Should the Experimental Use Exception Be Broadened?

On June 28, 2024, the United States Patent and Trademark Office (“USPTO”) requested the public’s views on the current state of the common law experimental use exception to patent infringement and whether legislative action should be considered to enact a statutory exception (89 FR 53963). Coincidentally, the International Association for the Protection of Intellectual Property (“AIPPI”) was also examining this issue and prepared a report based on submissions from various countries. On October 25, 2023, the Standing Committee on Pharma of the AIPPI issued a report summarizing the results of the investigation. It is summarized below, along with brief summaries of how various countries handle the experimental use exception.

A comparison of the experimental use exception across different jurisdictions can help readers prepare informed responses to the USPTO’s request for comments.

1. United States

The experimental use exception to patent infringement originated in Whittemore v. Cutter, 29 Fed. Cas. 1120 (C.C.D. Mass. 1813). The exception is an affirmative defense on which the defendant bears the burden of proof. It is “very narrow and limited to actions performed ‘for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry.’” Madey v. Duke University, 307 F.3d 1351, 1362 (Fed. Cir. 2002) (quoting Embrex, Inc. v. Service Engineering Corp., 216 F.3d 1343, 1349 (Fed. Cir. 2000)). Furthermore, the exception does not “immunize use [of a patented invention] that is in any way commercial in nature” nor ”any conduct that is in keeping with the alleged infringer’s legitimate business.“ Id. at 1362. While the exception continues to exist, it is “in a very limited form.” Id. at 1360. Third parties who assist one protected by the Hatch-Waxman safe harbor (35 U.S.C. § 271(e)(1)) are themselves protected. It is unclear whether this third-party protection extends to the experimental use exception.

In the seminal decision of Roche Products, Inc. v. Bolar Pharmaceuticals Co, 733 F. 2d 858 (Fed. Cir. 1984), the Federal Circuit held that the defense did not apply, and that Bolar infringed a Roche patent. Bolar had used the patented Roche product prior to patent expiration to prepare a Food and Drug Administration submission to enable it to market its own version of the patented drug after the Roche patent expired. This decision was subsequently overruled with the enactment of the Hatch-Waxman safe harbor. Since the Hatch-Waxman safe harbor does not exempt “basic research,” Merck KGaA v. Integra Life Sciences I, Ltd., 545 U.S. 193, 205 (2005), essentially all experimentation with a patented invention may be subject to an infringement allegation. It is noteworthy that in Madey, Roche, and Embrex, the accused use had distinct commercial purposes and was not undertaken to improve or advance a patented invention.

As the following discussion demonstrates, the situation in the U.S. is radically different from the law in various European countries. The laws, however, seem consistent on two points: (1) the use of a research tool is not exempt, Proveris Sci. Corp. v. Innovasystems, Inc, 536 F.3d 1256, 1265 (Fed. Cir. 2008); and (2) stockpiling is not exempt. Amgen, Inc. v. Hospira, Inc., 944 F.3d 1327 (Fed. Cir. 2019).

2. AIPPI

The AIPPI’s position essential adopts the conclusions set forth in the Tokyo Resolution, 3.2, 3.3. The defense allows testing to evaluate the teachings and validity of a patent, including the scope of its protection. It also allows experimentation for improving the invention, advancing the invention, or finding alternatives for making or using the invention. It does not, however, cover the commercial exploitation of the patented invention. Third parties assisting one in the performance of conduct protected by the experimental use defense are themselves protected, even if the former has a commercial purpose. The exception, however, is narrowly construed by the courts.

Experiments using a research tool for its claimed or originally intended use do not qualify as an experimental use, even if used for experimental or non-commercial purposes. The claimed or originally intended use of the tool is determined by the applicable patent specification. However, use of information gained from improper use of a research tool is not prevented. Lastly, stockpiling is not protected as an experimental use.

3. United Kingdom

The patent laws of the UK dictate that acts done for experimental purposes are protected from infringement. Section 60(5)(b) Patents Act 1977 states that an activity will not constitute patent infringement if it is “done for experimental purposes relating to the subject-matter of the invention.” This applies to all patented subject-matter and is not limited to only academic activities having no commercial value. The exception serves to protect work that is directed to improving upon the patented invention rather than just using it.

