Case, Board, and file wrapper citations in this article link to the documents in ProSearch. A free Design Patent Pro account is all that is needed to open them.
Introduction
An obviousness rejection of a design claim now requires a record-supported reason to combine the references. See LKQ v. GM, 102 F.4th 1280, 1299 (Fed. Cir. 2024) (en banc). This article examines what examiners are offering as that reason, how the Board has treated it, and what the record suggests would satisfy it. Under Rosen-Durling, the “so related” inquiry supplied the required connection between a primary and a secondary reference: the appearance of a feature in one had to suggest its application to the other. No reason for the combination independent of that relationship was required. LKQ rejected that formulation. The court held that “the motivation to combine these references need not come from the references themselves,” and that “there must be some record-supported reason (without hindsight) that an ordinary designer in the field of the article of manufacture would have modified the primary reference with the feature(s) from the secondary reference(s) to create the same overall appearance as the claimed design.” Id. at 1299.
The court named three sources for that reason and stated that the list was not exclusive. It did not say what evidence would establish any of them. It did not say how an examiner, who has no expert and no discovery, would supply one. This article addresses those questions from the examination record. It draws on 171 design patents issued between August 1, 2024 and September 20, 2026 whose file wrappers contain an obviousness rejection over a combination of references and LKQ‘s “record-supported” language. For scale, 1,942 design patents issued in the same period have an obviousness-type rejection of any kind in the wrapper. The article also draws on the seven Board decisions contained in those files, on the Board’s institution decisions, and on the district court decisions to date.
The article first traces the sources of the post-LKQ rationale, then examines examiner practice, applicant responses, the Board’s decisions, the reasons the rejections have not held, the litigation decisions, and what applicants and challengers might consider. A conclusion and a note on the limits of the record follow.
1. Where the Sources Came From
LKQ‘s three sources are the ordinarily skilled designer’s experience and creativity, the market demands and industry customs in the relevant field, and the ornamental features that are commonplace in the relevant field. 102 F.4th at 1299-1300. The court took them from the brief of the United States as amicus.
Each comes from KSR. MPEP § 2143.01, quoting Zup v. Nash Manufacturing, 896 F.3d 1365, 1371 (Fed. Cir. 2018), states KSR‘s sources as “market forces; design incentives; the ‘interrelated teachings of multiple patents’; ‘any need or problem known in the field of endeavor at the time of invention and addressed by the patent’; and the background knowledge, creativity, and common sense of the person of ordinary skill.” Market demand corresponds to KSR‘s market forces. Industry custom and commonplace features correspond to KSR‘s background knowledge. Experience and creativity corresponds to KSR‘s creativity and common sense. The KSR sources the court did not name are the ones that depend on a problem the combination solves. The court did not exclude them. It said that “in the area of motivation to combine, the problem to be solved may have less relevance in the design patent context than in the utility patent context.” Id. at 1299-1300. A documented design need, market pressure, or constraint remains available under the court’s non-exclusive list.
The court had a design precedent for the sources it kept. In Smith v. Whitman Saddle Co., 148 U.S. 674 (1893), the record showed several hundred styles of saddle and a custom among saddlers of varying them to suit the purchaser. The Court held that joining the halves of two known saddles was the exercise of the ordinary skill of the trade in the ordinary way. Id. at 680-81. The reason to combine came from evidence about the trade, not from the two saddles resembling each other.
The Office added to the court’s list in two steps. The May 22, 2024 memorandum names no sources. It quotes the record-supported-reason sentence and refers examiners to “MPEP 2141 et seq. for the guidelines to follow in utility applications.” The July 2024 examiner training deck names five: “the ordinarily skilled designer’s experience, creativity, and common sense; market demands and industry customs that exist in the relevant field; ornamental features that are commonplace in the relevant field; the extent to which ornamental features are motivated by functional considerations; and whether industry designers face similar design problems as designers in other industries or otherwise look to other industries for design ideas,” citing MPEP § 2143.01. The fourth and fifth are the Office’s additions. The deck then supplies two sample rejections. The first states that the modification “is no more than a simple substitution of one known design element for another” and “would have been within the skill of an ordinarily skilled designer.” The second states that the secondary reference “demonstrates that the ornamental feature … is commonplace in the field … and would therefore have been an obvious design choice.”
