contextualized. By statute, from petition filing to institution decision takes up to six months; from institution to final written decision takes 12 months. This 18-month maximum is substantially faster than the 24-36 month timeline to trial in a district court case. But the speed advantage interacts with the Hatch-Waxman 30-month stay in ways that create timing risk.
Fintiv: The Coordination Problem and Its Management
The PTAB’s Apple Inc. v. Fintiv precedent allows the Board to exercise discretion to deny institution of an IPR petition if a district court case involving the same patent is proceeding on a schedule that would place the trial before the IPR’s final written decision. Innovator companies learned quickly to exploit this by pushing for early trial dates in district court, specifically to disqualify concurrent IPR petitions.
Managing Fintiv risk requires precise tactical planning. Generic and biosimilar challengers should file IPR petitions as early as possible — ideally within 12 months of the Paragraph IV notice — to minimize the chance that district court proceedings have advanced to a point that triggers discretionary denial. They can also make strategic stipulations to the district court, committing not to raise the same prior art-based invalidity arguments in the district court case that are pending in the IPR. This reduces the ‘overlap’ factor that drives Fintiv denials.
Estoppel: The Highest-Stakes Feature of IPR
The estoppel provision of IPR is the feature that demands the most careful pre-petition analysis. If a petitioner requests IPR and the proceeding reaches a final written decision, the petitioner is statutorily barred from asserting in any other proceeding any ground of invalidity that was raised in the IPR or that ‘reasonably could have been raised’ in the IPR based on prior art patents and publications. This is a broad and powerful estoppel. A failed IPR on a strong patent can leave a generic manufacturer substantially disarmed in its parallel district court case.
The implication is that an IPR petition must be filed with the company’s strongest prior art in hand, and the petition must be constructed with full awareness that the arguments in it may be the only arguments available if the PTAB declines to invalidate. Weak petitions filed hastily are not just likely to fail at the PTAB — they can critically undermine the overall invalidity strategy.
Comparison Framework: District Court vs. IPR
Feature
District Court (Hatch-Waxman)
Inter Partes Review (PTAB)
Available Grounds
All invalidity grounds
Novelty (§102) and Obviousness (§103) only, prior art patents and publications
Standard of Proof
Clear and Convincing Evidence
Preponderance of the Evidence
Presumption of Validity
Yes
No
Decision-Maker
Generalist judge, often jury
Panel of three technically qualified APJs
Discovery
Extensive: documents, depositions, expert reports
Limited: cross-examination of declarants, cited exhibits
Timeline to Decision
24-36 months to trial
12-18 months from petition
Estimated Cost
$4M-$10M for complex pharma cases
$300K-$900K
Estoppel
Narrow (res judicata on decided issues)
Broad (all grounds raised or that reasonably could have been raised)
Strategic Primary Use
Full-scope invalidity attacks, inequitable conduct, enablement; final infringement and damages determination
Fast, cost-effective prior art attack; prior to district court trial with potential estoppel management
Key Takeaways: Part 4
U.S. pharmaceutical patent litigation is a multi-front operation. The 180-day first-filer incentive drives aggressive Paragraph IV timing. IPR’s lower burden, technical decision-makers, and high invalidation rates make it an almost mandatory tool for challengers — but estoppel risk demands rigorous petition construction. Coordinating IPR timing with district court proceedings to manage Fintiv risk is now a standard feature of U.S. challenge strategy.
Part 5: The European Frontier — EPO Oppositions and the Unified Patent Court
The EPO Opposition: Nine Months That Can Change Everything
The European Patent Office (EPO) opposition is a centralized administrative proceeding that allows any third party — with the explicit exception of the patent owner — to challenge the validity of a recently granted European patent. It is the most efficient first-strike mechanism in the European IP landscape.
The timing constraint is absolute: the opposition must be filed within nine months of the date on which the grant of the European patent is published in the European Patent Bulletin. This deadline is not extensible. Once the window closes, the EPO opposition pathway is permanently foreclosed for that patent.
The grounds for opposition under Article 100 EPC are limited to three categories: lack of patentability under Articles 52-57 (encompassing novelty and inventive step); insufficient disclosure under Article 83; and added subject matter under Article 123(2), which bars applicants from broadening their claims during prosecution beyond what was disclosed in the original filing.
The proceedings unfold before a three-member Opposition Division of the EPO. Written rounds are followed by oral proceedings — now frequently conducted by videoconference — in which parties and their counsel argue the case directly before the Opposition Division. The possible outcomes are revocation of the patent in full, maintenance in amended form with narrowed claims, or rejection of the opposition with the patent maintained as granted.
The statistical outcomes justify the nine-month urgency. In recent years, EPO opposition proceedings have resulted in approximately 38% of challenged patents being revoked in full and another 32% being forced into significant amendment. That 70% combined rate of outright revocation or meaningful narrowing represents exceptional return on investment relative to the official opposition fee of approximately €880.
Strategically, the EPO opposition has always served primarily as a long-term instrument for achieving central invalidation of a patent across all EPC member states through a single proceeding. The Unified Patent Court has added a second, more immediate function to the EPO opposition, which is analyzed in the next section.
