IP Update: China’s Supreme People’s Court Issues First Public Decision Defining China’s Inventorship Standard

In theory China's inventorship standard is similar to others around the world. An inventor is anyone who "makes creative contributions to the substantive features" of an invention. But what does this mean, and how does one apply this standard? Up to this point, this issue hasn't really been litigated much in China (at least according to publicly available records).

We finally have an important case (from the top!) that clarifies this issue. We analyse a binding judgment by the Intellectual Property Court of the Supreme People’s Court (SPC) of China1. It clarifies China’s applicable inventorship standards and the potential legal consequences of omitting qualified inventors. For U.S. companies conducting R&D cooperation and filing patents in China, this case sets important factors to consider.

1. Case Background

Key Parties

Collaboration Relationship

X Company is the owner of Proxalutamide, an androgen receptor (AR) inhibitor. In early 2020, Dr. Guo first conceived the idea of using AR inhibitors to block COVID-19 infection and reduce severe inflammation. He and the team shared this idea with X Company, provided technical protocols, and guided the design of global clinical trials, all without a written collaboration agreement in place. X Company subsequently applied for a patent application directed towards the use of proxalutamide in treating COVID-19 (including variants). They did not list Dr. Guo as an inventor.

Later, during the patent examination process, X Company faced a novelty rejection. Initially, the company argued that the D1 should be excluded from prior art3. However, just two weeks later, it voluntarily deleted Claims 1–4 (which covered wild-type COVID-19) without providing a valid reason.

2. SPC’s Inventorship Determination Standards

Beyond the patent prosecution arguments, the SPC’s ruling clarified the standards for inventorship in China, which differ significantly from the U.S.’s claim-centric approach.

First, the SPC emphasized that “substantive features of an invention” for the purpose of inventor identification4 are not equivalent to those used in determining inventiveness 5. Inventor identification focuses on whether a natural person made substantive technical contributions to the invention process (e.g., method of participation, extent of involvement, and creative input), while inventiveness review focuses on distinguishing the invention from prior art. These two standards serve distinct legal purposes and cannot be conflated— inventor status is independent of whether the invention itself is inventive, for both applications and granted patents.

Second, the SPC stressed that the invention process is iterative, and the discovery of technical problems and proposal of inventive concepts—the starting point of any invention—constitute critical creative contributions. In this case, Dr. Guo and the team first proposed the inventive concept (using AR inhibitors to treat wild-type COVID-19) in early February 2020 and shared research materials with X Company on February 8, 2020. This concept laid the foundation for all subsequent R&D activities.

Third, as an invention involving “a new medical use of a known compound,” the patent’s innovation point lies in the new therapeutic application of proxalutamide (not the compound itself). X Company failed to demonstrate that it independently discovered proxalutamide’s potential for treating COVID-19 prior to collaborating with Dr. Guo, whereas Dr. Guo explicitly proposed this second medical use concept in the materials he provided to X Company.

Notwithstanding these arguments, the CNIPA rejected the claims related to wild-type COVID-19 (citing lack of novelty due to D1, X company’s own public disclosure). X Company deleted the wild-type COVID-19 claims and retaining only those related to variant-specific use. The first trial court held that, in order to emphasize its exclusive ownership interests in the subject patent, X Company intentionally and proactively modified the claims during the litigation stage to “technically evade” the creative contributions of Dr. Guo to the patent in discussion, and that such conduct violated the principle of good faith. The SPC affirmed this holding on appeal.

Fourth, the patent’s substantive features include both proxalutamide’s use for general COVID-19 treatment (the foundational element) and its use for variants (a continuation of the original concept),contrary to X Company’s assertion that only the variant-specific use was inventive. Dr. Guo and the team also participated in other substantive R&D activities, including clinical trial design, data sharing, and technical discussions.

In summary, the SPC confirmed Dr. Guo’s status as an inventor because his inventive concept and substantive contributions formed the starting point and core of the patent—regardless of X Company’s subsequent claim modifications.

3. Final Judgment of SPC

The SPC upheld the lower court’s decision and ordered the following:

  1. Inventorship Confirmation: Dr. Guo is legally recognized as an inventor of the patent.
  2. Recordal Change: The patentee must amend the inventor list on the patent with the China National Intellectual Property Administration (CNIPA) within 30 days.
  3. Public Notice: The patentee must publish a 30-day correction statement on its official website and social media platforms to rectify the inventorship record.
  4. Cost Liability: The patentee shall bear all litigation costs.

It is uncertain whether there are any other civil litigation cases between the two parties seeking economic compensation, though this particular case did not involve any such claims

EIP thoughts:

This SPC ruling offers important insights regarding how inventorship is determined in China. Having said that, inventorship plays a different role in China than in the US. A large percentage of inventions in China are Employee Service Inventions, where ownership of inventions made during the course of employment or using an employer’s resources flow directly to the employer. In a majority of these cases, adding or removing an inventor likely will not affect ownership.

Still, it’s important to understand that China’s standards for inventorship are quite different from US standards, and there could be some serious consequences if applicants amend inventorship in Chinese applications in the same way as in US applications.

We’ve summarized some key takeaway lesson below based on this case:

  1. Different Standards for Inventorship: Unlike the US, which determines inventorship based on who contributed to each granted claim (on a claim-by-claim basis), Chinese courts prioritize creative contributions to the invention as a whole, i.e., based on the full technical concept behind the invention, not just the final granted claims.
  2. Amending Inventorship After Filing: Be very, very mindful of removing inventors after filing, especially if they have made contributions to the invention described in the specification as filed. As demonstrated in this case, deleting patent claims to “cancel” the contributions made by an inventor is a “violation of the good faith principle in China.”
  3. Potential Consequences: Purposely omitting “rightful” inventors in China carries potential risks, including court-ordered inventorship corrections, mandatory public notices (which can harm a company’s reputation in China), litigation costs, and even potential challenges to patent validity6.

Beyond the direct ruling, we also highlight potential subsequent issues of which U.S. clients should be aware:

  1. Inventor Remuneration Claims: Confirmed inventors may pursue statutory rewards or reasonable compensation under Chinese law.7 If no previous contract was in place, this remuneration amount could theoretically be quite large, up to >5% of the profits from the inventor’s contributed portion of the invention.8
  2. Patent Administrative Delays: Because amendments to correct inventorship in China are relatively rare, the CNIPA may conduct targeted reviews of patents/applications that have corrected inventorship, which could cloud licensing, assignment, or commercialization efforts.
  3. Cross-Border Agreements: Unclear R&D agreements may lead to disputes over inventorship and IP ownership, particularly given the divergent standards between the two jurisdictions.

Take Away

Based on these insights, below are a few recommendations:

Be mindful of the Chinese standard of inventorship when determining the list of inventors, focusing on contributors to the entire R&D process (not just final claims). Remember that even if a contribution is not novel or inventive, it may still be considered worthy of inventorship.

Document all evidence relating to creative contributions in writing (especially from R&D partners), creating a clear record to resolve potential inventorship disputes.

Be very, very mindful of removing inventors after filing, especially if they have made contributions to the invention as described in the specification as filed. Because most inventions are service inventions and modifying inventorship does not impact ownership, usually the act of modifying inventorship may present more risk to a patent than if the inventor list was left alone.

Explicitly define inventorship and IP ownership standards in cross-border R&D contracts, understanding there are significant differences between Chinese and U.S. patent practices to avoid ambiguity down the line.