Examiners added two more. The sentence that a designer “is charged with knowledge of the related art; therefore, the combination of old elements, herein, would have been well within the level of ordinary skill,” citing In re Antle and In re Nalbandian, appears in neither the memo nor the deck. It was carried over from the pre-LKQ form paragraph. The Glavas “so related” sentence also continued to appear in post-LKQ actions, although the deck’s assessment exercise marks it as improper.
The sources in use at the Office therefore come from four places: the court’s three, the deck’s two additions, the deck’s two sample sentences, and the two carryovers.
Figure 1. The four places the reasons to combine in the examination record came from.
2. What Examiners Are Writing, and How Each Formulation Has Fared
A file often contains more than one formulation. The counts are files in which the formulation appears in the last rejection. The table summarizes the formulations. The text that follows quotes representative files.
Simple substitution (79 files). The sentence has been before the Board once. In Ex parte Dascola, Appeal No. 2025-003515, Application No. 29/785,114 (P.T.A.B. Feb. 10, 2026), the examiner had replaced the “so related” sentence with the simple-substitution sentence in the Examiner’s Answer, with a footnote that the change was not a new ground of rejection. The Board reversed because the examiner did not provide “any findings that a designer of ordinary skill in the art would be inclined to further modify an oblong bottom element into a pill/capsule shape, much less adequate reasoning as to why a designer would do so.” No examiner in the 171 files explained the sentence beyond restating it. Examiners who defended it treated it as proof of itself. Application No. 29/898,552: “the demonstration of known alternative design approaches shown in the identified reference designs constitutes the ‘record-supported’ reason for the modifications proposed by the rejection.” Application No. 29/837,995: “there is proper evidence and motivation to combine the prior arts. Bin demonstrates that changing the angle of return and the length of the concave surface is commonplace in the field of designing litter box openings.” One examiner conceded the sentence where the primary reference had nothing in the claimed location to substitute. Application No. 29/884,865: “Examiner lacked proper motivation to combine the primary and secondary references.”
Commonplace features and design choice (66 files). This rationale is on LKQ‘s list. It is also the rationale examiners can support with the tools they have, because it is proved with references. In 42 of the 66 files the only support was the applied references themselves. In 24 files the examiner cited additional references, not applied in the combination, to show the feature was common. In Application No. 29/886,662 the examiner listed six United States design patents, two Chinese documents and a product listing to show that a two-stepped mug profile “is a common and well-established feature in the art of mugs.” In Application No. 29/868,173: “The ornamental feature of smaller sized circles can be seen in D946594, D910057, D806735, and D603418.” The Board has addressed that practice twice. Part 5 sets out what it held.
Figure 2. Support offered for the 66 commonplace or design-choice rejections.
Within the skill of the ordinary designer (58 files). This appears usually with the Antle/Nalbandian sentence, or with In re Stevens on changes in proportion (24 files). The Office’s own manual rejects it as a reason. MPEP § 2143.01(IV): a statement that modifications “would have been ‘well within the ordinary skill of the art'” because the references “teach that all aspects of the claimed invention were individually known in the art is not sufficient to establish a prima facie case of obviousness without some objective reason to combine the teachings of the references,” citing Ex parte Levengood, 28 USPQ2d 1300 (B.P.A.I. 1993). The sentence states the level of skill, which is Graham‘s third inquiry. It does not state a reason to combine.
Similarity between the references (28 files). In most of these files the rejection predates LKQ. In a few the sentence was typed into post-LKQ actions as the reason. In Application No. 29/972,805 a June 2025 final rejection answered a no-motivation argument with In re Glavas and with MRC Innovations v. Hunter, 747 F.3d 1326, 1334-36 (Fed. Cir. 2014), for the proposition that “it is the mere similarity in appearance that itself provides the suggestion that one should apply certain features to another design.” MRC applied the “so related” framework that LKQ abrogated. LKQ did not overrule MRC by name. Its holding rejects the use of that relationship as a substitute for a record-supported reason. The Board has reversed the “so related” rationale four times since LKQ (Part 5). The district court has treated it differently (Part 7).
