# In re Fisher: Gene Fragments and the Limits of 'Useful'

> The Federal Circuit refused patents on five expressed sequence tags whose only disclosed uses were generic research applications, sharpening the 'specific and substantial' utility standard for the genomics era.

Topic: Patents  |  Author: Lidiia Levitska  |  Source: Intellectual Property Law (outsideipcounsel.com)
Canonical: https://outsideipcounsel.com/blog/in-re-fisher-ests-specific-substantial-utility/


*In re Fisher*, No. 04-1465, was decided by the United States Court of Appeals for the Federal Circuit on September 7, 2005, in a majority opinion by Chief Judge Michel over a vigorous dissent from Judge Rader. The case took the "specific and substantial" utility standard (descended from the Supreme Court's 1966 decision in *Brenner v. Manson*) and applied it to the defining technology of its decade: fragments of DNA whose function nobody yet knew. By affirming the rejection of claims to five expressed sequence tags, the court drew a line that still governs how the utility requirement of 35 U.S.C. § 101 treats research tools and informational molecules.

## At a glance

- **Case:** *In re Fisher*, No. 04-1465 (Fed. Cir.)
- **Decided:** September 7, 2005; majority opinion by Chief Judge Michel; Judge Rader dissenting
- **Posture:** Appeal from the Board of Patent Appeals and Interferences, which affirmed the examiner's rejection of the claims; Federal Circuit affirmed
- **Technology:** Five purified nucleic acid molecules (expressed sequence tags, or ESTs) derived from maize, encoding portions of proteins of unknown function
- **Holding:** The claimed ESTs lacked utility under § 101 (and failed § 112 enablement for the same reason) because their only disclosed uses were generic to any EST and provided no specific, substantial, presently available benefit
- **Status:** Controlling Federal Circuit precedent on EST and research-tool utility

## What an EST is, and why utility was contested

An expressed sequence tag is a short sub-sequence of a transcribed gene. Researchers generate ESTs by sequencing fragments of complementary DNA; each tag corresponds to part of a gene that a cell was actively expressing. ESTs are immensely useful *as a class* for the genomics enterprise. They help locate genes, build genetic maps, and identify which genes are switched on in a given tissue. But an individual EST, standing alone, often tells you little about the function of the full-length gene from which it came or the protein that gene encodes.

The applicant (assignee Monsanto) claimed five specific ESTs from maize. The specification did not identify the function of the underlying genes or the proteins they encode. Instead it recited a list of generic uses to which the ESTs (like essentially any EST) could be put: serving as molecular markers, measuring messenger-RNA expression levels, identifying polymorphisms, isolating promoters, and similar laboratory operations. The examiner rejected the claims for lack of utility, the Board affirmed, and Fisher appealed.

## The framework: specific, substantial, and credible

The Federal Circuit grounded its analysis in the tripartite utility test that the Patent Office had codified in its 2001 Utility Examination Guidelines, drawing on *Brenner* and *In re Brana*. To satisfy § 101, an asserted utility must be **specific**, **substantial**, and **credible**. Credibility was not seriously in dispute; the litigation turned on the first two prongs, which the court defined with unusual care.

A **substantial** utility, in the Utility Guidelines language the court adopted, is one that defines a "real world" use; in the court's own words, an asserted use must show that the claimed invention "has a significant and presently available benefit to the public." A utility that is merely "hypothetical" is not substantial, and the Guidelines the court endorsed state that "[u]tilities that require or constitute carrying out further research to identify or reasonably confirm a 'real world' context of use are not substantial utilities." A **specific** utility is one particular to the subject matter claimed, as opposed to a general utility that would be true of any member of a broad class. The two prongs work together to exclude inventions that are valuable only as inputs to additional, undirected investigation.

## Applying the test: research intermediates are not enough

Measured against that framework, the court held, Fisher's seven asserted uses failed both prongs.

They failed the **substantial** prong because each use described the EST as a tool for *finding out* something: a step toward understanding the gene, not a present benefit flowing from it. Using an EST as a marker to monitor expression, or as a probe to fish out the corresponding gene, advances research; it does not deliver a real-world benefit until that downstream gene and its function are actually known. The court invoked *Brenner* directly: just as Manson's steroid was an object of further study rather than a thing of established use, Fisher's ESTs acted as "no more than research intermediates that may help scientists to isolate the particular underlying protein-encoding genes," their value depending on discoveries yet to be made.

