On June 8 2026, the CNIPA issued invalidation decision No. 610889, which upheld the validity of patent ZL200780013339.3 (‘Boron-Containing Small Molecules as Anti-Inflammatory Agents’). The patent at issue, which is owned by Anacor Pharmaceuticals, LLC (ANACOR), relates to the pharmaceutical use of crisaborole.
Crisaborole was developed by ANACOR and was first approved by the US Food and Drug Administration under the trade name Eucrisa. In July 2020, crisaborole ointment was approved for marketing in China, becoming the country’s first hormone-free topical prescription PDE-4 (phosphodiesterase 4) inhibitor for the treatment of mild-to-moderate atopic dermatitis in children and adults. Over the years, generic drug makers vying for the market have filed for regulatory marketing approvals for their generic crisaborole ointment, and the National Medical Products Administration has granted 32 drug production approvals so far. The petitioner in this invalidation case is one of the generic drug manufacturers that has obtained approval.
The China Drug Patent Information Registration Platform shows that ANACOR has registered two patents for crisaborole: a compound patent and a pharmaceutical use patent (the patent at issue). The compound patent expired in February 2026, and the patent at issue will expire in February 2027. This invalidation decision will bar generic drug manufacturers from launching cheaper counterparts in the short term.
Independent claim 1 of the patent at issue reads: “Use of a compound or a pharmaceutically acceptable salt thereof in the preparation of a medicament for treating or preventing inflammation-related diseases in humans or animals, wherein the inflammation-related diseases are members selected from the group consisting of psoriasis, arthritis, dermatitis […] The compounds are members selected from:
Sufficient disclosure
The petitioner asserted that the description did not sufficiently disclose the invention, arguing that the two experimental models and their evaluation indicators selected in the embodiments of the patent cannot prove the therapeutic effect of crisaborole on atopic dermatitis. Specifically, the phorbol ester-induced mouse model is not a dedicated model for atopic dermatitis. Therefore, even if the local phorbol ester mouse ear test is effective, it does not prove that the treatment is directly effective for atopic dermatitis.
The CNIPA rebutted the argument. It reasoned that the approach in assessing sufficient disclosure is two-pronged:
The information disclosed by the applicant shall be sufficient to support implementation of the invention’s solution, thereby meeting the threshold of patentability; and
The scope of disclosure shall be reasonable, thereby striking a balance between patent exclusivity and the substantive technical contribution.
The CNIPA opined that for a pharmaceutical use invention, the experimental data submitted during the filing stage need not be on a par with clinical research in terms of completeness, provided that the existing data is already sufficient to enable a person skilled in the art to confirm a “reasonable expectation” over the therapeutic effect of the drug.
In other words, the selection and construction of an experimental model should not be premised on the model being a dedicated model for a specific disease or fully simulating the entire course of the disease. The examination scrutinises whether the model is reasonably associated and matched with the core pathogenesis or key pathological link of the indication, such that the experimental data from the model is sufficient to objectively determine that the drug exerts a substantive effect on the key target or pathological link in the disease pathway.
The CNIPA found that as the phorbol ester-induced ear swelling model in the description can simulate the core pathological mechanism of atopic dermatitis, it would be improper to deny the probative value of the experimental results obtained using the model just because the model is a dedicated model for another disease in clinical research. In addition, the cytokine inhibition assay in the description further attests that crisaborole can be used to treat atopic dermatitis.
Therefore, those skilled in the art would be able to reasonably expect that crisaborole has a therapeutic effect in the treatment of atopic dermatitis, based on the experimental data in the patent. The CNIPA consequently dismissed the petitioner’s insufficient disclosure argument.
Inventive step
The petitioner also challenged the inventive step of the patent at issue, arguing that Evidence 1, the closest prior art, discloses a compound featuring a highly similar chemical structure, which is particularly effective against microorganisms, especially fungi. The distinguishing features of the patent are the substituent group CN and the compound’s use in the treatment of atopic dermatitis. Evidence 2 and 3 teach that CN substituting at the benzene ring can improve the antimicrobial activity of a compound, and other evidence teaches that atopic dermatitis and eczema can be treated by inhibiting some related bacteria. The use of crisaborole in the treatment of dermatitis thus does not involve an inventive step.
After affirming the technical effects of the patent at issue, the CNIPA held that the evidence does not provide the teaching to use crisaborole to treat dermatitis based on the below reasoning:
The experimental results in Evidence 1 indicate that subtle structural differences in substituents may lead to significant differences in effects. Those skilled in the art would thus lack a reasonable expectation over the efficacy after CN substituting at positions on the benzene ring.
The structure-activity relationship conclusions in Evidence 2 and 3 are drawn from compounds with a specific core structure and cannot be applied to other compounds with substantially different core structures, such as that described in Evidence 1.
The CNIPA therefore ruled that Evidence 1–3 does not teach the structure of crisaborole, and other evidence does not disclose a chemical structure similar to crisaborole that can be used to treat atopic dermatitis. The use of crisaborole in the treatment of atopic dermatitis is thus not obvious and involves an inventive step.
Key takeaways
The CNIPA decision offers key takeaways in assessing sufficient disclosure and inventiveness for pharmaceutical invention patents.
In assessing sufficient disclosure, the CNIPA focuses on the technical effects described in the patent per se and underscores that it would be erroneous to set a high bar and benchmark such disclosure against clinical research in terms of completeness. Moreover, the CNIPA affirms that the inventive step of compounds hinges on whether a clear structure-activity teaching can be obtained from the prior art, rather than relying on the outlining of a variety of substituents in generalities.