Consider engaging Chinese IP counsel earlier to advise on Chinese patent strategy, especially if cross border activity is going on that involves multiple jurisdictions and different standards of inventorship and ownership.

If you would like to have more information on this matter or would like to have our advice, please feel free to contact us at [email protected].

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

  1. （2023）最高法知民终2911号 Case No.: (2023) Highest IP Civil Appeal No. 2911
    https://ipc.court.gov.cn/zh-cn/news/view-3387.html?f_link_type=f_linkinlinenote&flow_extra=eyJpbmxpbmVfZGlzcGxheV9wb3NpdGlvbiI6MCwiZG9jX3Bvc2l0aW9uIjowLCJkb2NfaWQiOiI4ZWVjZDE4ZGFmOTZkODk5LTkzOWE2MDQyMjU0NTM0YzUifQ%3D%3D ↩︎
  2. Patent No. ZL202110657053.0 ↩︎
  3. D1: 《普克鲁胺治疗COVID-19获突破：显著降低住院率和重症率，生物探索》. D1 disclosed the use of proxalutamide for treating wild-type COVID-19. On 16 Sep 2022, company X argued that D1 was made public for the first time for the purpose of the public interests when an extraordinary situation occurred in the country, and the application was filed within 6 months from the publication date of D1. Therefore under Art. 24 a grace period should be applicable. In short, D1 should be excluded from prior art for all claims in the application, meaning none of the technical schemes should lose novelty due to D1. We are of the view that this argument had a good chance to be accepted. On 24 Sep 2022, a Request for Suspension of Procedure was submitted to CNIPA due to a dispute over inventorship rights. On 10 Oct 2022, X company, knowing about the request, filed a response again deleting claims 1-4. ↩︎
  4. R13 of the Implementing Rules of the Patent Law. An inventor or creator referred to in the Patent Law means any person who makes creative contributions to the substantive features of an invention-creation. ↩︎
  5. Art. 22.3 of the Patent Law. Inventiveness means that, as compared with the prior art, the invention has prominent substantive features and represents an obvious progress, and that the utility model has substantive features and represents a progress. ↩︎
  6. It is important to clarify that the mere omission of eligible inventors, standing alone, does not constitute a ground for invalidation according to R69 of the Implementing Rules of the Patent Law.  Instead, the act of omitting inventors accompanied by a violation of the principle of good faith—such as intentionally modifying claims to evade recognition of the omitted inventors’ creative contributions, as seen in this case—may lead to challenges to patent validity according to R11. ↩︎
  7. R93 of the Implementing Rules of the Patent Law. Where patentee has neither reached an agreement with the inventor or designer, nor stipulated in its lawfully formulated rules and regulations the method and amount of the reward referred to in Art. 15 of the Patent Law, it shall, within three months from the date of the announcement of the grant of the patent right, issue an award to the inventor or designer. The award for an invention shall not be less than 4,000 RMB; the award for a utility model or design shall not be less than 1,500 RMB. ↩︎
  8. Law of the People's Republic of China on Promoting the Transformation of Scientific and Technological Achievements, Art 45.
    …(1) Where the work-related S&T achievements are transferred or licensed to others for implementation, at least 50% of the net income from the transfer or licensing of such S&T achievements shall be withdrawn for rewards and compensation; (2) Where the work-related S&T achievements are used for trade-in investment, at least 50% of the shares or capital contribution proportions formed from the S&T achievements shall be withdrawn for rewards and compensation; (3) Where the work-related S&T achievements are implemented by the organizations themselves or through cooperation with others, at least 5% of the operating profits from implementing such S&T achievements shall be withdrawn for rewards and compensation each year, for three to five years, consecutively, after their conversion into practical applications and the successful commissioning thereof. ↩︎

Breaking: China Released Drug Trial Data Protection Implementation Measures

On May 15, 2026, China’s National Medical Products Administration (NMPA) officially announced the final Measures for the Implementation of Drug Trial Data Protection1 (“Measures”), effective May 15, 2026.

These Measures provide important details supporting the recently announced Implementing Regulations of the Drug Administration Law (“Regulations” – also effective May 15, 2026), which formally introduced a statutory market exclusivity of up to 7 years for qualifying orphan drugs, up to 2 years for pediatric drugs, and up to 6 years data exclusivity for “new chemical entity drugs and other drugs”.

Key Highlights

Under the finalized Measures, the NMPA2 grants a maximum regulatory data protection (“RDP”) period of up to 6 years from the date of a drug’s marketing authorization in China. During this protection window, the NMPA will not approve subsequent marketing authorization applications for generics, biosimilars, or modified drugs that rely on the original drug marketing authorization holder’s (MAH) “undisclosed experimental data and other data” without the MAH’s consent.

During the RDP period other applicants can submit drug marketing applications based on their own obtained data, but such data will not receive RDP protection and cannot be used by other following applicants.

What Data is Protected?

The protected “undisclosed test data and other data” refers to unpublished data (including partially-published data) used for the first time in China in a marketing authorization application. For innovative drugs, this includes data that demonstrate the safety, efficacy, and quality controllability of the drug. For improved innovative drugs, the protection includes all supporting and necessary clinical trial data showing clinical advantages of the improved drug over the original innovative drug, with the exception of bioavailability, bioequivalence, and immunogenicity data for vaccines.

The data protection periods are summarized below (full charts at the end of this article):

Type of Product

Class

RDP Period

Innovative Drugs, Vaccines, and Biologics

Class 1

6 years

Improved Innovative Drugs, Vaccines, and Biologics

Class 2

4 years

The first approved domestic generic drug/biosimilar of an originator drug marketed overseas but not yet in China

Class 3.2 / 5.2

3 years

Some Notable Observations

Removal of "Penalty” for Drugs Entering China Later: The 2025 draft included a restrictive penalty for drugs approved overseas before entering China, proposing that the protection period be reduced by the time difference between the overseas MA approval date and the China MA acceptance date. The final 2026 Measures completely removes this deduction formula, treating innovative and improved innovative drugs, vaccines, and biologics equally, regardless of whether they were first approved elsewhere.

Extended Data Protection for Improved Innovative Drugs: Compared to the 2025 draft, data protection period for Improved Innovative Drugs has increased from 3 years to 4 years.

Broadened Definition of "Self-Obtained" Data: In its official policy interpretation, the NMPA clarified that "self-obtained" data does not require the applicant to have physically conducted the clinical trials themselves. Data acquired through commercial purchases or exclusive licensing agreements also qualify for data protection, creating additional value for such data.

Clarification on "Undisclosed" Status: The NMPA clarified that mandatory global clinical trial registrations or summary abstract disclosures do not compromise the required "undisclosed" status of the protected data under these Measures, provided the comprehensive, underlying raw trial data remain confidential.

Timeline for Generic Entry: Generic applicants can submit filings relying on the protected data starting 1 year before the exclusivity expires. The NMPA will complete its technical review, pause the application process, and release the final marketing authorization immediately after the data protection period ends.

Subsequent Generics: To improve drug accessibility, once a "first generic" of an overseas original drug is approved (gaining a 3-year protection period), subsequent generic applicants do not need to duplicate clinical safety and efficacy trials if the first generic has already been listed on the domestic market for 2 years.

EIP Thoughts:

We are pleased to see that the final Implementation Measures have removed any reductions for RDP, providing the same data protection for all new drugs regardless of whether they obtained MA outside of China first. This was an important issue for many foreign drug companies, and we are encouraged that the NMPA appears to have taken into account feedback3 it received following the Draft Measures to arrive at this final version. These finalized Regulations and Measures will certainly encourage even more foreign companies to consider entering China.

