# LKQ v. GM: The Federal Circuit Dismantles the Design-Patent Obviousness Test

> Sitting en banc, the Federal Circuit overruled the four-decade-old Rosen-Durling framework and folded design-patent obviousness into the flexible Graham analysis used for utility patents. Design patents just became easier to challenge.

Topic: Patents  |  Author: Lidiia Levitska  |  Source: Intellectual Property Law (outsideipcounsel.com)
Canonical: https://outsideipcounsel.com/blog/lkq-v-gm-design-patent-obviousness/


*LKQ Corp. v. GM Global Technology Operations LLC*, 102 F.4th 1280 (Fed. Cir. 2024), decided en banc on May 21, 2024, is the most significant design-patent decision in more than forty years. In an opinion by Judge Stoll, the en banc Federal Circuit overruled the two threshold requirements of the long-governing *Rosen-Durling* test for assessing whether a design patent is obvious, concluding they were "improperly rigid" and imposed "limitations absent from § 103's broad and flexible standard." The statutory scheme and Supreme Court precedent, the court reasoned, including *Whitman Saddle*, *Graham*, and *KSR International Co. v. Teleflex Inc.*, all suggest a more flexible approach than *Rosen-Durling*. The decision recalibrates both prosecution and litigation strategy for anyone who relies on design protection.

## At a glance

- **Case:** *LKQ Corp. v. GM Global Technology Operations LLC*, 102 F.4th 1280 (Fed. Cir. 2024) (en banc)
- **Decided:** May 21, 2024; opinion by Judge Stoll
- **Holding:** The *Rosen-Durling* test's threshold requirements ("basically the same" primary reference; "so related" secondary references) are overruled as improperly rigid; design-patent obviousness is governed by the flexible *Graham v. John Deere* framework used for utility patents, though a primary reference must still be identified
- **Status:** Affirmed-in-part, vacated-in-part, and remanded (No. 2021-2348). The court reinstated the panel's holding that substantial evidence supported the Board's finding of no anticipation, vacated the Board's nonobviousness determination, and remanded for the PTAB to apply the new standard

## What Rosen-Durling required

Design patents protect the ornamental appearance of an article of manufacture, and like utility patents they are subject to the nonobviousness requirement of § 103. For decades, the Federal Circuit assessed design-patent obviousness through a rigid two-step gate derived from *In re Rosen* and *Durling v. Spectrum Furniture*. First, a challenger had to identify a single primary reference: a prior design "basically the same" as the claimed design as a whole. Only if such a reference existed could the analysis proceed. Second, any secondary references used to modify the primary reference had to be "so related" to it that features in one would suggest application to the other.

Those threshold requirements were demanding, and they frequently defeated obviousness challenges at the outset: if no single prior design was "basically the same" as the claim, the inquiry ended before it began.

## Why the court overruled it

The en banc court held that *Rosen-Durling*'s rigid prerequisites could not be squared with the statute or with Supreme Court precedent. Section 103 applies to design and utility patents alike, and *KSR* had rejected, for utility patents, precisely the kind of rigid, formalistic gatekeeping that *Rosen-Durling* imposed on designs. The Supreme Court in *KSR* condemned tests that constrain the obviousness analysis with mandatory thresholds rather than allowing a flexible, evidence-based inquiry. Because nothing in § 103 justifies a separate, more rigid regime for designs, the court held, design-patent obviousness must be analyzed under the same framework as everything else.

That framework is the four-part inquiry from *Graham v. John Deere Co.*: (1) the scope and content of the prior art; (2) the differences between the prior art and the claim at issue; (3) the level of ordinary skill in the pertinent art; and (4) secondary considerations of nonobviousness, such as commercial success, copying, and industry praise. The rigid "basically the same" and "so related" thresholds are gone. The primary reference itself, however, survives: the court held that in applying Graham factor one, "a primary reference must be identified," reasoning that doing so protects against hindsight. What changed is the standard it must meet. The primary reference need only be "something in existence," and not, in language the court borrowed from the C.C.P.A.'s 1950 decision in *In re Jennings*, something "brought into existence by selecting individual features from prior art and combining them." It "will likely be the closest prior art, i.e., the prior art design that is most visually similar to the claimed design." It must be analogous art, and so must any secondary reference, but the two no longer need to be "so related" to each other.