The UPC: Pan-European Stakes and the New Preliminary Injunction Reality
The Unified Patent Court became operational on June 1, 2023 and represents the most significant structural reform to European patent litigation in a generation. It is an international court with jurisdiction across (currently) 18 EU member states, with the authority to issue rulings on infringement and validity that have effect across all participating countries simultaneously.
The UPC’s Court of First Instance operates through a central division (with seats in Paris, Milan, and Munich) and an expanding network of local and regional divisions across member states. Cases are heard by multinational panels that include both legally and technically qualified judges — a deliberate design choice that addresses the expertise deficit that has sometimes afflicted national patent courts.
The procedural design targets a first-instance decision within 12 to 14 months of filing. Whether that ambition is consistently achievable as caseload grows remains to be demonstrated, but the early evidence from the court’s first two years of operation is broadly positive.
The strategic implications run in two directions simultaneously.
For innovators: the UPC is a powerful enforcement tool. A single infringement action can produce a pan-European injunction covering all 18 member states, halting a competitor’s launch across the continent’s major markets in a single proceeding. Preliminary injunctions, which are notoriously difficult to obtain in U.S. courts, are a much more realistic remedy in the UPC, particularly for patents that have survived examination and opposition without significant amendments.
For challengers: the UPC centralizes downside risk to a degree that was impossible under the old fragmented national court system. A central revocation action filed at the UPC can — if successful — invalidate a European patent across all participating countries simultaneously. The ‘all-or-nothing’ character of the UPC’s jurisdiction makes defending a weak patent exponentially more perilous for innovators than it was when a challenger had to litigate country by country.
The German local divisions in Munich and Dusseldorf emerged quickly as the busiest UPC venues, drawing on Germany’s long-established expertise in pharmaceutical patent disputes. Other active divisions in the Netherlands, France, and Italy are developing their own jurisprudence, and the patterns of forum selection are still being established.
EPO-UPC Interplay: The Integrated Chess Board
The UPC has transformed the strategic meaning of an EPO opposition. This is most visible in the context of preliminary injunctions, and understanding the interplay is now a prerequisite for any European patent challenge strategy.
To grant a preliminary injunction, the UPC must assess the likely validity of the asserted patent. A pending EPO opposition — particularly one that has been filed with strong prior art and detailed claim charts — provides a defendant in UPC infringement proceedings with a concrete argument that the patent’s validity is genuinely in doubt. Early UPC cases including Alexion v. Amgen and 10x Genomics v. NanoString have demonstrated that UPC judges do not automatically stay their own proceedings because of a parallel EPO case, but they do weigh the existence and quality of an opposition when evaluating PI requests. A credible, well-argued opposition can shift the balance against granting an injunction.
This creates a new tactical calculus. For a biosimilar or generic company, filing a timely and substantive EPO opposition is no longer primarily about the long-term outcome of that opposition itself — though that outcome matters enormously. It is also a defensive move that can immediately affect the company’s litigation position in the UPC. A well-constructed EPO opposition, filed within the nine-month window, gives the company a document in the public record demonstrating that serious prior art exists against the patent. This document can be deployed within weeks or months in UPC proceedings.
Conversely, a patent that survives an EPO opposition without significant amendment enters UPC litigation with substantially enhanced credibility. The Opposition Division’s decision to maintain the patent as granted, after reviewing the challenger’s best prior art arguments, is powerful evidence of validity in the UPC’s preliminary injunction calculus.
The modern European strategy requires the EPO and UPC to be treated as a single integrated system rather than two separate proceedings. The arguments advanced, the prior art relied upon, and the claim amendments proposed in EPO proceedings all have direct consequences in concurrent or subsequent UPC litigation. Failure to coordinate these positions creates tactical contradictions that can be exploited by the other side.
Key Takeaways: Part 5
The nine-month EPO opposition deadline is one of the most time-sensitive and consequential decision points in European pharmaceutical IP. The UPC has elevated the strategic importance of that nine-month window by making a well-filed opposition a key defensive tool in UPC preliminary injunction proceedings. European patent strategy must now be built as an integrated EPO-UPC plan, not as two separate tracks.
Part 6: Evergreening Technology Roadmap
Evergreening is not one tactic — it is a system of overlapping IP strategies designed to maintain market exclusivity beyond the expiry of the primary composition-of-matter patent. Understanding the full taxonomy of evergreening mechanisms, and the specific invalidity arguments that can dismantle each one, is essential for both challengers and the policymakers and portfolio managers who assess the durability of any drug’s IP position.
The Taxonomy of Evergreening Mechanisms
Polymorph patents cover specific crystalline or amorphous solid-state forms of an active ingredient. Different polymorphs can have different solubility, stability, or bioavailability profiles, which can constitute genuine improvements but are also frequently challenged as inherently anticipated (if the polymorph would have been produced in prior art synthesis) or obvious (if polymorph screening was routine practice in the relevant therapeutic class at the time of filing).
Salt-form patents protect specific acid-addition or base-addition salts of a known compound. Salt selection is a standard pharmaceutical development step with established methodologies. Courts and patent offices have grown increasingly skeptical of salt-form patents that do not demonstrate unexpected improvements over known salts or free base forms.