China’s RDP policy allows a single drug to potentially receive data protection separately for each new indication. However, the protection scope for a new indication is limited to the clinical trial data used to support the launch of the new indication.

The NMPA explained that “the core purpose of RDP is to protect original trial data generated to prove a drug's safety and efficacy.” Therefore, new data protection for a drug can be granted only if the newly generated data was essential for evaluating a drug or vaccine's safety and efficacy (e.g., in the new indication). This principle applies equally to new combination therapies, as well as for bioavailability, bioequivalence, and vaccine immunogenicity data.

Interesting, in China PTE (patent term extension due to regulatory delay) is only available to Innovative Drugs (Class 1) and certain Improved Innovative Drugs (Class 2) that have not yet been marketed in China. Additionally, one can only receive one PTE per drug, one PTE per patent. Drug companies should consider RDP strategy carefully in conjunction with a drug’s overall patent strategy, especially when multiple indications may exist. This is especially true in China, where certain patent decisions (e.g., when to file divisional applications) need to be made relatively early in a patent’s prosecution lifetime. Drug companies will need to consider various aspects of how to map out a product’s patent, clinical, PTE, RDP, and market entry strategies to optimize the overall protection of a drug product.

One Important Deadline

Applicants must formally submit their data protection request to the Center for Drug Evaluation (CDE) of the NMPA simultaneously with their initial marketing authorization application. Companies who already have approved drugs in China or pending marketing applications submitted prior to May 15, 2026 must submit their data protection requests within 15 working days of the announcement, i.e no later than June 5, 2026.

Chemical Drug Registration Classification and Data Protection Period

Class

Drug

Data protection period

Remark

Class 1

Innovative Drugs that have not been marketed in China or overseas

6 years

 

Class 2

Modified (or “improved”) Innovative Drugs that have not been marketed in China or overseas

4 years

For Innovative Drugs that are already marketed overseas but not in China, if a registration application for a new indication that has not been approved in either China or overseas is submitted for the first time, and meets the requirements of Article 7, Paragraph 1 of these Measures, the data protection period is 6 years.

Class 3

Drugs manufactured by domestic applicants that are Generics of Innovative Drugs that have been marketed overseas but not in China

3 years

The first drug of its kind to be approved in China

Class 4

Drugs manufactured by domestic applicants that are Generics of Innovative Drugs that have been marketed in China

None

No data protection

Class 5

Drugs that have been marketed overseas

and are under application for marketing in China

5.1

Innovative Drugs that are already marketed overseas but not in China

6 years

If the conditions in Article 7, Paragraph 2 are met, the data protection period is 4 years.

Modified (or “improved”) Innovative Drugs that are already marketed overseas but not in China

4 years

 

5.2

Generic Drugs that are already marketed overseas but not in China

3 years

The first drug of its kind to be approved in China; or a generic version of an innovative drug that is already on the market overseas but not in China.

Vaccines Registration Classification and Data Protection Period

Class

Drug

Data protection period

Remark

Class 1

Innovative Vaccines

6 years

 

Class 2

Modified (or “improved”) Innovative Vaccines

4 years

 

Class 3

3.1 Vaccines manufactured overseas that are already marketed overseas but not in China

6 years

 

3.2 Vaccines that are already marketed overseas but not in China and are submitted for approval for manufacturing and marketing in China.

3 years

 

3.3 Vaccines that already marketed in China

None

No data protection

Biologic Drugs Registration Classification and Data Protection Period

Class

Drug

Data protection period

Remark

Class 1

Innovative biological products

6 years

 

Class 2

Modified (or “improved”) Innovative drugs

4 years

For original drugs that are already marketed overseas but not in China, if a registration application for a new indication that has not been approved in either China or overseas is submitted for the first time, and meets the requirements of Article 7, Paragraph 1 of these Measures, the data protection period shall be 6 years.

Class 3

3.1 Biological products manufactured overseas that are already marketed overseas but not in China

6 years

For cases meeting the criteria in Article 7, Paragraph 2, the data protection period is 4 years. For non-innovative drugs, the data protection period is 3 years.

3.2 Biological products that are already marketed overseas but not in China and are submitted for approval for manufacturing and marketing in China.

3 years

 

3.3 Biosimilars

None

No data protection

3.4 Other biological products

None

No data protection

If you would like to have more information on this matter or would like to have our advice, please feel free to contact us at [email protected].

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

 About the Authors

Jennifer Che, J.D. is President & Managing Director and a US Patent Attorney at Eagle IP, a boutique patent firm with offices in Hong Kong and Shenzhen.

Yolanda Wang is a Principal, Chinese Patent Attorney, and Chinese Patent Litigator at Eagle IP, a boutique patent firm with offices in Hong Kong and Shenzhen.

  1. 药品试验数据保护实施办法 Sources: official announcement and follow up notice (Chinese)

    ↩︎

  2. The Center for Drug Evaluation (CDE) of the NMPA will oversee the granting of RDP.

    ↩︎

  3. Many organizations, including the American Intellectual Property Law Association (AIPLA), Intellectual Property Owners Association (IPO), and other organizations representing the interests of foreign drug companies provided Comments in response to the Draft Measures released in March 2025. Eagle IP worked with many other attorneys on the IP Practice in China Committee of AIPLA to submit Comments on behalf of AIPLA to the NMPA about these concerns.

    ↩︎

Please “Mind the Traps”: Four Common Misconceptions About Hong Kong Patents

For many foreign companies interested in the broader Asian market, Hong Kong is a strategically important jurisdiction. However, because of its unique legal framework, domestic and foreign companies alike are confused by how IP protection works in Hong Kong. Missteps in such understanding can lead to critical, costly "gaps" in IP protection.

To safeguard innovations effectively in Hong Kong, be aware of these four common misconceptions:

Hong Kong IP protection goes well beyond a simple registration of a Chinese or European/UK patent. There are many strategies and multi-layered approaches for protecting IP in this key jurisdiction.

If you would like to have more information on this matter or would like to have our advice, please feel free to contact us at [email protected].

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

A Stratification Too Far? How the CNIPA Draws the Line on Patient Subgroups in a Recent Precision Medicine Invalidation Case

Precision medicine is rapidly transforming the global healthcare landscape, providing more effective therapies and better patient outcomes through targeted solutions. As therapeutics move away from the traditional “one-size-fits-all” model, unique challenges are presented in the patent examination process.

The 2024 Top 10 Patent Re-examination and Invalidation case highlighted below provides further insight into how the China National Intellectual Property Administration (CNIPA) interprets claim scope and inventive step during the examination of precision medicine technologies.

Background

Generic Drug maker Chia Tai Tianqing Pharmaceutical Group filed an invalidation request before the Patent Re-examination Board (the “Board”) against an invention patent titled "Use of Degarelix1 in the Preparation of a Medicament for Treating Metastatic Prostate Cancer"2, owned by Ferring International Pharma (Switzerland), on the basis of lack of inventiveness.

Claim 1 of the captioned application recites the following:

Use of Degarelix in the preparation of a medicament for treating metastatic prostate cancer in a subject, wherein the treatment comprises:

wherein the subject has a pre-treatment baseline serum alkaline phosphatase level (sALP) of approximately 150 IU/L or higher.

The closest prior art was a clinical trial publication which disclosed most features of the claim, with the exception of the pre-treatment sALP level subject requirement.