To fall within the exception, the experiment must be carried out to discover something unknown, test a hypothesis, or find out whether something that is known to work in specific conditions will work under different conditions. It is important to note that tests conducted for demonstrative purposes or to amass information to satisfy a third party (such as a regulatory body) do not constitute experimental use. If an experimental act has multiple objectives, the defendant may rely only on the experimental use exception if the main purpose of the conduct was to generate information of scientific or technical value. The scale of the act is also important. A large-scale trial of a product that involves a supply of products in quantities that exceed those required to test how the product works may fall outside the exception.

The defendant bears the burden of proof to show that: (i) the acts complained of were done for experimental purposes; and (ii) those purposes relate to the subject-matter of the invention. If the patented invention is being used as a research tool, the subject-matter of the invention must be the focus of the experimental work to fall within the exception. A third party committing any of the cited acts in the UK risks being liable for patent infringement unless they can rely on the experimental use exception. The UKIPO Guidance suggests that the commercial supply by a third party to the party using the patented product would not be covered by the experimental use exception, which is a position also taken by the UK courts. Despite this, the extent to which other third parties (like those engaged by the sponsor of a clinical trial) could benefit from the experimental use exception remains unclear. Stockpiling is not protected as an experimental use.

4. Germany

In Germany, the experimental use exception is a defense in infringement proceedings. Section 11.2 of the German Patent Act states that an experiment can be excluded from the effect of the patent if and only if the research objective is related to the patented subject matter. As such, Germany defines an experiment as any planned act directed to gain knowledge, irrespective of the final purpose of this knowledge. Even if an experiment ultimately has a commercial purpose, it can fall under experimental use if there is a knowledge-seeking component. In sum, Germany has a broad interpretation of the experimental use exception.

Regarding the use of patented inventions as research tools, the German Patent Act specifically allows it very narrowly if the usage is for the purpose of breeding, finding, or developing a new plant variety (§ 11 No. 2b of the German Patent Act). It also allows for the use of patented experimental or clinical data for the development of new health technologies and the general advancement of science.

As always, the burden of proof is on the party that asserts the exception. There has been very little case law on the experimental use exception in Germany, so there has been limited guidance concerning how far the exception extends to acts of contributors, such as suppliers of a protected molecule. It remains unclear whether a third party contributing to the privileged act is protected by the exemption. If, however, the third party is a “co-experimenter” with a protected party, the exemption would apply. When asked if Germany wanted to modify its experimental use exception, commentators stated that further clarification regarding third party contributors should be included in the law. Stockpiling is not exempt unless covered by a Supplementary Protection Certificate “SPC” manufacturing waiver.

5. The Netherlands

Article 53 (3) of the Dutch Patent Act contains the experimental use exception to the rights of a patentee. It states that “the exclusive right [of a patent] shall not extend to acts solely serving for research on the patented subject matter, including the product obtained directly as a result of using the patented process.” Experimental use is considered a defense to patent infringement in the Netherlands, though the Dutch Supreme Court has held that the provision is to be interpreted narrowly.

The Dutch Patent Act implies that strictly academic and private non-commercial acts will fall within experimental use and are not covered by a patent. But the research exception is not limited to academic or private non-commercial acts, however. Acts that have been considered to fall within the scope of the research exemption include evaluating whether to take a license, further developing technology, and identifying a further medical use.

There is no existing Dutch case law regarding whether the contribution by a third party would constitute patent infringement. That being said, if the contribution by a third party consists of aiding researchers in conducting experiments that fall within the research exemption, it will not constitute patent infringement. As long as the research is purely scientific in nature by all parties concerned and/or is only aimed at any purpose that fulfills the purpose of the patent law, it can fall under the exception. If the patented invention is used to aid in research, it must solely be used for research on the patented subject matter. Merely using the invention as a research tool does not fall within the exception. Stockpiling of the patent product is not exempt unless covered by the SPC manufacturing waiver.

6. Italy

In Italy, the experimental use exception is provided in Article 68(1) of the Industrial Property Code and is considered a defense to patent infringement. Relevant case law in Italy is limited. The country excludes experimental acts relating to the use of biological material for cultivation purposes or the discovery of other plant varieties from the effect of a patent. Whether courts interpret the exception as broad or narrow has been authoritatively argued between research ”having as its object” the patented invention (which can be covered by the exception) and research conducted merely ”with” the invention (which cannot). The possible presence of a commercial interest is irrelevant to the lawfulness of the experiment — the exception is not only limited to acts done for “academic purposes” having no “commercial value.” Though there is no established guideline, the rationale and legal wording suggests that the contribution of third parties to the experimental act does not negate the applicability of the exception. As expected, the burden of proof lies on the party who raises the defense. It is likely that the use of a research tool is not protected. Stockpiling is not permitted.