Formulation patents cover specific dosage forms, drug delivery systems, controlled-release mechanisms, excipient combinations, or routes of administration. Extended-release formulations, for example, can extend exclusivity by several years and are particularly common in primary care categories. The invalidity arguments center primarily on whether the formulation choices were obvious design decisions given the state of drug delivery technology at the time.
Method-of-use patents claim the treatment of specific conditions with the drug, often conditions that were not in the original label. Pediatric exclusivity can add six months of exclusivity to existing patents by incentivizing clinical trials in pediatric populations. Combination drug patents cover co-formulation of two or more known agents, and are particularly vulnerable to obviousness challenges when both components were known and their combination had been discussed in the clinical literature.
Enantiomer patents, which cover the specific optically active isomer of a racemic mixture drug, are among the most controversial evergreening tools. The chiral switch — from a racemate approved in the 1990s or 2000s to a single-enantiomer product filed as a ‘new’ patent — has been the subject of significant litigation, particularly in Europe. The legal durability of enantiomer patents varies significantly by jurisdiction.
Invalidity Strategies Against Each Evergreening Mechanism
Polymorph and salt-form patents are attacked most effectively through inherent anticipation (showing the claimed form was produced in prior art syntheses) and enablement (showing that practitioners could not reliably produce the claimed form without extensive experimentation, or conversely, that the alleged improvement in dissolution or bioavailability was obvious to try given known formulation science).
Formulation patents fall primarily to obviousness challenges supported by prior art teaching standard controlled-release mechanisms, excipient combinations, or dosage optimization approaches. The ‘obvious to try’ doctrine has particular resonance here, because the number of known formulation approaches for any given molecule is typically finite and well-characterized.
Method-of-use patents face novelty challenges through inherent anticipation when the claimed use would necessarily result from prior art disclosures, and obviousness challenges when the clinical literature pointed toward the new indication before the patent was filed. Evidence that clinical investigators had already hypothesized or begun studying the claimed use is powerful prior art against a method-of-use patent.
Technology Roadmap: Where Evergreening Strategies Are Heading
Prodrug patents — covering inactive precursors that convert to the active drug in the body — are an area of increasing evergreening activity. They can offer genuine delivery advantages for drugs with poor absorption profiles, but are also used to extend exclusivity when the underlying active metabolite is itself in the public domain.
Biomarker-selected patient population patents, which claim the administration of a drug to a subset of patients identified by a genetic or proteomic biomarker, are particularly complex from an IP standpoint. They sit at the intersection of the obvious-to-try doctrine and §101 patentable subject matter constraints. When the clinical literature had identified the biomarker as predictive and the drug company’s claims simply formalize that correlation into a method of treatment, both obviousness and §101 challenges have merit.
Nanoformulation patents, claiming drug particles at specific nanometer-scale size ranges with corresponding improvements in bioavailability, are another active area. Their vulnerability depends heavily on whether the prior art taught nanoparticle formation for the specific drug class and whether the claimed particle size ranges and resulting bioavailability improvements were predictable from that prior art.
Key Takeaways: Part 6
Evergreening creates stacked layers of IP protection with sharply differentiated invalidity risk. Composition-of-matter claims carry the lowest risk; formulation, polymorph, and method-of-use claims carry substantially higher risk under post-KSR obviousness standards. Any financial model assigning full exclusivity value to a drug protected primarily by secondary patents is likely mispricing the IP risk.
Part 7: The Offensive Playbook — A Challenger’s Operational Guide
Step 1: Continuous Patent Monitoring and Target Prioritization
An effective invalidity campaign begins not at the petition stage but years earlier, during the continuous monitoring of competitor patent portfolios. Tools including DrugPatentWatch provide real-time tracking of patent grants, Orange Book listings, continuation applications, and prosecution history documents, enabling challengers to identify high-value targets at the earliest possible moment.
Target prioritization should begin with commercial potential. The return on a successful invalidity campaign is a direct function of the revenue protected by the patent being challenged. Drugs with annual U.S. revenues above $1 billion are natural targets, and the 180-day first-filer prize makes the first Paragraph IV applicant’s economics particularly attractive for the highest-grossing drugs. The patent monitoring operation should feed a live commercial opportunity dashboard, scoring each potential target on market size, patent strength, and challenger readiness.
Readiness assessment should include: whether the COM patent has already expired (making the challenge a secondary patent attack, with different risk profile); whether other generics have already filed ANDAs (affecting first-filer eligibility); whether relevant prior art has already been cited in prosecution (potentially strengthening the case that the examiner considered but distinguished the best available arguments); and whether any related patents have been previously litigated (creating both precedent and estoppel considerations).
Step 2: The Nullity Search — How to Find What the Examiner Missed
The nullity search is the technical foundation of any invalidity challenge. Its goal is to find prior art references that either single-handedly anticipate the asserted claims or, collectively, render them obvious under the post-KSR framework. The examiner who granted the patent performed a search, but that search had resource constraints, time limits, and database access limitations. A well-resourced nullity search — conducted with no time constraints, access to global databases, and domain experts with deep knowledge of the relevant scientific literature — can find what the examiner missed.