## The consequence: a lower bar to invalidation

The practical effect is to make design patents easier to challenge as obvious. By removing the threshold that often ended challenges prematurely, *LKQ* allows challengers to assemble and combine prior-art designs under the flexible *Graham* approach, much as they do against utility patents. The case itself arose from an inter partes review concerning GM's design patent for a vehicle front fender. That context signals the stakes for the automotive-parts and repair industries, where design patents have been a central tool, but the ruling reaches every field that relies on design protection, from consumer products to fashion.

## Open questions

The en banc court deliberately left the contours of the new analysis to be worked out case by case. Several questions are now live. How visually close must a primary reference be in practice, now that "basically the same" is gone but the reference itself is still required? How analogous must prior-art designs be to be combinable? On that second question the court expressly declined to draw the line, saying only that art in the same field of endeavor qualifies, that it did not foreclose other art, and that it would "leave it to future cases to further develop the application of this standard." How will the level of ordinary skill be defined for an *ornamental* design, where "skill" is aesthetic rather than technical? And how much weight will secondary considerations carry in the design context? The PTAB and the Federal Circuit will answer these on remand and in the cases that follow.

## Implications for design-patent owners and challengers

- **Owners: build and preserve secondary-considerations evidence.** With the rigid threshold gone, evidence of commercial success, copying, and industry praise becomes a more important bulwark against obviousness.
- **Owners: expect more robust challenges.** Portfolios that looked secure under *Rosen-Durling* may be more vulnerable; reassess litigation and licensing posture accordingly.
- **Challengers: think like a utility-patent litigant.** Marshal combinations of prior-art designs under *Graham*/*KSR* rather than searching for a single near-identical reference.

## Frequently asked questions

**What was the Rosen-Durling test?** A rigid two-step framework that required a single prior-art design "basically the same" as the claimed design before any obviousness analysis could proceed, and that limited which secondary references could be combined with it.

**What replaces it?** The same flexible four-factor *Graham v. John Deere* analysis used for utility-patent obviousness, consistent with the Supreme Court's *KSR* decision.

**Does this make design patents weaker?** It generally makes them easier to challenge as obvious by removing a threshold that often ended challenges early. Owners can still defend validity, including through secondary considerations of nonobviousness.

## Authorities and sources

- *LKQ Corp. v. GM Global Technology Operations LLC*, 102 F.4th 1280 (Fed. Cir. 2024) (en banc), No. 2021-2348, decided May 21, 2024: [slip opinion (PDF), U.S. Court of Appeals for the Federal Circuit](https://www.cafc.uscourts.gov/opinions-orders/21-2348.OPINION.5-21-2024_2321050.pdf). Opinion by Judge Stoll; Judge Lourie filed an opinion concurring in the judgment, agreeing with the remand but arguing the court should have modified rather than discarded *Rosen* and *Durling*.
- *Graham v. John Deere Co. of Kansas City*, 383 U.S. 1 (1966): [Cornell LII](https://www.law.cornell.edu/supremecourt/text/383/1).
- *KSR International Co. v. Teleflex Inc.*, 550 U.S. 398 (2007): [official U.S. Reports text (PDF), Library of Congress](https://tile.loc.gov/storage-services/service/ll/usrep/usrep550/usrep550398/usrep550398.pdf).
- *Smith v. Whitman Saddle Co.*, 148 U.S. 674 (1893), the design-patent decision the en banc court read as applying a more flexible approach: [official U.S. Reports text (PDF), Library of Congress](https://tile.loc.gov/storage-services/service/ll/usrep/usrep148/usrep148674/usrep148674.pdf).
- 35 U.S.C. § 103 (conditions for patentability; non-obvious subject matter): [Cornell LII](https://www.law.cornell.edu/uscode/text/35/103).
- 35 U.S.C. § 171 (patents for designs, applying the conditions of Title 35 to designs): [Cornell LII](https://www.law.cornell.edu/uscode/text/35/171).