Owning the Improvement Is Not Freedom to Operate: Exelixis v. MSN Shows the Reach of Earlier Genus Patents

A company can obtain a patent on a new form of a technology and still infringe an earlier patent covering a broader category that includes it. The Federal Circuit’s decision in Exelixis, Inc. v. MSN Laboratories Private Ltd., No. 25-1236 (Fed. Cir. Aug. 31, 2026), provides a concrete example of that distinction and an important reminder for businesses developing improvements within an existing technology landscape.

MSN Laboratories obtained its own patent on a crystalline form of cabozantinib (L)-malate known as Form S. Yet when MSN sought approval for a generic pharmaceutical product using Form S, it conceded that the product infringed earlier Exelixis patents broadly claiming crystalline cabozantinib (L)-malate. MSN therefore challenged those broader patents under the written-description requirement of 35 U.S.C. § 112(a).

The Federal Circuit affirmed the district court’s rejection of that challenge. The result illustrates two related points with consequences beyond pharmaceutical patent litigation. First, patentability and freedom to operate are separate inquiries. Second, the commercial reach of a broad patent may depend on whether its specification adequately establishes possession of the claimed territory, not merely the particular examples the patent expressly describes.

Exelixis holds the New Drug Application for Cabometyx, a cancer treatment containing cabozantinib (L)-malate. During development, Exelixis investigated different solid-state forms of the compound and identified two crystalline polymorphs designated N-1 and N-2.

Exelixis ultimately obtained several patents with claims extending beyond those particular forms. U.S. Patent No. 11,091,439 claims crystalline cabozantinib (L)-malate salts, while related patents claim formulations and treatment methods using those salts.

MSN later developed Form S, another polymorph of cabozantinib (L)-malate, and obtained a patent directed to that form. The existence of MSN’s patent did not place Form S outside Exelixis’s earlier claims. MSN conceded infringement and instead argued that Exelixis had not adequately described the broader genus of crystalline forms it claimed.

That posture highlights a recurring commercial issue in innovation-driven industries. A patent establishes a right to exclude others from practicing the claimed invention. It does not necessarily establish that the patent owner may practice the invention without permission from someone holding broader, earlier rights.

For technology development, licensing, acquisition, and commercialization decisions, the distinction matters. The question whether an improvement is patentable is different from the question whether commercializing it requires rights under another party’s portfolio.

The central validity question was whether Exelixis’s specification demonstrated possession of the genus it claimed even though it specifically characterized only two polymorphs.

Under the Federal Circuit’s en banc decision in Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co., § 112(a) requires a patent specification to show that the inventor possessed the claimed invention as of the filing date. For a genus, Ariad identifies two principal ways to make that showing: disclosure of a representative number of species, or disclosure of structural features common to the genus that allow a skilled artisan to recognize its members.

The Federal Circuit concluded that Exelixis could rely on the second route. Its specification identified cabozantinib (L)-malate by chemical name and formula and required the claimed material to be crystalline. Those characteristics allowed skilled artisans to distinguish members of the claimed genus from other materials, including amorphous cabozantinib (L)-malate.

The court therefore did not need to determine whether Exelixis’s two disclosed polymorphs constituted a representative number of species. The common structural characteristics were sufficient to sustain the district court’s written-description finding on the record before it.

MSN emphasized that different polymorphs can possess materially different physical properties, including solubility, stability, density, melting point, hygroscopicity, and vapor pressure. Those differences can be extremely important when deciding which form of a pharmaceutical compound to develop commercially.

But the Federal Circuit focused on what Exelixis had actually claimed.

The claims did not require members of the genus to achieve a particular solubility, stability, therapeutic effect, or other performance characteristic. They required cabozantinib (L)-malate in crystalline form. Accordingly, differences among polymorphs in properties that were not claim limitations did not establish that the specification failed to describe the claimed genus.

That distinction helped make GlaxoSmithKline LLC v. Banner Pharmacaps, Inc. a more useful analogy than AbbVie Deutschland GmbH & Co. v. Janssen Biotech, Inc. In GSK, common structural characteristics allowed skilled artisans to recognize the claimed genus. In AbbVie, by contrast, the disclosed antibodies occupied a comparatively narrow structural group while the claims extended to materially different antibodies.

Exelixis therefore does not establish a broad safe harbor for genus claims. The potential polymorph genus was comparatively limited, with the district court finding a maximum of fourteen pure forms, and the claims tracked characteristics actually identified in the specification. The Federal Circuit also reviewed the district court’s factual written-description findings only for clear error.

The commercial significance of the decision lies in the interaction between layers of intellectual property rights.

Innovation often proceeds through successive improvements. One company may establish a broad technological platform or genus, while another later develops a narrower implementation with meaningful advantages. The later development may itself satisfy the requirements for patent protection without escaping claims directed to the broader technology.

That creates a practical need to evaluate patentability and freedom to operate separately. A business deciding whether to invest in development, acquire technology, enter a license, or launch a product should not treat issuance of its own patent as evidence that third-party rights have been cleared.

The same distinction can affect technology transactions. A license covering a particular improvement may not be sufficient if commercialization also implicates earlier platform or genus patents. Conversely, ownership of a broader portfolio can remain commercially significant even as competitors patent later species or refinements within the claimed territory.

Exelixis also shows why claim scope must be evaluated together with the disclosure that supports it. Broader claims can have substantial commercial value, but that value depends on satisfying § 112 across the territory actually claimed. The Federal Circuit did not excuse incomplete disclosure merely because Exelixis had identified some members of the genus. It upheld the claims because the specification supplied common characteristics that, on this record, meaningfully identified the claimed class.

The most useful lesson from Exelixis v. MSN Laboratories is straightforward: obtaining intellectual property protection for an improvement and obtaining the right to commercialize that improvement are different legal and business questions.

MSN’s Form S was sufficiently distinct to support its own patent, yet remained within earlier Exelixis claims that survived a written-description challenge. For businesses developing technology in crowded patent landscapes, that distinction should inform product development, licensing, diligence, and commercialization strategy well before launch.

At the same time, Exelixis explains why earlier genus claims can retain that blocking power. Section 112(a) does not invariably require a patent owner to describe every species individually. Where the specification identifies meaningful common structural characteristics that allow skilled artisans to recognize the claimed territory, a broader genus may be adequately described.

The strategic issue is therefore not simply who patented the newest technology. It is how the different layers of patent rights fit together, which claims actually cover the commercial product, and whether the disclosure supporting those claims is strong enough to sustain the scope on which the business depends.