A world-class nullity search for a pharmaceutical patent requires searching across patent databases (USPTO, EPO Espacenet, WIPO PATENTSCOPE, national patent offices in Japan, China, South Korea, and elsewhere), non-patent literature databases (PubMed, Web of Science, Embase, SciFinder, Reaxys), conference proceedings and abstract archives, academic dissertations and theses, FDA drug approval documents and clinical trial registries, and technical and trade publications from relevant scientific fields.
AI-powered search tools have substantially enhanced the efficiency and coverage of nullity searches. These tools can perform semantic search across multilingual databases, identify conceptual connections between structurally distinct prior art references, and flag obscure non-patent literature that keyword-based searches would miss. The practical result is a larger and richer prior art universe for any given patent, with meaningful improvement in the probability of finding a highly relevant reference.
The nullity search is not complete until the search team has also surveyed the patent’s prosecution history. The examiner’s rejections and the applicant’s responses reveal which prior art was already considered and how the claims were amended or argued around. This history is a map of the patent’s known vulnerabilities and its surviving defenses. It tells challengers which arguments will be foreclosed (because the applicant successfully distinguished the same references) and which new references — not cited in prosecution — might be more powerful because the applicant has had no opportunity to craft arguments against them.
Step 3: Building the Claim Charts and Selecting the Forum
With prior art in hand, the next step is constructing detailed claim charts that map the elements of each asserted claim to the disclosures in the prior art references. A claim chart is a formal document that will become the basis for invalidity contentions in district court or the petition in an IPR. Its quality — the precision of the element-by-element mapping and the cogency of the obviousness narrative linking multiple references — directly determines the strength of the legal argument.
The forum selection decision requires weighing all of the considerations detailed in Part 4. For a U.S. challenge, the standard strategy is parallel attack: file the Paragraph IV ANDA to trigger Hatch-Waxman litigation and simultaneously prepare an IPR petition targeting the patent’s composition-of-matter or secondary claims on prior art grounds. Time the IPR petition to be filed early enough to manage Fintiv risk. Identify which invalidity arguments will be deployed at PTAB vs. reserved for district court, to avoid estoppel exposure from a failed IPR while preserving full invalidity options in court.
For a European challenge, assess whether the nine-month EPO opposition window is open. If it is, filing a strong opposition is nearly always the right decision given the cost-effectiveness and the growing importance of opposition in UPC preliminary injunction proceedings. If the window has closed, assess whether the patent is held in unitary patent form (subjecting it to UPC jurisdiction) or as a bundle of national patents (requiring national litigation or a UPC central revocation action for validated countries).
Investment Strategy Note: Challengers
For investment professionals tracking generic and biosimilar companies, the quality and timing of Paragraph IV filings against high-revenue branded drugs are material indicators of a company’s pipeline value. First-filer status on a $3 billion drug creates an expected value in excess of $500 million for the first-filer’s 180-day exclusivity period, before any royalty-bearing license. Companies with strong patent challenge capabilities — large nullity search teams, experienced PTAB litigators, and productive relationships with academic prior art experts — carry a pipeline value premium that is systematically underestimated by analysts who focus primarily on ANDA approval counts.
Part 8: The Defensive Playbook — How Innovators Fortify Patents Under Fire
Step 1: Robust Patent Prosecution as the First Line of Defense
The best defense against an invalidity challenge is a patent that was drafted, prosecuted, and strategically structured to withstand one. In the post-Amgen, post-KSR environment, this means:
The specification must be comprehensive. Following the Amgen ruling, a biologic patent that relies on a small number of exemplified species to support a broad functional genus claim is vulnerable from the moment it is granted. Innovators must invest in generating and disclosing a large number of working examples — not as a scientific formality but as a legal necessity. For antibody programs, this means characterizing and disclosing dozens of antibody structures with confirmed functional activity. For small molecule programs targeting a structural class, it means providing synthesis and activity data for a representative cross-section of the claimed chemical space.
Claim architecture should be layered from broad to narrow. The broadest claims provide the widest protection if they survive challenge; the narrower claims act as fallback positions if the broadest claims fall. A portfolio that includes COM claims, sub-genus claims, species claims, formulation claims, and method-of-use claims provides multiple layers of protection, with each layer independently defensible.
Continuation applications are the most underutilized defensive tool in pharmaceutical prosecution. By filing continuations before the parent patent issues, an innovator keeps the prosecution process open and retains the ability to file new claims tailored to the specific products competitors develop, to the specific arguments raised in litigation, or to the specific guidance provided by court decisions on claim scope. A continuation filed today can be a critical defensive asset in litigation five years from now.
Secondary considerations — the objective real-world evidence that an invention was not obvious — are most powerful when they are built systematically over time rather than assembled reactively during litigation. The required components:
Nexus documentation must begin at launch. The company needs evidence, gathered prospectively, establishing that the commercial success of its product is attributable to the specifically novel and non-obvious features of the patent, not to brand recognition, marketing spend, formulary positioning, or prior-art features of the drug. This is a challenging evidentiary burden, and it requires deliberate planning. Launch-phase market research, clinician survey data, and payer decision analysis can all contribute to a nexus record that will be substantially more credible than retrospective reconstructions in litigation.
