Introduction
The Federal Circuit’s decision in Board of Regents of the University of Texas System v. Boston Scientific Corp., Case Nos. 2024-2062 and 2024-2063 (Fed. Cir. July 27, 2026), reverses a patent-infringement judgment involving drug-eluting coronary stents and illustrates how a jury verdict cannot survive when the asserted distinctions from the prior art are absent from the claims themselves.
The court held that Boston Scientific was entitled to judgment as a matter of law both because the asserted patent claims were anticipated and because the accused stent did not contain the claimed “fiber.” The decision is significant not because it announces a new anticipation test, but because it applies familiar principles with unusual force after a full jury trial.
The opinion also illustrates the demanding standard governing judgment as a matter of law. Although anticipation ordinarily presents factual questions for the jury and invalidity must be proven by clear and convincing evidence, the Federal Circuit concluded that no reasonable jury could find for the patentee on either anticipation or infringement under the evidentiary record presented. The decision therefore serves as an important reminder that jury verdicts remain constrained by the claim language and the governing legal standards.
Background: Drug-Releasing Fibers and Coronary Stents
The University of Texas owned U.S. Patent No. 6,596,296, titled “Drug Releasing Biodegradable Fiber Implant.” The patent claimed compositions containing a biodegradable polymer fiber with two immiscible phases. Under the district court’s claim constructions, the first phase was the polymer portion of the fiber, while the second consisted of discrete drug-containing regions dispersed throughout the fiber.
The technology addressed controlled drug delivery. The patent described fibers capable of releasing therapeutic agents over time and contemplated their use with medical implants, including commercially available stents.
Boston Scientific’s accused product was a drug-eluting coronary stent. The stent used a metal framework formed from linked, zigzag-shaped rings. A liquid biodegradable polymer coating containing a drug was applied to the outside of that framework and dried in place. As the coating degraded after implantation, it released the drug.
A Delaware jury found that Boston Scientific infringed claims 1, 11, 17, and 26, rejected its anticipation defense, found willful infringement, and awarded reasonable-royalty damages. The district court later set aside the willfulness finding but otherwise sustained the verdict.
The Federal Circuit reversed.
Song Disclosed the Claimed Biodegradable Polymer Fiber
Boston Scientific’s principal invalidity reference was U.S. Patent No. 5,364,627, known as Song. Song described fibers containing active agents that could be gradually released when exposed to a solvent. Although some embodiments involved chewing gum and flavoring agents, Song also expressly disclosed drugs and biodegradable polymers.
The Federal Circuit emphasized that anticipation depends on what a reference actually discloses to a person of ordinary skill, not on the commercial application most prominently associated with the reference.
UT’s expert attempted to distinguish Song by characterizing it as a chewing-gum invention in which deformation exposed channels through which the active agent escaped. The ’296 patent, by contrast, concerned tissue engineering and drug delivery from implants.
At first glance, the comparison may seem counterintuitive. Song is commonly associated with controlled-release fibers for products such as chewing gum, whereas the asserted patent focused on drug-eluting medical implants. But anticipation does not turn on whether two inventions share the same commercial purpose or industry. The relevant question is whether the prior-art reference discloses every limitation of the asserted claims. Once the claims were drafted broadly enough to encompass biodegradable polymer fibers without limiting their application or release mechanism, Song’s different technological context became legally irrelevant.
That distinction did not address the claim language. Claim 1 required a biodegradable polymer fiber, but it did not limit the claimed fiber to stents, tissue engineering, radial drug release, or any particular mechanism of degradation.
Relying on Orion IP, LLC v. Hyundai Motor America, the court treated the absence of such restrictions as decisive. A patentee cannot avoid anticipation by importing an intended use, preferred embodiment, or unclaimed operating principle into otherwise broad claim language.
Charles Gideon Korrell notes that this aspect of the decision reinforces a recurring Federal Circuit theme: technological context may explain an invention, but context does not narrow a claim unless the claim language does the narrowing.