7. Sweden

Like the aforementioned countries, the experimental use exception in Sweden, Swedish Patent Act, Article 3, para. 3, item 3, excludes the “use of the invention for experiments which relate to the invention itself” from patent infringement and is used as a defense. As long as the experiment refers to the patented invention, the exception can even be applied to acts having commercial value. The burden of proof lies on the party who invokes the exception as a defense. As long as the main act carried out by the original party falls within experimental use, any contributions made by a third party will not constitute patent infringement. If the patented product is used as a tool for research with the purpose of making a new invention, the exemption does not apply. Although there is no established case law, Swedish legal literature guides that further development in medically therapeutic substances can fall within experimental use if a new effect is discovered or a smaller number of substances with better results can be uncovered. Essentially, if the research results in a product with added value, it is likely it can be covered by the experimental use exception. Like other European countries, stockpiling is not exempt except for the SPC manufacturing waiver. Similarly, the use of a research tool is not exempt.

8. Spain

Experimental use is considered as a right in Spain as opposed to a defense as stated in Spanish Patent Act 24/2015, Article 61.1.b. It has been interpreted narrowly by Spanish courts, though the exception is not limited to acts done for academic purposes. Experiments merely using the object of an invention for any purposes not related to the patented subject matter will be considered infringement. Contribution to the experiment by a third party will be protected only if the original act falls within the scope of the exception. Spanish law does not have any provisions regarding the use of the invention as a research tool. However, it seems clear that if used in an experiment, the research must be related to the original invention to be excluded from patent infringement. Stockpiling is not exempt unless the SPC manufacturing waiver exists.

9. Hungary

In Hungary, the experimental use law is considered both a right and defense. In Hungary, the experimental use exception and protection under the Bolar amendment are not separated. Hungary Patent Act, Article 19(6). The extent of the experimental act is important. If the amount and source of the patent substance is considered to be more than reasonably required, then the exception is unlikely to apply. The acts should be pursued for scientific or academic purchases, and if the party does not engage in such activities, the court may consider this a factor against applying the exception. As opposed to other countries, Hungary’s Patent Act states that third-party contributions to the original experiment constitute patent infringement. It notes that third-party contributions usually do not serve experimental purposes and are highly likely to be regarded as purely commercial interests, and therefore outside of the exemption’s scope. Hungary does not allow for the exploitation of patented inventions as research tools. Stockpiling is not exempt except for the SPC manufacturing waiver.

Analysis

There are important distinctions between the experimental use exception in the U.S. and European countries. However, a few similarities exist. The use of a research tool is not exempt in the U.S. or in other countries. Moreover, stockpiling — usually viewed as a part of the Bolar amendment in Europe — is not protected. Arguably, it should be for the reasons set forth below.

I. Stockpiling

The legislative history and court decisions indicate that stockpiling should be permitted under the Hatch-Waxman safe harbor. Yet, it is not. Courts have agreed that “[t]he basic idea behind [271(e)(1)] was to allow competitors to begin the regulatory approval process while the patent was still in force, followed by market entry immediately up on patent expiration.” Proveris Scientific Corp. v. Innovasystems, Inc., 536 F.3d 1256, 1261 (Fed. Cir. 2008) (emphasis added). As such, “Congress must have intended to allow competitors to be in a position to market their products as soon as it was legally permissible.” Telectronics Pacing Sys, Inc. v. Ventritex, Inc. 982 F. 2d 1520, 1525 (Fed. Cir. 1992) (emphasis added). Section 271(e)(1) was enacted to allow competitors “to enter the commercial marketplace in a large scale way as soon as the relevant patents expired.” Intermedics Inc v. Ventritex, Inc. 775 F. Supp. 1269, 1277 (N.D. Cal. 1991) (emphasis added), aff’d sub nom, 991 F.2d 808 (Fed. Cir. 1993).