No reason stated (14 files). The modification is asserted and no “because” follows.
A reason grounded in the field (13 files). The examiner states why designers of the claimed article would look to the secondary reference. Application No. 29/816,824 (lighted mirror, outside the corpus): a designer of lighted mirrors “would look to the lighting fixture industry for design ideas.” Application No. 29/838,006 (duffel bag): “A designer looking to modify a bag would reasonably look to any other bag, specifically ones adapted for similar purposes, for modifications.” Application No. 29/924,967 (transitional interface): “one of ordinary skill in the art of graphic design and mobile device operating systems would know to look to other examples of mobile phone unlock screen transitions.” In several files the field was defined by the claimed features. Application No. 29/810,478 refers to “the field of endeavor of the design of graphical user interfaces including overlapping circles.” None of the 13 was tested on the motivation question. Each ended in an amendment, an allowance on a visual difference, or a Board reversal on another ground.
Market demand and industry custom (8 files). No file in the corpus identifies evidence beyond the applied references for the asserted demand or custom. Four bra applications (29/952,839, 29/953,335, 29/953,591, 29/953,592) state: “it is common in the art to modify bras to have different strap structures. Therefore, it is not an unreasonable conclusion that one strap system can be applied to a different bra. Straps are not exclusive to individual bras but are known to be applied to a large variety of bras.” Two drainage endcap applications (29/955,323, 29/962,478) state that “consumer preference favors familiar product appearances” and that “manufacturers routinely modify existing products to match prevailing commercial designs,” supported by “the various cited references.” One examiner outside the corpus supported an industry-custom finding with a source. In Application No. 29/977,938 the examiner cited a 1979 National Depression Glass Association page “demonstrating that the stem and decoration (or pattern) of a drinking glass are equally interchangeable features in the art.” That examiner maintained the reason through allowance and withdrew because the prior art did not show the claimed sharp ridge between the triangles.
Functional considerations (3 files). The deck’s addition appears as a named reason in three files and is explained in none. The two endcap files state that the modifications “are driven by market customs, functional considerations and simple substitution of one known element for another in the field of art.” In Application No. 29/824,050 the examiner stated that seams extend down the side of the plank “because this is a natural result of joining three dimensional planks together.”
Other KSR rationales (10 files). “Combination of prior art elements according to known methods to yield predictable results” appears in Application No. 29/877,958 (seat) and a few others. “Known design option” appears in nine files. “Obvious to try” appears in one file outside the corpus, Application No. 29/951,373.
Aesthetic appeal (3 files). In Application No. 29/898,552 the examiner answered an objection that the reason was not record-supported with the proposition that “the recognized purpose of ornamental design is to add aesthetic appeal to the article of manufacture,” citing MPEP § 1504.04 and Ex parte Cady. In Application No. 29/776,397 the reason was that “visual simplification of a design is generally regarded as advantageous among industrial designers working in the space of personal electronic devices.” A reason that applies to every modification of every design does not explain the choice of one modification over another.
3. What Applicants Are Arguing
Twenty applicants made the selection argument in terms: that the existence of an element in the prior art does not explain why a designer would have chosen it, or this version of it, over the alternatives. The authorities cited were Next Step Group v. Deckers Outdoor Corp., IPR2024-00525, Paper 16 (P.T.A.B. Aug. 6, 2024), cited in 19 files. A&A Global Imports v. Lerman Container Corp., IPR2024-01138, Paper 7 (P.T.A.B. Jan. 22, 2025), for the “as opposed to all other possible configurations” language, cited in 14 files. Ex parte Chen, Appeal No. 2024-002635 (P.T.A.B. Aug. 12, 2025), described by applicants as holding that “so related” and “could be adapted” are not record-supported reasons, cited in 4 files. And Gema USA v. First in Finishing (S.D. Ind. Mar. 27, 2025), for the point that using the patent itself to supply the motivation is hindsight, cited in 2 files.