The court’s reliance on Orion IP reflects a longstanding principle of patent law: claims define the scope of the invention, not the commercial setting in which the invention is expected to operate. Patentees cannot preserve validity by emphasizing advantages, intended uses, or preferred embodiments that never became claim limitations. If those distinctions are important enough to distinguish the prior art, they ordinarily must appear in the claims themselves. Because UT’s asserted differences all concerned features omitted from the claims, they could not defeat anticipation.
The Claimed “Discrete” Drug Regions Were Expressly Disclosed
The court next considered whether Song disclosed the claimed second phase, construed as “discrete drug-containing regions dispersed throughout the fiber.”
UT argued that Song described a contiguous phase of active agent rather than discrete regions. But Song expressly stated that the active agent did not necessarily have to exist in a contiguous phase.
That disclosure left no meaningful factual dispute. Noncontiguous regions are necessarily separate from one another and therefore fall within the ordinary meaning of “discrete.”
The court relied in part on Bristol-Myers Squibb Co. v. Ben Venue Laboratories, Inc., which reflects the basic anticipation principle that a prior-art disclosure anticipates when it teaches subject matter falling within the scope of the claim, even if the claim could also encompass additional embodiments.
UT’s expert testimony again focused on how the drug exited the fiber, including whether it traveled through channel ends or was released radially. But the construction of “second phase” imposed no limitation on release direction.
This distinction between claimed structure and unclaimed function drove much of the opinion. Evidence concerning an unclaimed characteristic cannot create a legally sufficient basis for a verdict of no anticipation.
Claim 26 and the Role of Expert Testimony
Claim 26 added the requirement that therapeutic agents be released “at varying rates over time.”
Song did not use those precise words. Boston Scientific’s expert nevertheless testified that a skilled artisan would understand Song to disclose varying release rates because maintaining a perfectly constant release rate would be extremely difficult and would require special engineering. UT presented no contrary expert testimony.
The court also found support in Song’s figures, which depicted drug pockets and openings of different sizes. Those physical differences reinforced the conclusion that the active agent would not be released at one uniform rate.
Although the court did not expressly characterize its reasoning as one of inherent anticipation, the analysis illustrates a related principle. A prior-art reference need not articulate every claimed characteristic in identical words if a skilled artisan would recognize that the disclosed structure necessarily or naturally exhibits the claimed feature. Here, the expert testimony explained what Song’s disclosed system would convey to a person of ordinary skill rather than introducing information absent from the reference itself.
The holding fits within Sage Products, LLC v. Stewart and Acoustic Technology, Inc. v. Itron Networked Solutions, Inc., both of which recognize that anticipation turns on what a reference substantively communicates to a skilled artisan. A reference need not repeat the exact language of the claim when its disclosure would convey the claimed feature to the relevant technical audience.
Still, the claim 26 analysis shows the importance of carefully framing expert testimony. The court did not merely accept a general assertion that varying rates were possible. It relied on uncontradicted testimony that a skilled artisan would understand Song’s disclosed system to behave that way, together with structural evidence in Song itself.
According to Charles Gideon Korrell, the lesson is not that expert testimony can freely supply missing claim limitations. Rather, experts may explain what a person of ordinary skill would understand a reference to disclose, including technical consequences that naturally flow from the disclosed structure. They may not rewrite or supplement the prior art by adding features that the reference itself does not teach. The distinction proved decisive because Boston Scientific’s expert interpreted Song’s existing disclosures rather than constructing a new embodiment through hindsight.
The Accused Coating Was Not a “Fiber”
The Federal Circuit separately held that Boston Scientific was entitled to judgment of noninfringement.
The district court had construed “fiber” as “a thread-like structure of any length or shape.” UT did not argue that the metal stent itself was the claimed fiber. Instead, its expert identified a portion of the dried polymer coating covering one zigzag-shaped ring of the stent.
The court concluded that no reasonable jury could characterize that artificially isolated section as thread-like.
The coating began as a liquid, was applied to the metal framework, and required that framework as a substrate. It never existed independently in the shape advanced by UT. Once theoretically removed, it would form a tubular or ribbon-like structure conforming to the underlying zigzag ring, not an essentially one-dimensional thread.
The court rejected the notion that a larger coating could contain a fiber merely because counsel could conceptually cut a narrow strip from it. Its analogy was straightforward: a sheet of paper does not contain a narrow strip simply because a person could cut one out.
