The inventiveness of crystal type patents has never been judged by “whether a new crystal mold has been made,” but rather on “whether this new mold brings unexpected technical effects.”

As a typical case selected for the top ten patent invalidation cases, the invalidation of the crystalline patent for fruquintinib reflects this examination standard and provides important references for pharmaceutical companies’ patent drafting and responses.

---

1. Case Background: Why was the crystalline patent for a domestically produced innovative drug invalidated?

Fruquintinib is a Class I innovative drug independently developed by a domestic pharmaceutical company. It passed priority review and was approved for market launch in 2018 for the treatment of metastatic colorectal cancer. As of 2023, its cumulative domestic sales have exceeded 2 billion yuan. Its compound patent has already been granted in advance, and the patent in question is a crystal form patent laid out based on a compound. Claim 1 protects the crystal form I of fruquintinib, limited by six XRD characteristic peaks.

In March 2024, the claimant for invalidation filed an invalidation request with the National Intellectual Property Administration, citing reasons including the claim not being supported by the specification and lacking novelty and inventiveness. The petitioner, based on evidence such as the marketing drug package insert for fruquintinib, argues that the active compound used in the preclinical experiment is form I, constituting public use; At the same time, it is believed that this crystal form has not achieved unexpected technical results and lacks creativity.

The patentee argued that the generic name “fruquintinib” only refers to the drug substance itself and does not involve crystal form, and the protected crystal form was not disclosed before the application date; Based on the specification and three supplementary experimental data, it is argued that the crystal form has advantages in high temperature, high humidity, and light stability, and has achieved commercial success.

After review, the panel found that the chain of evidence regarding the petitioner’s “use of public information” was insufficient to prove that Type I was already in a state where the public could easily know, and the novelty claim was not supported. However, regarding the inventiveness claim, the final invalidation request review decision No. 581990 was made, declaring the entire patent in question invalid. The examination decision provided exemplary discussions on the identification of “unexpected technical effects” and the review of supplementary experimental data.

---

2. Identifying “unexpected effects”: the more conventional the method, the higher the threshold

To understand the fruquintinib case, one must first grasp the fundamental difference in inventiveness judgment between crystal type patents and compound patents.

The inventiveness of compound patents usually lies in whether the structure itself is not obvious.

However, crystal form patents differ: different crystal forms of the same compound have identical molecular formulas, differing only in how molecules are arranged in the lattice.

For those skilled in the field, after completing compound development, continuing crystal form screening and obtaining different crystal forms by changing conditions such as crystallizing solvents or temperature is a routine step in drug development and a common pursuit in this field. This means that existing technology itself provides strong insights for “conducting crystal type screening.”

Regarding the threshold for inventiveness in crystal form patents, the Supreme People’s Court has systematically explained this in previous judicial precedents: inventions of crystal forms of known compounds usually only produce unexpected technical effects before inventiveness can be recognized; Moreover, the more conventional the method of obtaining crystal forms, the higher the requirements for effectiveness.

The review decision in the fruquintinib case applied this logic quite specifically. The patent specification in question uses 25 embodiments, employing conventional solvents such as acetone, ethanol, n-propanol, and isopropanol, achieving crystal type I over a wide temperature range. Based on this, the panel determined that under such broad conditions, the ease of obtaining this crystal form makes it extremely difficult for those skilled in the art to attempt it, and the insights from the prior art are very strong. Therefore, it must deliver extremely robust “unexpected effects” to pass.

In other words, in the examination of crystal type patents, the “difficulty of the preparation method” directly determines the “level of technical effect requirements.” The more conventional the method, the weaker the innovation, so applicants must prove themselves with extremely remarkable and unconventional effect data. If a crystal form with decent performance is obtained solely through conventional screening, such a level of technical contribution cannot support creativity under current review standards.

---

3. Review and submit supplementary experimental data: You can’t make things up out of thin air, only add embellishment

Facing extremely high examination thresholds, the patentee claimed that Type I exhibited high temperature, high humidity, and light stability, and submitted three supplementary experimental data as evidence.

This touches on a common practical controversy: Can the data for the supplementary payments be used?

According to the provisions of the Patent Examination Guidelines, the technical effects proven by supplementary data should be obtained by technical personnel in the relevant technical field from the content disclosed in the patent application.

From paragraph 9 of the specification and Example 44, the collegial panel determined that the effects of the original disclosure of this patent were concentrated on “stability” and “non-hygroscivity.” The effect demonstrated by the supplementary data is consistent with the relevant and represents a different form of technical effect recorded in the application documents, and thus was included in the review.

But “inclusion in review” and “supporting creativity” are two separate issues.

The panel further pointed out that the routine goal of crystal type screening is to achieve good stability and low moisture absorption. No crystal transfer under accelerated conditions is a conventional expectation for those skilled in the field, rather than an unexpected discovery. Therefore, the case reached a clear conclusion: although the patentee’s claim for stability falls under the category of “patent drug characteristic effects,” it has not yet reached the level of “unforeseeable.”