The core argument centered on how to interpret the claim scope of the precision medicine invention, specifically how claim limitations are read. The Board assessed whether such a physiological biomarker – though related to disease progression – was significant enough to be considered a valid limitation in order to support inventive step. Ultimately, the CNIPA invalidated the patent, offering important guidance on claim interpretation and inventive step in the context of precision medicine.

Understanding Scope of the Claims

Under Article 25 of the Chinese patent law, methods of treatment are not patent eligible subject matter. Treatment methods can only be protected as Swiss-type claims, which are normally read as “use of x for the manufacture of a medicament for treating y”. Typically, there are three types of features which are considered limiting in this type of claim:

In this case, the Examiner considered certain additional features of claim 1, (e.g., the initial dosage, maintenance dosage, and frequency of dosing) to relate solely to treatment administration (which are only involved after the manufacturing process, e.g., per the doctor’s instructions). Therefore, the CNIPA did not consider these features as being limiting to the manufacturing process, and thus not considered valid limitation to the claim scope.

Examining Inventiveness

Excluding the non-limiting features of claim 1, the Board analyzed the remaining features in the claim:

The closest prior art presented4 at the invalidation proceeding was a published phase III clinical trial study using Degarelix in prostate cancer patients. The study disclosed a 97.2% positive treatment response rate across patients of various disease stages, of which 20% were metastatic prostate cancer patients.

The only differentiating feature of the claimed invention from the prior art was the pre-treatment baseline sALP level range in the indicated patient group (the patentee also indicated it should be understood as a novel disease sub-type in the claim during the invalidation procedure).

Determining Technical Problem Solved

In order to evaluate inventive step, the CNIPA needs to determine a) the technical problem solved by the differentiating feature and b) determine whether the prior art as a whole provides motivation to solve the problem through structural modification.

Was the pre-treatment baseline sALP level a valid claim limitation that affected the technical problem that was solved? A feature can be considered “limiting” if it contributes to a specific technical effect over the prior art. In this case, for example, limitations to the treatment of a specific patient subgroup may be acceptable if the treatment is directly linked to a differentiating technical effect in that patient subgroup.

The patentee argued that the limitation of metastatic prostate cancer in combination with pre-treatment baseline sALP level range resulted in a specific target patient group having a disease subtype not described in the prior art, namely prostate cancer patients with bone metastasis.

While the Board acknowledged that the specification demonstrated that “sALP reduction in the selected patient group was more effective than in patients with other disease stages”, they reiterated there is no evidence that persons having ordinary skill in the art (PHOSITAs) agreed that pre-treatment sALP levels alone could be used to classify a new prostate cancer subtype.5 As a result, the Board concluded that the technical problem solved was limited solely to the use of Degarelix in treating patients with metastatic prostate cancer.

Determining Technical Effect of the invention

Despite the fact that the applicant had shown effective sALP reduction in certain subjects in the specification, the Board emphasized that the applicant needed to provide evidence supporting that sALP level changes over time can act as a single marker to measure disease progression and treatment outcome. The Board stated that the patentee’s data was merely “an analysis of physiological parameter changes during treatment of patients with specific pre-treatment baseline sALP level”. Without a clear link to a specific technical effect, these types of changes would not rise to the level of being a valid limitation in a Swiss-style claim.

As no new disease indication or corresponding treatment effect (against prostate cancer) was identified, the Board invalidated the patent on the grounds that inventiveness cannot be determined without these elements.

EIP Thoughts

So what is the standard for protecting certain patient sub-groups?

When protecting precision medicine technologies, it is common to add limitations to certain features that identify patient sub-groups within a broader disease indication.

The present decision emphasizes that the Board looks for “clinical guidance value” in these claimed features when determining whether to consider such “disease marker” type limitations. Specifically, the Board looks at inventiveness – whether a PHOSITA would think that the feature (combined with the knowledge in the art) is valuable in guiding clinical decisions.

In the present case, the limiting disease marker, sALP level alone, was not definitive on its own to identify a new and meaningful patient group. In fact, sALP also correlates to other physiological outcomes. For example, even healthy individuals or individuals with other diseases can have the same elevated levels of sALP.

A disease marker needs to be clear, significant and definitive for PHOSITAs, as these markers are only meaningful if they can accurately stratify patients, enabling subsequent treatment to be assigned appropriately.

We believe that a crucial step to protecting these features is to present a clear relationship and connection between the limiting disease marker, the disease indication, and subsequent treatment outcomes in the specification. In this example, if the patentee had provided evidence showing that only patients having sALP levels at 150 IU/L or above responded surprisingly well to the treatment, then the Board may have recognized the sALP level as a limiting feature in the claim because a patient’s sALP levels would guide the clinical decision of a doctor. The claim(s) would likely be considered inventiveness based on the surprising technical effect.

What is “agreed” in the art?

While we now understand the importance of linking disease markers with how they guide clinical decisions, how can we determine what is “agreed upon” between PHOSITAs? When is a particular mutation or sub-population considered a new disease?

In an article published by members of the Board in charge of this case6, the authors provided further discussion into this issue of classifying diseases.

Specifically, they pointed out that there are a number of disease classification systems (whether it be international systems, such as the International Classification of Diseases (ICD), national systems, or systems categorized by academic associations under specific disciplines), that may not be compatible with each other. At the same time, it is almost certain that technological and clinical advances will evolve faster than the systems themselves, as these classification systems go through meticulous and long processes in producing updates. For example, while HER2-positive breast cancer has been classified clinically and treated with targeted therapy for decades, it was only just recently added in the FY 2025 ICD update7[7].

The article confirmed that the standard for claiming new patient subgroups or disease classifying markers is not based on whether a disease is listed in well-established systems (e.g., covered in the ICD). Instead, patent examiners will use a more flexible approach, taking into account other factors to determine what is considered a clinically recognized disease.

As suggested by the Board’s comments, as long as a claimed disease marker carries substantial “clinical guidance value”, we believe having sufficient clinical data demonstrating such guidance value can support limiting a disease to a new patient subgroup. However, it is still unclear if earlier stage data, such as using only pre-clinical data, would be sufficient to support a new subgroup.

This case highlights the ever-present tension that clinical stage companies face between the (earlier) public disclosures of clinical trial information with the ultimate confirmed results of the clinical trial. Prior published clinical trial information (required by the US FDA) typically discloses the medicament, the disease to be treated, and expected outcomes. Because China’s patent laws do not allow method of treatment claims (thus excluding dosing schedules, dosing regimens, and other types of new IP that could arise from a clinical trial), clinical stage companies have more limited options when it comes to protecting IP in China at this later stage of a product’s development. Finding strong relationships between various disease markers and disease outcomes is a solid way to obtain additional IP protection from clinical trial results in China.

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

  1. Degarelix is a GnRH antagonist that lowers testosterone levels in the body.

    ↩︎

  2. Patent No. ZL200980104713.X

    ↩︎

  3. Features such as physical characteristics of the manufactured product or routes of administration may be considered as valid limitations that limit part of the manufacturing process. However, whether the CNIPA will accept these limitations is based on if these features substantially affect the claimed manufacturing process. For example, one of skill in the art would expect a different manufacturing process for an oral formulation and a subcutaneous injection carrying the same active compound. However, if in fact the manufacturing process is the same, even with a different delivery method (e.g. same formulation injected at different locations with known routine modified process), the CNIPA will likely not consider such different routes of administration as limiting.