Failure of others documentation requires identifying who was working on the problem, what they tried, and why they failed. For genuinely novel mechanisms or unexpectedly potent molecular series, contemporaneous evidence of competitor programs that did not reach clinical success is powerful support for non-obviousness. This evidence often exists in the scientific literature, in conference presentations, or in competitor regulatory filings, and it should be identified and preserved as part of routine competitive intelligence.
Industry recognition — peer-reviewed publications, clinical guideline endorsements, medical society awards, or technical prizes — provides third-party corroboration that the invention was viewed as a genuine advance by practitioners in the field at the time of its disclosure.
Step 3: Active Litigation Management
When a challenge materializes, the defensive strategy must be actively managed across all active proceedings simultaneously.
Claim amendments in EPO opposition proceedings and U.S. continuation applications allow the patent holder to narrow the claims to carve out the prior art cited by the challenger, preserving a more limited but defensible scope. The decision to amend must be made carefully, weighing the commercial value of the narrowed claim scope against the risk of maintaining the broader claims and losing the patent entirely.
Expert witness preparation is among the most consequential investments in patent litigation defense. The POSITA standard means that the court’s understanding of what was known and predictable in the field at the time of filing is heavily influenced by competing expert testimony. An expert who can clearly articulate why the claimed invention was not an obvious design choice, and who can credibly distinguish the prior art from the claimed invention, is a decisive advantage. Finding and preparing that expert should begin early.
Cross-forum coordination requires the legal team to ensure that positions taken in EPO proceedings, PTAB proceedings, and district court are not in conflict. Inconsistent claim construction positions or contradictory statements about the scope of prior art can be exploited aggressively by challengers. The patent holder’s legal strategy must present a unified and consistent narrative of the invention’s novelty and non-obviousness across all forums.
Investment Strategy Note: Innovators
For investors and analysts evaluating branded pharmaceutical companies, the quality of a company’s patent prosecution practices and its ongoing relationship with its IP portfolio are forward-looking indicators of franchise durability. Companies that file thin specifications, neglect continuation strategies, and fail to build secondary consideration records are accumulating IP risk that will not become visible until a Paragraph IV is filed or an IPR petition lands. Analysts conducting due diligence on branded pharmaceutical companies should evaluate not just the nominal patent expiry dates but the structural quality of the underlying patents, the diversity of the patent portfolio for each major asset, and the company’s history of success or failure in patent litigation.
Part 9: High-Profile Case Studies
Amgen Inc. v. Sanofi (2023): Functional Genus Claims and the New Enablement Standard
The central facts of this dispute are detailed in Part 2. The business consequences warrant separate attention here.
Amgen’s Repatha (evolocumab) reached peak annual revenues of approximately $1.3 billion in 2023. The patent litigation with Sanofi over competing PCSK9 antibodies Repatha and Praluent (alirocumab) lasted nearly a decade and involved multiple rounds of trial and appeal. The Supreme Court’s final ruling in Sanofi’s favor did not immediately destroy Amgen’s market position — both drugs remain commercially available — but it stripped Amgen of the ability to exclude Sanofi through broad genus patent claims, fundamentally altering the competitive dynamics of the PCSK9 inhibitor market.
The decision’s IP valuation implication is systematic: any biologic patent that was drafted and filed using the pre-Amgen strategy of broad functional genus claims plus a small number of exemplified species should be reassessed for enablement vulnerability. This includes a substantial number of antibody patents currently protecting high-revenue biologic drugs. Companies with concentrated exposure to this type of patent structure face elevated risk of biosimilar entry or licensing pressure.
For biosimilar challengers: Amgen created a clear doctrinal path for challenging biologic patents with broad functional claims. The first question in any biologic invalidity assessment is now whether the patent’s disclosed examples are sufficient to enable the full scope of what it claims — and in a large fraction of existing biologic patents filed under the pre-2023 practice, the honest answer is ‘no.’
Gilead Sciences v. Merck & Co.: When Conduct Overrides Validity
Merck’s Hepatitis C patent litigation against Gilead produced one of the most studied reversals in modern pharmaceutical patent litigation history. In March 2016, a jury found that Merck’s patents were valid and infringed by Gilead’s sofosbuvir-based HCV drugs Sovaldi and Harvoni, awarding $200 million in damages. The award was vacated by the district court following a bench trial on equitable defenses and affirmed by the Federal Circuit in 2018.
The factual basis of the unclean hands finding centered on a 2003 confidential due diligence call between Merck and Pharmasset (later acquired by Gilead as the source of sofosbuvir). The call was conducted under a firewall agreement that explicitly excluded Merck patent prosecution personnel. A key Merck in-house patent attorney with active HCV prosecution responsibilities nonetheless participated in the call, obtained confidential information about Pharmasset’s lead compound, and used that information to amend Merck’s pending claims to better cover the molecule. The court found that the same attorney then provided ‘inconsistent, contradictory, and untruthful’ testimony about his participation in the call.
The ‘unclean hands’ doctrine is broader than inequitable conduct before the USPTO. It applies to any serious misconduct — including pre-litigation business conduct and litigation misconduct — that has a direct nexus to the subject of the lawsuit. Unlike inequitable conduct, which renders a patent unenforceable against the world, unclean hands is a personal defense that operates against the specific party who engaged in the misconduct. But in practice, the effect was the same: Merck could not enforce the $200 million verdict.