This also confirms the core criterion for inventiveness review of crystal form: merely achieving “better than a known crystal form” or “meeting conventional drug requirements” is not enough; To support creativity, the technology must achieve a substantial breakthrough that cannot be inferred from conventional screening expectations.

---

4. Case Insights: A Guide for Pharmaceutical Companies to Avoid Pitfalls When Planning Crystalline Patents

The value of the fruquintinib case lies not in declaring a patent invalid, but in exposing several commonly overlooked practical issues in the inventiveness judgment of crystal patents.

First, the display of embodiments should not be blindly piled up; caution should be taken to avoid backlash.

In compound patents, with many examples and a wide synthesis route, protection coverage is usually enhanced; But in crystalline patents, the logic is exactly the opposite. In this case, the specification uses 25 embodiments to demonstrate that the same crystal form can be easily obtained using various conventional solvents and at a wide temperature range. The applicant’s original intention was to prove the process was stable, but the examiner determined that since conventional means could be easily obtained, it showed a lack of technical difficulty and inventiveness. Therefore, if the target crystal type truly has preparation barriers, the manual should restrain the failure or shortcomings of conventional methods, turning “difficulty in obtaining” into a creative argument.

Second, effect comparison cannot be superficial; it is essential to precisely pinpoint the “closest known crystal type.”

In invalid programs, merely general comparisons with existing technology often fail to hold the position. Companies must clearly state in the instructions what the closest known crystal form to this drug was. Then, closely focusing on the differences from the specific crystal form, experimental data and a framework for demonstration were constructed. Only by establishing a benchmark during the application stage can your effectiveness advantage be realized in subsequent reviews or supplementary data submissions.

Third, you need to look at the “type” and “degree” of the effect; stability alone is hard to support creativity.

Stability and bioavailability belong to the “efficacy of drug characteristics,” which is the passing line for crystal type patents. However, crystal type screening itself aims for stability and non-moisture absorption, so no crystal transfer under conventional accelerated conditions is a common industry expectation. To support creativity, the specification must present that this advantage is a substantial breakthrough that cannot be inferred from conventional screening expectations. For example, in some extremely special environments, there is an unexpected retention rate, or a crystal transformation challenge that conventional methods cannot overcome. In short: the type of effect is right, but the degree beyond expectations is insufficient, still unable to support creativity.

Fourth, supplementary data cannot be created afterward; space must be reserved during the application stage.

Due to the first-to-file system, supplementary data submitted by companies during the invalidity phase cannot be used to prove effects that are not mentioned in the specification. Supplementary submission data can only be used for “reinforcement,” not for “filling in gaps.” In this case, the supplementary data was considered by the examiner because paragraph 9 of the specification and Embodiment 44 already describe stability and non-hygroscopicity. If the manual originally only states “good stability” and invalidates the data to prove “significantly better than competitors,” the examiner will reject it on the grounds that it “exceeds the original public scope.” Therefore, the specification at the application stage must establish a framework for comparative testing to allow room for future invalid procedures.

Fifth, the effort in patent layout must be done in advance, not remedial after the fact.

Crystal type screening is usually conducted after a compound patent, but when it comes to selecting comparison targets and describing effect data, it’s best to finalize them in advance when drafting the crystal type application. This requires pharmaceutical companies to continuously monitor the crystal form dynamics of global competitors from the compound R&D stage. If you wait until the patent is invalidated to look for comparison documents or piece together the effect data, it is often too late. The layout of crystalline patents essentially coordinates R&D pace and patent strategy; thorough preparation before filing is far more effective than rushed responses during invalidity procedures.

---

5. Conclusion

Looking back at the entire case, the fruquintinib case not only provided clear judicial rules regarding the use of public recognition and supplementary data review, but also redefined the inventiveness of crystalline patents at the legal level: the more conventional the preparation methods, the stricter the requirements for “unexpected technical effects.”

This ruling sends a clear industry signal: the era when new crystal types were discovered through conventional screening means immediate authorization is over. For pharmaceutical companies, the true patent barrier must be built on the pre-development layout — precisely identifying comparative targets during the application phase and solidly building core data systems. Only in this way can innovation be maintained under increasingly stringent scrutiny standards.

---

About Us

MeritsIP: Your Global Intellectual Property Partner

At MeritsIP, we provide comprehensive IP services in biomedicine, medical devices, manufacturing, semiconductors, and AI. We support 200+ top clients globally with strategic, high-quality IP services.

Stay Connected

• • Website: https://www.meritsipglobal.com

• • LinkedIn: Follow us on LinkedIn

• • Email: info@meritsip.com

© 2025 MeritsIP. All Rights Reserved.