The Act’s legislative history demonstrates that the goal of safe harbor was to give the public immediate access to generic products after patent expiration. For example, House Reports state: “It is the Committee’s view that experimental activity does not have any adverse economic impact on the patent owner’s exclusivity during the life of a patent but preventing of such activity would extend the patent’s owner’s commercial exclusivity beyond the patent expiration date. H.R. Rep. No. 98-857, pt. 1 at 46 (1984). Further, “[The Committee on Energy and Commerce] reasoned that with [§ 271 (e)(1 )] generic manufacturers would be required to engage in ... bioequivalency tests after the expiration of the patent. This would result in delays of about two years after the expiration of the patent before a generic could go on the market." H.R. Rep. 98-857, pt. 2 at 8-9 (1984).

Under the Federal Circuit’s decision in Amgen v. Hospira, supra, stockpiling is not exempt under the safe harbor. Thus, a competing product cannot be marketed “as soon as the relevant patent expires.” Certainly, a competitor cannot enter the commercial marketplace in a “large-scale way” when the relevant patent(s) expire. The courts have not analyzed this inconsistency.

II. Scope of Protection

Despite the Madey, Roche, and Embrex decisions, there is little guidance on the precise scope of the experimental use exception. Regardless, it is narrow and “in a very limited form.” Thus, the scope of the U.S. exception is significantly different from that of European countries. One could reasonably contend that the European view, which specifically allows experimentation for improving and/or advancing a patented invention, is preferable to the narrow U.S. view.

Most decisions addressing “experimental use” relate to its applicability as a defense to a prior use invalidity challenge under Section 102. These decisions are not relevant as the purpose of the defense in a validity challenge is significantly different than when used as a defense to infringement. A number of decisions, however, do address the experimental use defense. While the facts differ, the holdings demonstrate that the defense is unlikely to succeed. See, e.g., Verinata Health, Inc. v. Ariosa Diagnostics, 329 F. Supp. 3d 1070 (N.D. Cal. 2018); Ares-Serono, Inc. v. Organon International BV, 862 F. Supp. 603 (D. Mass. 1994); Pitcairn v. United States, 547 F.2d 1106 (Fed. Cir. 1977); Baxter Diagnostics Inc. v. AVL Scientific Corp., 798 F. Supp. 612 (C.D. Cal. 1992); SmithKline Beecham Corp v. Apotex Corp. 439 F.3d 1312 (Fed. Cir. 2006); Infigen, Inc. v. Advanced Cell. Technology, Inc., 65 F. Supp. 2d 967 (W.D. Wis. 1999). In Infigen, the accused use was “part of the ongoing business activities,” but arguably it was directed to improving/enhancing the patented invention. Id. at 981. See generally Applera Corp. v. MJ Research, 311 F. Supp. 2d 293 (D. Conn. 2004); Ciese v. Pierce Chemical Corp., 29 F. Supp. 2d 33 (D. Mass. 1998).

Early decisions provide no solace to an alleged infringer. See Poppenhusen v. New York Gutta Percha Comb Co., 19 F. Cas. 1059 (C.C. S.D. N.Y 1858), Poppenhusen v. Falke, 19 F. Cas. 1048 (C.C. S.D. N.Y. 1861). Bonsack Mach. Co. v. Underwood, 73 F. 206 (C.C. N.C. 1896).

In a somewhat ambiguous decision in Chesterfield v. United States, 159 F. Supp. 371 (Ct. Cl. 1958) (per curiam), the experimental use defense succeeded. See also Ruth v. Stearns-Roger Mfg. Co., 13 F. Supp. 697 (D. Colo. 1935) (partially successful); Dugan v. Lear Avia, Inc., 55 F. Supp. 223 (S.D.N.Y. 1944) (partially successful).

Although the particular facts in the decisions vary, the clear takeaway is that as stated in Madey, the defense is “very limited,” and is unlikely to succeed under existing law. It can be argued that even if the alleged infringing conduct has some commercial benefit — maybe undirect — the conduct should be protected if it enhances or improves the patented invention, even if it otherwise relates to the user’s legitimate business activities. This would seem particularly applicable if the conduct in no way harms or affects the patentee. But see Amgen v. Hospira, supra, where significant damages were awarded for stockpiling even though the patentee’s sales were not affected.

Undoubtedly, it would be difficult to distinguish a legitimate experimental use from a purely commercial endeavor. Regardless, allowing some expanded experimental use defense by adopting the European approach would seemingly “promote the Progress of science and useful Arts” (U.S. Const., Art. 1, Sec. 8, Cl. 8) and should be considered.

---

The author wishes to thank Mia Pakola, legal intern, for her assistance with this article.