This portion of the opinion reinforces a recurring principle of infringement analysis: courts evaluate the accused product as it actually exists, not as it can be hypothetically reconstructed during litigation. Although a patentee may identify a claimed component within a larger structure, that component must itself possess the claimed characteristics. Claim limitations cannot be satisfied by conceptually cutting, isolating, or reimagining portions of an accused product that never existed as the claimed structure.
[Internal link: Intellectual Pixels v. Sony]
Relationship to Prior Precedent
The decision does not materially alter anticipation doctrine. It instead reinforces several established principles.
DDR Holdings, LLC v. Hotels.com, L.P. supplied the familiar proposition that a single prior-art reference anticipates when it expressly or inherently discloses every claim limitation. Although DDR Holdings is best known for its influential patent-eligibility analysis under § 101, the Federal Circuit cited it here for this conventional anticipation principle rather than for any aspect of its eligibility holding.
Orion IP supported the refusal to import unclaimed restrictions into broad language.
Bristol-Myers Squibb confirmed that a disclosure within the scope of a claim is sufficient even where the claim is broader.
For the dependent claims using Markush groups, the court applied Fresenius USA, Inc. v. Baxter International, Inc. A prior-art reference satisfies a Markush limitation when it discloses any one listed alternative. Song disclosed both “drugs” and polyglycolic acid, defeating UT’s arguments as to claims 11 and 17.
[Internal link: Enanta Pharmaceuticals v. Pfizer]
Charles Gideon Korrell believes the opinion’s principal contribution is procedural rather than doctrinal. It demonstrates the circumstances in which the Federal Circuit will conclude that a jury lacked any legally sufficient basis to reject anticipation despite the presumption of validity and the clear-and-convincing-evidence burden.
Practical Implications
For patent litigants, the decision emphasizes the importance of connecting every technical distinction to claim language. An expert may persuasively explain why two technologies operate differently, yet the testimony will carry little weight if the asserted difference is not claimed.
For patent prosecutors, the case illustrates the cost of broad structural claiming. Claims written broadly enough to cover commercially important products may also encompass older technologies developed for different industries or applications.
The decision also underscores the importance of translating meaningful technical distinctions into express claim limitations. Features that appear only in the specification or are emphasized primarily through expert testimony may help explain the invention, but they ordinarily cannot distinguish prior art or establish infringement unless they are reflected in the claims themselves.
For infringement analysis, the opinion cautions against identifying claim elements only after conceptual cutting, extraction, or reconstruction of an accused product. A proposed component must have a defensible structural existence, not merely be a shape that could be created from a larger whole.
The ruling also underscores the value of a complete expert record. Boston Scientific obtained JMOL on claim 26 in substantial part because its expert testimony was unrebutted. A party bearing the burden of proof may still prevail as a matter of law where the evidence is one-sided and the opposing party offers no reasonable basis for rejecting it.
Conclusion
Board of Regents v. Boston Scientific is both a forceful application of settled anticipation principles and a significant reminder of the limits of jury fact-finding. The Federal Circuit did not merely disagree with the jury’s assessment of competing experts. Rather, it concluded that, viewing the record under the governing JMOL standard, no legally sufficient evidentiary basis supported either the verdict rejecting anticipation or the finding of infringement.
Charles Gideon Korrell views the decision as a reminder that patent trials remain anchored to claim language. Scientific complexity can obscure that point, but it cannot replace it.
Key Takeaways
- Song anticipated the asserted claims because it disclosed biodegradable polymer fibers, immiscible phases, discrete drug-containing regions, and the additional limitations of the dependent claims.
- Differences in intended use or drug-release mechanism could not preserve validity because those distinctions were absent from the claims.
- Uncontradicted expert testimony may establish what a prior-art reference communicates to a skilled artisan, particularly when the reference’s figures and structure support that testimony.
- An infringement theory cannot create a claimed structure by hypothetically cutting or isolating part of an accused product.
- The Federal Circuit will not hesitate to reverse a jury verdict when the record permits only one legally permissible conclusion, even where invalidity must be proven by clear and convincing evidence.