The operational lesson for all companies engaging in partnered research, due diligence processes, or collaborative clinical development is unambiguous. Firewall agreements protecting patent prosecution personnel from exposure to third-party confidential information are not administrative formalities — they are legally essential risk management structures. A single breach, compounded by litigation misconduct, can extinguish a favorable jury verdict and expose the company to attorney fee awards in the other direction.
Xarelto (Rivaroxaban): A Lesson in Jurisdictional Fragmentation
The multi-jurisdictional patent litigation over Bayer’s Xarelto (rivaroxaban) is a textbook study in how the same patent, evaluated under broadly similar legal standards by technically sophisticated courts, can produce opposite outcomes in different countries.
EP 1 845 961, covering rivaroxaban’s use in treating thromboembolic disorders, faced validity challenges from generic manufacturers including Sandoz, Stada, and Teva across numerous European markets as Xarelto’s patent life wound down. The results were contradictory. Courts in the Netherlands, Austria, Sweden, Lithuania, and Belgium upheld the patent. The German Federal Patent Court revoked it for lack of inventive step. UK, French, Norwegian, Swiss, Australian, and South African courts also found it invalid. The EPO Opposition Division initially revoked the patent before the Boards of Appeal reinstated it, after which national courts continued to reach conflicting conclusions.
The Xarelto experience illustrates the strategic limitation of relying on a single patent claim or a single national victory as the basis for pan-European exclusivity. A win in one jurisdiction provides no guarantee of protection in adjacent markets, and a loss in Germany — Europe’s largest pharmaceutical market — can cascade into commercial exposure that cannot be fully compensated by favorable rulings elsewhere.
The UPC was designed precisely to eliminate this fragmentation for future disputes. A patent validated as a unitary patent under the new system could be revoked in a single UPC central revocation action or enforced through a single UPC infringement action. Whether this centralization ultimately benefits innovators or challengers will depend on the court’s developing jurisprudence, but it fundamentally changes the risk profile of European patent litigation relative to the Xarelto era.
Part 10: AI-Assisted Drug Discovery — The New Validity Flashpoints
Artificial intelligence has moved from a research novelty to a core drug discovery infrastructure tool. AI platforms — including generative molecular design systems, protein structure prediction models, and ADMET property predictors — are now used routinely in compound selection, target identification, lead optimization, and formulation development at most major pharma and biotech companies. This integration has created a new set of patent validity questions that the legal system is still actively resolving.
AI as a Challenger’s Tool: Enhanced Prior Art Searches and the Rising Non-Obviousness Bar
For challengers, AI provides two distinct advantages in patent validity campaigns. The first is operational: AI-powered prior art search tools dramatically increase the coverage and efficiency of nullity searches. These tools can query global databases in multiple languages simultaneously, identify semantic relationships between conceptually related but terminologically distinct prior art references, and surface obscure non-patent literature that keyword-based search strategies would miss.
The second advantage is more structural and longer-term. The non-obviousness standard is pegged to the level of skill possessed by a hypothetical POSITA in the relevant field. As AI tools become standard equipment for practitioners in medicinal chemistry, computational biology, and formulation science, the capabilities of that hypothetical practitioner are effectively being elevated by the tools available to the field. What would have required genuine inventive insight from a human medicinal chemist five years ago may now be a routine AI-generated output. Courts and patent offices have not yet fully grappled with how widely deployed AI tools affect the POSITA’s implied capabilities, but the trajectory is clear: the non-obviousness bar will rise as AI becomes more deeply integrated into standard scientific practice.
AI as an Innovator’s Liability: The Inventorship Exposure
The most acute legal risk created by AI-assisted drug discovery is the inventorship problem. U.S. patent law requires inventors to be natural persons. This principle was reaffirmed by the Federal Circuit in Thaler v. Vidal (2022), which held that an AI system cannot be listed as a patent inventor. The same principle applies in virtually every major patent jurisdiction globally.
The legal risk for companies using AI in drug discovery arises when the AI’s contribution to the invention is so substantial that it is difficult to identify a human contributor who made the required ‘significant contribution to the conception’ of the claimed invention. If a company uses a generative AI system to design a novel molecular scaffold, and the human scientists’ role was limited to inputting parameters, selecting outputs, and funding the process, a challenger may have grounds to argue that the patent lacks valid human inventorship.
This is not a theoretical future risk. It is a present, actionable invalidity ground that challengers are already beginning to identify in AI-assisted discovery programs where the company’s public disclosures reveal an AI-centric invention narrative without the corresponding human inventive contribution. Companies must document human inventive contribution contemporaneously and granularly: the specific decisions made by human scientists to define the problem, select the AI approach, evaluate outputs, design the experiments that validated or refined the AI’s suggestions, and exercise scientific judgment at each decision point in the process.
AI-Generated Prior Art: The Minefield Ahead
AI systems designed specifically to generate prior art are a growing phenomenon. Services that use large language models to produce vast libraries of scientific disclosures — effectively flooding the prior art landscape with machine-generated content — raise complex questions about what constitutes legally qualifying prior art. Is a disclosure that no human has reviewed or validated ‘publicly accessible’ in the sense required for prior art? Does it ‘enable’ a POSITA to practice what it describes? These questions do not yet have settled legal answers, but the volume of AI-generated content entering the public domain is growing, and its impact on pharmaceutical patent prosecution and litigation will need to be addressed by courts and patent offices in the near term.