    ↩︎

  4. Klotz L, Boccon-Gibod L, Shore ND, Andreou C, Persson BE, Cantor P, Jensen JK, Olesen TK, Schröder FH. The efficacy and safety of degarelix: a 12-month, comparative, randomized, open-label, parallel-group phase III study in patients with prostate cancer. BJU Int. 2008 Dec;102(11):1531-8. doi: 10.1111/j.1464-410X.2008.08183.x. PMID: 19035858.

    ↩︎

  5. Changes in bone turnover and osteoblastic activity, caused by cancer bone metastasis, may be reflected in ALP level changes. Although the priority date of this patent is in 2009, even at least up to 2023, serum ALP levels was still not a well-recognised marker for prostate cancer bone metastasis due to its correlation to several non-neoplastic illnesses, low diagnostic sensitivity and lack of well accepted cut-off value used across different researchers. See more at: Biomarkers for Prostate Cancer Bone Metastasis Detection and Prediction

    ↩︎
  6. https://www.cnipa.gov.cn/art/2025/6/6/art_2648_200011.html ↩︎
  7. https://libmaneducation.com/icd-10-cm-for-2025-new-codes-for-hormone-status-in-breast-cancer/ ↩︎

What Microsoft’s Recent Patent Invalidation Case in China Teaches Us About User Interface Patents

Each year the CNIPA publishes its Top Ten Patent Re-examination and Invalidation Cases for the previous year. These cases are meant to be guiding cases, showcasing exemplary real-world decisions that clarify certain aspects of the law.

Today we’ll be sharing about one of the Top 10 Invalidation Cases in 2024 involving Microsoft (China) Co., Ltd. petitioning to invalidate a patent by Newman Infinite Inc. in the field of human-computer interaction (HCI).

Case summary

The patent claimed methods and devices for manipulating a touch screen user interface that could help differentiate between various touch targets, preventing undesired interactions with non-target elements. The patentee used a self-defined term "clutch user interface element", which was not a common term used in the art.

“Claim 1. A method for manipulating a user interface element presented on a display screen, the method comprising:

displaying a target user interface element on a touch-sensitive display screen; displaying a clutch user interface element;

determining that a selection touch event has occurred when the target user interface element is approached while no engagement touch event is present when the clutch user interface element is approached;

in response to determining that the selection touch event has occurred, selecting the target user interface element for manipulation;

determining when the engagement touch event is occurring; and

when the engagement touch event is occurring, activating, by a processor, the manipulation function associated with the target user interface element.”

Key Issues

1. Lacking Essential Technical Features in the Claims

The crux of Microsoft’s argument was that claims 1, 9, and 17 were invalid for insufficient disclosure because they did not contain the essential technical feature of "after a touch engagement event occurs, another user interface element cannot be selected."

The CNIPA disagreed, indicating that the essential technical feature was present in the self-defined term “clutch user interface element”, which was defined in claim 1. Additionally, the meaning was further supported by the specification and further clarified during the Second Office Action response. The CNIPA concluded that a skilled person would understand its role in preventing undesired interactions with non-target elements. Therefore, claims 1, 9 and 17 included all essential features necessary to solve the technical problem and complied with Rule 20(2).

2. Novelty and Inventiveness

Microsoft also argued that the claims were not novel and inventive in view of the prior art. The CNIPA disagreed, concluding that the claims were novel and inventive because none of cited prior art references disclosed at least certain distinguishing features involving backend processing.1

Why Top 10?

1. Clarification of Claim Scope of Self-defined Terms

The court’s ruling provides essential guidance on how to determine the protection scope for claim terms. In general, claim terms should be interpreted as their ordinary meaning in the relevant field. For self-defined terms, the interpretation should be based on limitations in the claims, combined with teachings in the specification, to determine the technical problem to be solved and the technical effects achieved.

Additionally, the prosecution history files (e.g., OA responses) are accessible by the public and can also be used to explain the claims.

When interpreting the claims, internal evidence should be considered first (specification, drawings, prosecution files); if still unclear, external sources like technical textbooks shall be considered. Expert infringement opinions are for reference only and are not determinative.

2. Backend Processing Disclosure in human-computer interaction related inventions

In human-computer interactions, backend processing serves as a bridge between frontend input operations and output results, typically remaining invisible to users. When the backend processing is not explicitly disclosed in the prior art, the analysis should adopt the perspective of a person skilled in the relevant technical field. The standard is whether one skilled in the art can directly and unambiguously identify the backend processing method by objectively examining technical features demonstrated during software interactions based on user input operations and the corresponding output results.

Additionally, technical features related to backend processing should be considered holistically with frontend interaction characteristics.

EIP thoughts

This case serves as a reminder of the complexities involved in patent prosecution, especially in rapidly evolving fields like human-computer interaction technologies. Here are some practical steps to consider:

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

  1. Direct English translation of the Court’s comment: “when it is determined that there is no engagement touch event occurring in proximity to the clutch user interface element and a selection touch event has occurred in proximity to the target user interface element, the target element identifier associated with the target user interface element is stored in memory to select the target user interface element for manipulation, when it is determined that an engagement touch event has begun occurring in proximity to the clutch user interface element, the target element identifier is retrieved from memory, and the target user interface element is associated with an operational function, consequently, when the engagement touch event is occurring, the manipulation function associated with the target user interface element is activated by a processor”

    ↩︎

Beijing Supreme People’s Court Upholds Novo Nordisk Semaglutide Patent in China based on Post-Filing Data

On December 31, 2025, The Supreme People's Court upheld the Beijing IP Court's decision, meaning that the semaglutide patent remains valid and will expire on March 20, 2026. Novo Nordisk announced this positive news via a press release on the day the case was decided. The Chinese Ministry of Commerce has also confirmed this decision on their website.

This case is certainly positive news for those in the biopharmaceutical industry thinking of developing their drugs in China. Additionally, if you've been following this case, you'll know that the validity of the patent came down to whether the Court would accept post-filing data demonstrating semaglutide's surprising pharmacokinetic effects in animal models. After all, there was no experimental data in the original application as filed (which is why the CNIPA originally rejected the case).

This high-profile case further demonstrates China's willingness to consider post-filing data, provided that the data are submitted under China's stricter post-filing data requirements. In this case, the specification provided sufficient "proof" that the ideas of the surprising technical effects were already established in the original application as filed, even if the data were not present.

Thin Disclosure but "Bulletproof" Protection: How a Salt Form Patent was Upheld in China Despite Minimal Data

In China, innovative drug companies typically rely on their initial composition of matter patents to block out competitors. Other “follow-on” IP, such as those directed towards formulations, polymorphs, and salt forms, are often regarded as less likely to survive invalidation challenges, especially in China where the standard of “person skilled in the art” is so high.

Having said that, we share below today a positive counter example case in which a salt form of a JAK inhibitor survived invalidation. This case is one of CNIPA’s Top Ten Patent Re-examination and Invalidation Cases for 2024 (published in the 2025 list). These cases are meant to be guiding cases, showcasing exemplary real-world decisions that clarify certain aspects of the law.

Background

Incyte Corporation is owner of Chinese patent ZL200880102903.3 (‘033 patent) which claims three salt forms of Ruxolitinib ((R)-3-(4-(7H-pyrrolo[2,3-d]pyrimidin-4-yl)-lH-pyrazol-l- yl)-3-cyclopentylpropanenitrile). Ruxolitinib is a Janus Kinase (JAK) inhibitor used to treat certain blood cancers and was disclosed in an earlier Incyte patent application (US 11/637,545, hereinafter the ‘545 application).