They failed the **specific** prong because every asserted use was equally true of any EST whatsoever. Nothing in the recited utilities was particular to *these five* maize sequences; the same list could accompany an application for any randomly chosen gene fragment. A utility so generic that it attaches to an entire category of molecules is not "specific" to the claimed invention, and granting a patent on that basis would, the court warned, confer rights out of proportion to any genuine contribution.

Because the same deficiency that doomed the § 101 utility analysis also meant the specification taught no use a skilled artisan could practice, the court affirmed the parallel rejection under § 112's enablement requirement. That is a recurring feature of utility cases, where a missing use simultaneously undermines both doctrines.

## The Rader dissent: tools have utility

Judge Rader dissented sharply. ESTs, he argued, are genuinely useful as research tools, in the same way a microscope or a new reagent is useful. No one doubts that a better microscope is patentable even though its "use" is to enable further discovery. To Rader, the majority confused the utility of the *fragment* with the unknown utility of the *underlying gene*, holding ESTs to a standard that the law does not impose on other laboratory instruments. The disagreement frames the central policy question the case leaves behind: when is an enabling research input itself a patentable end, and when is it merely a license to keep hunting?

## Open questions

*Fisher* settled that *these* generic EST utilities were inadequate, but it did not draw a bright line for research tools generally. How much functional characterization of an underlying gene is "enough"? Would an EST tied to a gene of known function clear the bar? The opinion also leaves unresolved how its logic applies to other informational or platform inventions (antibodies of unverified target relevance, screening assays, diagnostic correlations) where value likewise lies partly downstream. And Rader's microscope analogy continues to press: the case marks a boundary it does not fully theorize, ensuring that "research tool" utility remains contested terrain.

## Implications

- **Disclose a downstream, specific use.** For gene fragments and similar molecules, recite a real-world benefit tied to the particular sequence claimed, not boilerplate true of any EST.
- **"Useful for research" is suspect.** A utility that amounts to enabling further investigation will often fail the substantial prong under *Fisher* and *Brenner*.
- **Generic utilities fail the specific prong.** If the asserted use would fit any member of a broad class, it is not specific to the claimed invention.
- **Utility and enablement travel together.** A specification that lacks a real use frequently fails both § 101 and § 112(a), giving challengers two routes to the same result.
- **A precedent with reach beyond genomics.** The specific/substantial framework now informs scrutiny of antibodies, diagnostics, and other claims whose payoff partly depends on future discovery.

## Frequently asked questions

**Did *In re Fisher* hold that ESTs can never be patented?**
No. The court held that *these* claims failed because their only disclosed utilities were generic and required further research to yield a benefit. An EST disclosed with a specific, substantial utility (for instance, tied to a gene of known function) could in principle satisfy § 101.

**How does *Fisher* relate to *Brenner v. Manson*?**
*Fisher* applies *Brenner*'s logic to genomics. Both cases reject patents on subject matter whose value lies in enabling further research rather than in a presently available benefit, and *Fisher* expressly relies on *Brenner*'s "hunting license" reasoning.

**Why did the claims also fail enablement under § 112?**
Because the specification did not teach a real use for the ESTs, it failed both to show utility under § 101 and to enable a skilled artisan to use the invention under § 112(a). The same gap defeats both requirements.

## Authorities and sources

- Federal Circuit opinion, *In re Fisher*, 421 F.3d 1365 (Fed. Cir. 2005) (full reporter text): https://www.law.berkeley.edu/archive/files/In_re_Fisher.pdf
- University of Houston Federal Circuit case summary, *In re Fisher*, No. 04-1465 (Sept. 7, 2005): https://www.law.uh.edu/faculty/gvetter3/fcplc/CaseSummaries/2005/In%20re%20Fisher,%2004-1465%20(Fed.%20Cir.%20Sept.%207,%202005)%20%5BSUMMARY%5D.pdf
- Patently-O pre-argument preview of the appeal (Donald Zuhn, May 2, 2005), published before the Federal Circuit heard the case: https://patentlyo.com/patent/2005/05/post-3.html
- Tashica T. Williams, "In re Fisher: Raising the Utility Hurdle for Express Sequence Tags," 21 Berkeley Tech. L.J. 123 (2006): https://btlj.org/data/articles2015/vol21/21_1_AR/21-berkeley-tech-l-j-0123-0146.pdf