Key Takeaways: Part 10
AI integration into drug discovery creates a parallel set of IP risks and opportunities. Challengers gain enhanced prior art search capabilities and a structural argument that AI-era POSITA capabilities are higher than pre-AI standards assumed. Innovators face new inventorship documentation burdens and must plan for a litigation environment in which AI-generated prior art is a growing feature of the prior art landscape. Companies that have not yet developed formal AI inventorship documentation protocols are accumulating future invalidity exposure with each AI-assisted discovery program.
Part 11: Investment Strategy for Portfolio Managers and Analysts
How to Quantify Patent Invalidity Risk in a Pharma Portfolio
The standard financial model for a branded pharmaceutical drug assigns patent-protected revenue for the full period between current date and the nominal expiry of the latest-listed Orange Book patent, then applies a generic entry discount at expiry. This model has two systematic errors: it ignores the probability of successful Paragraph IV challenge before expiry, and it treats all Orange Book-listed patents as equally durable.
A more accurate model incorporates the following inputs:
Revenue concentration analysis identifies what fraction of the drug’s U.S. revenue is protected by each category of patent (COM, polymorph, formulation, method-of-use). Higher concentration in secondary patents raises the invalidity probability for the effective exclusivity period.
Paragraph IV risk scoring uses the drug’s annual U.S. revenue as the primary predictor of challenge likelihood, combined with the time since NDA approval (challenge probability rises in the first five years post-approval for high-revenue drugs), the composition of the patent portfolio, and any public statements by generic companies about their development timelines.
PTAB invalidation probability, for drugs already facing IPR petitions, is estimated by analyzing the quality of the petitioner’s prior art (assessed by prior art citation overlap with the patent’s prosecution history), the breadth of the asserted claims relative to the disclosed examples, and the petitioner’s track record in PTAB proceedings against similar patents.
Litigation outcome history for the specific patent or patent family provides direct precedent. A patent that has already survived two district court validity challenges carries a different risk profile than one that has never been litigated.
Reading the IP Position of Specific Assets
Biosimilar opportunity assessment: for any major biologic drug with annual global revenues above $2 billion, an IP analyst should map the full patent portfolio, assess the enablement vulnerability of any broad functional genus claims in the context of the post-Amgen standard, and identify whether the biosimilar developer’s development timeline is likely to produce enough data to support a strong enablement challenge before the relevant patents expire. Biosimilar companies with strong IPR capability and domain expertise in biologic patent law are better positioned to accelerate market entry through active validity challenges than those relying exclusively on expiry-driven timing.
Pipeline licensing and M&A due diligence: when evaluating assets in a transaction, IP due diligence should include a formal invalidity risk assessment for each core patent in the portfolio. This includes a sampling nullity search against the two to four most important claims in each key patent, a review of prosecution history for any weaknesses in the written description or enablement record, and an assessment of the secondary consideration evidence available to support non-obviousness. Assets with thin specifications, broad functional claims, and weak secondary consideration records should be priced to reflect their elevated invalidity exposure.
Signaling Events to Monitor
Paragraph IV certification filings against a branded drug are public events, detectable through FDA ANDA database monitoring and court docket tracking. The timing and identity of the first Paragraph IV filer are material signals: a first Paragraph IV from a sophisticated generic company with strong IPR capability against a core patent rather than a secondary one signals a high-probability challenge to fundamental exclusivity.
IPR petition filings at the PTAB are public and searchable through USPTO’s Patent Center. A well-constructed petition with strong prior art is itself a signal that the petitioner’s legal team has found the patent’s vulnerabilities. Monitoring IPR petition quality is an underutilized source of competitive intelligence for both pharma companies and investors.
EPO opposition filings against European patents protecting high-value drugs are published in the EPO’s online opposition register within weeks of filing. A detailed opposition with strong prior art filed within the nine-month window by a credible opponent is an early warning signal for European exclusivity risk.
Part 12: Key Takeaways by Segment
For IP Teams and Patent Counsel
The post-Amgen enablement standard requires a fundamental reassessment of biologic patent prosecution strategy. Functional genus claims drafted under pre-2023 assumptions are structurally vulnerable. Continuation practice should be aggressive and systematic, building each new layer of claims on an expanded experimental record.
Secondary consideration documentation should be treated as a programmatic function of the commercial team, not a reactive litigation task. Nexus evidence assembled over years of market experience is categorically more credible than evidence assembled in the months before trial.
European patent strategy must now integrate EPO opposition planning with UPC litigation positioning as a single coordinated exercise. The nine-month EPO opposition deadline is a hard deadline with strategic consequences that extend beyond the opposition itself.
For R&D and Business Development
Inventorship documentation for AI-assisted discovery programs must be established as a formal standard operating procedure before those programs generate assets with patent potential. The window to document human inventive contribution contemporaneously closes as the program advances; retrospective reconstruction is significantly weaker.