Petitioner Chongqing Pharscin filed an invalidation request based on Articles 26.3, 26.4, (29) and 22.3 of the Chinese Patent Law. The petitioner argued that the ‘033 patent was invalid for lack of inventiveness and insufficient disclosure, alleging that the patent had an invalid priority claim and thus was not inventive in view of its own earlier Ruxolitinib prior art. If Incyte could not establish a valid claim to priority, its earlier US application directed towards Ruxolitinib would become the closest (and likely destructive) prior art.

After detailed analysis of a variety of issues, the CNIPA Patent Reexamination and Invalidation Department (PRD) upheld all claims as valid (Decision No. 4W117594).

What happened? How did this patent survive the allegations of insufficient disclosure?

This case highlights how invention patents in the chemical and pharmaceutical arts are examined in three key aspects:

  1. Priority entitlement and “same subject matter” in highly unpredictable fields
  2. Cited foreign documents “incorporated” into a priority application’s specification
  3. Sufficiency of disclosure for salt-form inventions

A Seemingly Faulty Specification

At first glance, the ‘033 specification appeared to be missing a lot of important details, such as the compounds’ abilities to inhibit JAK activity and treat JAK-related diseases. For example, when describing the results of Example A (an in vitro JAK kinase inhibition assay), the specification only stated “[b]oth the phosphoric acid[(phosphate)] salt of the invention, and the corresponding free base compound, were found to have IC50 values of less than 50 nM for each of JAKl, JAK2, and JAK3.” However, the specification failed to explain what technical effects were associated with IC50 values of less than 50 nM. Furthermore, out of the three salt forms (maleic acid, sulfuric acid and phosphate salts), only the phosphate salt had associated data. There were also no clinical data regarding the diseases that could be treated.

The priority document US 60/943,705 (filed on 13 Jun 2007) did not recite Example A either, making it questionable whether the priority claim was valid.

In short, this ‘033 Chinese patent appeared to be missing a lot of key information:

Priority Claiming in China

In order to properly claim priority in China, inventions need to have the “same subject matter of invention” as those in the priority application. The definition of the “same subject matter of invention” is similar to the standard for novelty: the same technical field, the same technical solution, the same technical problem to be solved, and the same expected technical effect.

As Example A was not present in the priority document, one could not directly determine whether the priority document shared the same expected technical effect with the ‘033 patent. At the same time, the advantageous properties of the salt forms over the free base form were clearly stated in the specification.

What ultimately saved the priority claim was a statement in the priority application’s background section that cited an even earlier patent US 11/637,545. This, together with the common knowledge in field of JAK research, made it such that the priority document included the same subject matter as the invention.

Cited Foreign Document Treated as Part of the Priority Document

This decision recognizes that an earlier foreign patent application (US 11/637,54) merely cited in the priority application’s specification is treated as part of the priority document itself, sort of as if it were incorporated by reference. Despite earlier Examination Guidelines requiring cited documents to be published before the priority date, the PRD applied the 2023 Guidelines retroactively in light of legislative intent and interest balancing, holding that the cited US application forms part of the priority document itself for assessing technical effects (despite the US application publishing after the Chinese application was filed).

“Same Subject Matter” Established via Common General Knowledge

In order to determine whether a priority claim is valid, the priority document and the present application must share “the same subject matter,” and “the same subject matter” is defined by having the same technical field, the same technical problem to be solved, the same technical solution and the same technical effects. As summarized in the invalidation decision:

In the field of chemistry and pharmaceuticals, due to the high degree of unpredictability of technical effects, it is generally necessary to include in the specification sufficient qualitative or quantitative experimental data to demonstrate the achievement of the stated purpose and/or expected effect, so as to assure those skilled in the art that the invention can achieve the stated purpose or effect. Accordingly, in the determination of “the same subject matter” in the field and possess the same technical effect [for determining whether priority claim is valid], there is a high degree of reliance on experimental data.

With the high standard on experimental data, even without Example A in the priority document, a person skilled in the art could reproduce the JAK assay from the cited US 11/637,545 and would immediately recognize that the phosphate (and other) salts retain the same potent JAK-inhibitory profile as the free base. Therefore, a person skilled in the art would understand that the priority document and the present application have the same technical effects and share “the same subject matter,” making the priority claim valid.

Sufficiency of Disclosure Upheld

The PRD rejected the insufficiency attack on three independent grounds, all rooted in the low predictability threshold required when the invention is a pharmaceutically acceptable salt of a well-characterized active compound.

First, salt formation and characterization are routine operations. As no special conditions or unexpected difficulties were provided as evidence by the petitioner, the preparation methods and structural confirmation of the salts were deemed sufficiently enabled by a person skilled in the art.

Regarding the point that only the phosphate salt’s result was disclosed: Because the three claimed salts share the same cationic core, the activity of the other two salts is predictable to a person skilled in the art and can be tested by the experimental methods cited in the specification. Therefore, the disclosure of only the phosphate salt’s result was sufficient to enable the others without undue experimentation.

Most importantly, regarding the claimed therapeutic use/use in manufacture of therapeutic drugs: The specification and the priority document cited multiple references in the background to show that researchers in this art have long focused on and studied the relationship between JAK and various diseases. Although the ‘033 patent discloses no experimental data or clinical results specifically linking the three salts to disease treatment, a person skilled in the art would still understand that the claimed salts, which possess the similar JAK-inhibitory activity, have potential therapeutic use for those diseases.

EIP Thoughts

Being the first ever JAK inhibitor approved, Ruxolitinib remains highly significant in the pharmaceutical market due to its inhibition of a fundamental pathway. Numerous clinical trials are ongoing to expand its indications. Several other companies have petitioned to invalidate this salt form patent, and one of the petitions challenging disclosure sufficiency was rejected under the "一事不再理原则" (ne bis in idem principle).

This patent remains robust in China, likely due to its narrow scope limited to the specific salt form compounds, which have different structures than the prior art compounds at that time. Furthermore, the surprising technical effects of the salts are clearly stated in the specification. While the specific patent strategy used to protect these Ruxolitinib salt forms might not be 100% applicable in other cases, this invalidation decision still provides some useful insights regarding priority entitlement, the determination of “the same subject matter,” the role of cited foreign documents, and sufficient disclosure. It provides a positive example of how technical effects can still be fully considered even when the priority document and the later application are not identical.

Further, as seen in this decision, the determination of whether two applications include the same subject matter in this particular field involves more common knowledge in the art, including the effect of Ruxolitinib itself and importantly, the common knowledge that salt formation does not change the property or effect of the drug. A person skilled in the art would recognize the claimed salt as a direct, foreseeable improvement derived from the parent compound.

Our general practice is not to leave out certain disclosures and rely solely on incorporation by reference. While the Examination Guidelines permit examiners to consult cited documents when necessary, Part II, Chapter 2, Section 2.2.6 also explicitly cautions that content essential for meeting Article 26(3) (sufficiency/enablement) “should not be written by citing other documents, but should be specifically included in the specification.” Omitting comparative examples, characterization data, or key technical effects and expecting examiners to retrieve them from prior applications or literature carries unnecessary risk of rejection for insufficient disclosure.

Therefore, the safest and most examiner-friendly approach is to incorporate all enabling and distinguishing technical information directly into the Chinese specification itself.

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

The Secret Prior Art Trap: A Case Study on Conflicting Applications in China

In patent prosecution, the concept of a "conflicting application" or “secret prior art” plays a critical role in determining the patentability of an invention. This article explores the framework for assessing conflicting applications under Chinese patent law, with a comparative perspective on the approaches adopted in the United States.

Case Background: The Image Encoding Dispute

The subject of the appeal was a patent application for an image encoding technology, filed on June 5, 2019. The applicant had not claimed any priority. The CNIPA Examiner had issued a Final Rejection, asserting that the application lacked novelty in light of two prior PCT international applications:

  1. D1-WO​ (WO2020207498A1): Claiming a single priority date of April 12, 2019.
  2. D2-WO​ (WO2020227393A1): Claiming three priority dates, the earliest being May 9, 2019.

The applicant appealed this decision, leading to a thorough review by the Re-examination Board (the Board). The Board demonstrated how to verify the “four elements” utilizing this case.1

Question 1: what is qualified as a valid publication to be considered a conflicting application?

Under Chinese patent law, a PCT application only acquires the status of a potential conflicting application after it has entered the Chinese national phase and has been published by the CNIPA. Chinese examiners should not use PCT publications as prior art directly, but instead should cite their Chinese counterparts. In this case, both of the PCT applications had corresponding CN national applications, allowing them to meet the “territory” elements (filed before CNIPA). In this case, the Board continued its review by replacing the PCT publications with their corresponding Chinese national phase publications: D1-CN​ and D2-CN.

Question 2: how can priority affect a conflicting application

The other elements were then reviewed for both D1-CN and D2-CN.

D1-CN: The subject application’s filing date (June 5, 2019) fell between D1-CN’s priority date (April 12, 2019) and its filing date (April 13, 2020). This makes priority verification essential.

D1-CN contained:

Only the content entitled to the April 12, 2019 date satisfied the "time" element for a conflicting application (filed before, published after) and disclosed identical technical solutions to those in the application in question, thereby meeting the “content” element. Consequently, D1-CN was deemed a valid conflicting application for novelty for those specific embodiments.

D2-CN: presented a more complex scenario with three priority dates. Only the earliest (May 9, 2019) was before the subject application’s filing date.

Consequently, D2-CN did NOT qualify as a conflicting application.​ The rejection based on this document was overturned.

Epilogue

The examiners in charge of this case followed up by writing an article, further suggesting that a potential conflicting application should be verified based on the following sequence: Territory → Time (with Priority Verification) → Content.​

  1. Territory: confirm if it is a national-phase publication (the CN publication), not the WO or other publications.
  2. Time (with Priority Verification): each technical embodiment must be traced back to its supporting priority document.
  3. Content: only the content that meets the “time” element and is identical to the technical solution in the application in question can be used as a conflicting application.

EIP thoughts:

Conflicting application prior art is not unique to China. The United States has a similar concept, often referred to as "secret prior art," primarily under 35 U.S.C. § 102(a)(2)2.

To facilitate understanding, we provide a comparison table below.

Feature

China (CN)

United States (USPTO)

Term for Concept​

Conflicting Application (抵触申请)

Secret Prior Art (Post-AIA: § 102(a)(2))

Used for Novelty?​

Yes​

Yes​

Used for Inventiveness/Non-Obviousness?​


No​



Yes


Territorial Scope​


Must have entered the Chinese national phase (CNIPA jurisdiction).



Any US patent application or WIPO publication designating the US. National phase entry is NOT required.​


Applicant's Own Prior Application​


Yes, can be a conflict​ (after Oct. 1, 2009).

3

No, if by the same inventor or commonly owned or as part of a joint research agreement.


Unlike the U.S., where a secret prior art reference can be used to challenge both novelty and non-obviousness, China's approach is narrower. A conflicting application in China serves as a novelty-only bar. Its effectiveness is strictly limited to technical solutions fully supported by a valid priority claim and only after the PCT application has entered the Chinese national phase. Successfully navigating this requires meticulous priority verification and a focus on precise claim differentiation, as obviousness-type arguments are inapplicable.

Understanding these differences is essential for managing global patent portfolios and developing effective prosecution strategies in each key jurisdiction. Feel free to contact us at [email protected] if you have further questions.

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

  1. In China, a conflicting application is a patent application that:
    - Was for the same invention
    - Was filed by any applicant​ (including the applicant itself under certain conditions)
    - Was filed before the China National Intellectual Property Administration (CNIPA).
    - Was filed before​ and published on or after​ the filing date (or priority date) of the subject application.
    This is normally understood as “four elements” for conflicting applications.
    Examination Guideline, Part Two, Chapter III, Section 2.2
    ↩︎
  2. A United States patent or a WIPO-published application designating the US ) will be considered prior art against a later-filed application if it:
    - Was filed by another​ (not the same inventor),
    - Was effectively filed​ before the later application's effective filing date, and
    - Was granted or published​ on or after the later application's effective filing date.
    35 USC 102(a)(2)
    ↩︎
  3. Before Oct. 1, 2009, only applicants filed by others were considered as conflicting applications.
    ↩︎

From Faces to Football Fields: How Applying AI Models to Different Application Fields Can Prove Inventive in China

As AI models become versatile and adaptable across multiple contexts and industries, questions about inventive step sit at the heart of patent examination in China. The 2023 Patent Examination Guidelines (hereinafter referred to as 'the Guidelines') included several sessions dedicated to explaining how inventive step should be examined for AI-related inventions, featuring examples related to AI algorithms, big data, and user experience1.

In the re-examination of Beijing ByteDance Network Technology Co. Ltd.’s application entitled “Method and apparatus for processing an image” (Application No. 201810734681.2)2, which was recognized as one of the Top 10 Re-examination Cases of 2025 by the CNIPA, the CNIPA focused on the following issue: when an existing algorithm or AI model is applied to a different application field compared to the prior art, how should its inventive step be judged?

Summary of the Invention

The invention relates to methods of processing captured images using a key point detection model. Its core lies in inputting a captured image of a target object—such as a sports field or other pre-defined object—into a pre-trained key point detection model that corresponds specifically to that target object. A "key point" is a pre-determined position on the target object (for instance, the corner point of a football goalpost).

The model outputs a position information set, which consists of: (1) Position coordinates indicating where the key points appear within the captured image; and (2) Visibility information reflecting the probability that the key point is actually visible in the captured image. These output data enable the system to recognize and align the same target object across multiple images.

One exemplary embodiment highlighted in the application was in the context of a football field. Using the pre-trained model of a specific football field, the method could analyze multiple frames of the same football field, determine geometric transformations across frames, and thus enable game data analysis.

A diagram of a football field<br><br>AI-generated content may be incorrect.

Prosecution History

The claims were initially rejected for lack of inventive step over the combination of D1, D2 and common knowledge in the art under Article 22.3 of the Patent Law.

Amendments

During re-examination, the applicant amended the independent claims to:

PRD’s Decision

Upon reviewing the amended claims and the applicant’s arguments, the board at the Patent Reexamination and Invalidation Department (PRD) (hereinafter referred to as “the Board”) revoked the examiner’s rejections, concluding that the amended claims did in fact possess an inventive step based on the reasons below.

Differences from Prior Art

The Board identified the following distinguishing features of the amended claims compared to the prior art:

Technical Problem Solved

Based on the distinguishing features above, the Board identified the technical problem addressed by the invention as how to determine the correspondence of the same target object across multiple captured images. This was materially different from what D1 and D2 achieved, which only addressed the quality of a single facial image.