Partnership and due diligence firewall procedures must be enforced at the operational level, not just the policy level. The Gilead v. Merck fact pattern — a single attorney participating in a firewalled call — is a risk that exists in any collaborative research environment. The consequences of a breach can survive a successful jury trial.
For Portfolio Managers and Institutional Investors
Patent invalidity risk is a systematic pricing factor that is underweighted in standard pharmaceutical equity models. Secondary patent portfolios protecting high-revenue drugs should be assessed against the published PTAB invalidation statistics and Paragraph IV challenge likelihood framework. Biosimilar companies with strong PTAB litigation capability represent an underappreciated source of premium over peers relying exclusively on expiry-driven market entry.
The UPC’s first two years of operation are generating a body of precedent that will materially affect European exclusivity assumptions for drugs across multiple therapeutic categories. Monitoring early UPC preliminary injunction decisions for signals about the court’s validity assessment standards is now a relevant input to European exclusivity modeling.
Part 13: Frequently Asked Questions
After Amgen v. Sanofi, can a biologic company still secure broad genus claims?
Yes, but the required disclosure to support those claims is substantially heavier than pre-2023 practice assumed. A broad functional genus claim covering a class of antibodies or other biologics requires enabling a POSITA to make and use the full scope of that class without undue experimentation. The practical implication is that a company seeking to file and maintain such claims must invest in generating and disclosing a large number of representative species with confirmed functional activity, not a handful of examples plus a screening methodology. Narrow species and sub-genus claims, layered underneath any broader genus claim, provide insurance when the genus claim is challenged.
Is it always strategically correct to file an IPR when challenging a U.S. drug patent?
The high invalidation rates at the PTAB make IPR an attractive option in most cases, but several specific circumstances argue against it or require careful management. IPR is unavailable for enablement, written description, inequitable conduct, and §101 challenges — if these grounds are the strongest available, district court is the only forum. The Fintiv risk, which can result in discretionary denial if the parallel district court case is too far advanced, must be managed through early petition filing. The estoppel provision, which can cripple the district court invalidity case if the IPR fails, requires that only the strongest prior art arguments be advanced in the petition. A dual-track strategy — IPR for prior art grounds, district court for all remaining invalidity theories — is the standard approach for high-value targets.
How has the UPC changed the strategic calculus for filing an EPO opposition?
The EPO opposition now serves a dual function. Its long-standing function — centrally challenging a European patent’s validity through a cost-effective administrative proceeding — is unchanged. Its new function, created by the UPC’s preliminary injunction practice, is to place a credible public challenge on record that can be deployed in UPC infringement proceedings to resist a PI request. Early UPC cases confirm that the court weighs the existence and quality of pending EPO oppositions when evaluating preliminary injunctions. A well-constructed opposition, filed within the nine-month window, creates an immediate tactical asset in UPC litigation. Companies that treat the nine-month EPO deadline as primarily a long-term validity question are undervaluing this second function.
What documentation must companies generate to protect AI-assisted drug discoveries from inventorship challenges?
The minimum required documentation establishes, contemporaneously and specifically: the human-defined problem or objective that guided the AI’s deployment; the specific parameters, constraints, or design criteria that human scientists set for the AI; the basis on which human scientists selected specific AI-generated outputs for further study over the alternatives the AI produced; the additional experimental design and validation work performed by humans to test, modify, or refine the AI’s suggestions; and the scientific judgments made by humans throughout the process that shaped the final claimed invention. The standard lab notebook that documented pre-AI experimental work needs to be adapted to capture these AI-specific decision points. Without this record, an inventorship challenge based on the absence of identifiable human conception becomes substantially more credible.
What is the difference between inequitable conduct and unclean hands, and which is more dangerous in pharmaceutical patent litigation?
Inequitable conduct is a defense specific to patent law, requiring proof that the patent applicant intentionally withheld material prior art from, or deliberately misled, the USPTO during prosecution. When proven, it renders the patent permanently unenforceable against the world. The Therasense standard for proving inequitable conduct is demanding — requiring both materiality and a specific intent to deceive, without inferring intent from materiality alone — which makes successful inequitable conduct findings relatively rare.
Unclean hands is a broader equitable defense applicable when either party engages in serious misconduct that has a direct relationship to the subject of the lawsuit, regardless of whether that misconduct occurred before the patent office. The Gilead v. Merck outcome demonstrates its potential: conduct that began with a firewall breach in a due diligence call and was compounded by false testimony in litigation was sufficient to vacate a $200 million jury verdict. Unlike inequitable conduct, unclean hands is technically a ‘personal’ defense that operates against the specific wrongdoer rather than rendering the patent itself invalid — but in practice, when it succeeds, the patent holder loses all enforcement rights against the defendant who prevailed on the defense.
This guide is intended for informational purposes for pharmaceutical IP professionals, portfolio managers, and business decision-makers. It does not constitute legal advice. Specific invalidity strategy should be developed with qualified patent counsel familiar with the relevant jurisdiction and the specific patent at issue.
Data on PTAB invalidation rates are drawn from PTAB statistics published through 2024-2025. Hatch-Waxman challenge frequency data reflects published academic research on ANDA filing patterns. UPC case references reflect public decisions available through the UPC’s online case management system.