The Board concluded that there was no technical inspiration in D1 or D2 to adapt their teachings to address the technical problem solved by the present invention. Accordingly, the amended claims were considered inventive.

EIP Thoughts

This case demonstrates both the challenges and opportunities for applicants seeking to protect AI-related inventions in China. On one hand, it appears that the. CNIPA maintains a close watch against attempts to claim generic algorithms without limiting to certain specific application field(s) to sufficiently distinguish from known art. On the other hand, if an applicant can pinpoint a specific technical problem in a particular application field, provide tailored model training, and identify distinct output uses that result in surprisingly improved technical effects not disclosed in the known prior art, they may have a strong chance of establishing inventive step.

Under the three-step framework for assessing inventive step, the Board emphasized that examiners must consider whether the change of application field results in substantial adjustments or modifications to the algorithm or model. Specifically, where a new application field leads to significant changes in:

and such differences produce a non-obvious technical effect relative to prior art, the invention should be recognized as inventive.

The reasoning parallels the 2022 Top 10 case3 of “Method for establishing a neural network model for classifying grades of scrap steel”, where the inventive step was recognized based on the adaptation of a neural network model to a highly specialized industrial context.

For patent applicants and patentees of AI-related inventions, this case highlights some strategic considerations when drafting the applications:

Looking ahead, similar disputes will almost certainly arise as AI technologies continue to develop and overlap across industries. The lesson is clear: broad yet detailed examples, explicit technical problems, and clear linkages between model training and application scenarios are extremely important.

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

  1. Sessions 6.1.3 and 6.2, Part II Chapter 9, Patent Examination Guidelines 2023

    ↩︎

  2. https://www.cnipa.gov.cn/art/2025/6/6/art_2648_200004.html

    ↩︎
  3. https://www.cnipa.gov.cn/art/2023/4/26/art_3207_184728.html ↩︎

When Standards Evolve: How China Judges Inventiveness in Next-Gen Tech Patents

Starting in 2023, Datang Mobile, one of the key players in China’s telecom standardization efforts, initiated infringement proceedings against Samsung in China, Germany, and the US. Samsung responded by filing multiple invalidation petitions in several jurisdictions, including China, the United States, and Europe. The case discussed below is one of above-mentioned invalidation cases in China, and was selected as one of the Top Ten Reexamination and Invalidation Cases of 2024 by CNIPA as

“a typical patent case in the field of communications involving standards evolution in that its assessment of inventiveness provides guidance for this type of case.”1

Legal & Technical Focus

The patent involved is titled “Carrier Aggregation Feedback Method, Device, and System,”2 owned by Datang Mobile3. Carrier aggregation (CA) is a key technology introduced in LTE-Advanced (4G) and further developed in LTE-Advanced Pro (sometimes referred to as “4.5G”) and 5G. The patent concerned the evolution from 4G intra-band CA to 4.5G inter-band CA.

The core issue of the invalidation request raised by the petitioner, Samsung4, was whether the patented invention—designed for inter-band CA scenarios—was obvious in view of prior art based on earlier intra-band CA standards.

Key considerations around the obviousness assessment included:

  1. The petitioner argued that combining 4G prior art with 4.5G references rendered the claim obvious.
  2. The patent’s central distinguishing feature was the method and system for determining the maximum number of subframes requiring feedback (M) based on carrier configurations from the system information tables—rather than from actual usage data as in the prior art.
  3. The Patent Office determined that the prior art did not disclose or suggest deriving M from configuration tables, nor did it provide a clear technical motivation for such a modification.
  4. Therefore, the invention solved a new problem specific to inter-band CA scenarios and achieved a technical effect of improved feedback efficiency.

Ultimately, the patent was maintained in amended form, and the parties later reached a global settlement.

Guidelines from the Examiners

In a subsequent CNIPA article5, the examiners in charge summarized their methodology for reviewing the inventiveness of the disputed patent. This methodoloy not only shows the CNIPA internal standards, it also practically serves as a guideline for assessing inventiveness of similar type of cases in the future.

Step 1: Determining, relative to the prior art, in which aspects the intergenerational change of the patent at issue is embodied—whether it lies in a change of system architecture, or in changes of technical scenarios or user applications;

Step 2: Determining the relationship between the technical problem solved by the distinguishing features and the intergenerational change—namely, whether the problem is a new problem introduced by the intergenerational change, or whether it is a traditional problem that has long existed in the development of communication technologies; and

Step 3: Making an overall assessment of whether the prior art provides any technical motivation, by standing in the position of a person skilled in the art, grasping the entire inventive concept, and at the same time paying attention to the technical field, technical problem, technical means, and technical effect.

EIP Thoughts

This case highlights the complexity of assessing inventive step in rapidly evolving industries like telecommunications. When standards evolve from one generation to another, the legal question is not simply whether features are known, but whether there is a real technical motivation to combine them across different contexts.

Key takeaways include:

  1. Standard evolution matters: Moving from 4G to 4.5G did not merely extend existing techniques; it introduced new technical problems (e.g., inter-band configuration mismatches).
  2. Problem-solution analysis is central: The invention’s distinguishing feature addressed a problem that did not exist in the 4G scenario.
  3. Combination obstacles: Even if individual features are known, if the prior art operates in different technical contexts, there may be no reasonable teaching or motivation to combine them.
  4. Strategic implications: For companies in standard-driven fields, careful patent drafting to emphasize “new scenario” problems and technical effects can make patents more robust in invalidation proceedings.

For patent applicants and owners, especially outside the telecom sector, this case offers broader lessons:

  1. Highlight the “new scenario”: When drafting applications, clearly explain how the invention addresses problems unique to the next generation of technology.
  2. Differentiate timing and context: Small distinctions, such as when feedback is determined (at configuration vs. during operation), can carry significant weight in validity disputes.
  3. Anticipate standard evolution: Patents that bridge multiple generations of standards often become central to licensing negotiations and litigation.
  4. Global litigation dynamics: Standard-essential patents (SEPs) frequently trigger parallel disputes across jurisdictions; consistent claim drafting and prosecution strategies are crucial.

The “Carrier Aggregation Feedback” case is a textbook example of how technical innovation, standard evolution, and global patent litigation intersect. For industries where standards evolve quickly—telecommunications, computer networking, and beyond—patent protection strategies should not only focus on current implementations but also anticipate how future standards may reshape technical problems and solutions.

If you would like to have more information on this matter or would like to have our advice, please feel free to contact us at [email protected].

This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.

  1. “Case Five: Invalidation Request for the Invention Patent ‘Carrier Aggregation Feedback Method, Device, and System’ 案例五：“载波聚合反馈方法、装置及系统”发明专利权无效宣告请求,” https://www.cnipa.gov.cn/art/2025/6/6/art_2648_200008.html, published on June 6, 2025, accessed on September 5, 2025.

    ↩︎

  2. Patent No.: ZL201210137097.1

    ↩︎

  3. Datang Mobile Communications Equipment Co., Ltd.

    ↩︎

  4. Samsung (China) Investment Co., Ltd.

    ↩︎

  5. “Case Five: Invalidation Request for the Invention Patent ‘Carrier Aggregation Feedback Method, Device, and System’ 案例五：“载波聚合反馈方法、装置及系统”发明专利权无效宣告请求,” https://www.cnipa.gov.cn/art/2025/6/6/art_2648_200008.html, published on June 6, 2025, accessed on September 5, 2025.

    ↩︎
Top