The fullest form is in Application No. 29/877,958, applied feature by feature: no reason why the designer would select the secondary reference’s side frames “from all prior art seat side frame members, such as simply leaving the side frame members of [the primary reference] for example or using the side frame members from [the other secondary reference],” and the same for the legs, the cushions, and the backrest. Other examples: Application No. 29/735,750 (firearm sight): the Office “has provided no explanation for why a designer of ordinary skill in the art would have chosen to use the horseshoe design element of the Glock rear iron sight as opposed to all the other ‘infilled horseshoe elements.'” Application No. 29/908,486 (exercise interface): the Office “fails … to cite any evidence that would suggest why such a designer would have been motivated to select and combine those two references, other than the fact that they may have separately disclosed individual elements.”
Twenty-two applicants argued that a reference was not analogous art. One applicant reserved an Official Notice objection under MPEP § 2144.03, which requires an examiner who asserts that a feature is common to support the assertion on the record when challenged.
4. How Examiners Answer
In 106 of the 171 files, the Office allowed the application after the applicant presented arguments and without a claim or drawing amendment identified as the basis of allowance. In 53 the examiner identified a feature or proportion the combination did not produce, or a difference in overall appearance. In 41 the examiner stated no reason. In 7 the primary reference was found not visually similar. In 1 the ground was analogous art (Application No. 29/787,268). In 4 the examiner conceded the reason to combine:
Application No. 29/884,865, April 11, 2025: the arguments “regarding the applicability of the ‘simple substitution’ rationale was found persuasive due to the lack of an element in-existence to the left of the dollar sign symbol within the primary reference. Examiner lacked proper motivation to combine the primary and secondary references.”
Application No. 29/841,323, May 9, 2025, after an appeal conference: the references did not “provide sufficient evidence for motivation to combine the references.”
Application No. 29/895,014: “the Examiner has found that it would not necessarily have been obvious to a designer of ordinary skill in the art to modify the primary reference using qualities or features shown in the supporting references.”
Application No. 29/918,972: the rejection “has been overcome by the applicant’s argument submitted on 12/31/2025 regarding the rationale for an ordinary designer to modify the reference to create the claimed design.”
The remaining 65 ended otherwise. Forty-three ended by drawing or claim amendment, usually broken lines converted to solid, often by continued prosecution application after a final rejection. Seven ended by Board reversal, 4 by declaration, and 3 by disqualifying a reference’s publication date. In 8 the wrapper does not state how the rejection was resolved.
Figure 3. How the last obviousness rejection ended in each of the 171 files.
In 54 files the examiner defended the reason after an LKQ argument and maintained the rejection at least once. Two allowances state the examiner’s position on the reason while withdrawing. In Application No. 29/797,092 the examiner rejected the hindsight and primary-reference arguments and withdrew because “The examiner did not find this tapered edge in the prior art.” In Application No. 29/917,871, after a declaration on overall visual impression: “Although the examiner now recognizes that the proposed modification would not have been obvious, this does not indicate that the original rejection lacked a justified, record supported reason. The examiner maintains that the initial rejection was proper, even though the conclusion of obviousness is no longer sustained.”
No examiner adopted the selection argument in terms. In the Board decisions reviewed that reached it, the Board did. Of the 20 files where it was made, 9 were allowed with no reason stated, 4 on a visual difference, 4 after drawing amendment, 1 on analogous art, 1 on the concession in 29/918,972, and 1 by the Board. Examiners accepted the analogous art argument as the ground of withdrawal once in 22 files, and rejected it in terms in at least four while allowing on other grounds. Application No. 29/937,714: “while examiner disagrees as to what is considered analogous art, the remainder of their argument has been found persuasive.”
5. What the Board Has Held
Seven of the 171 files contain a Board decision on an obviousness rejection after LKQ. In each of the seven the Board reversed, and in none did it sustain the examiner’s stated reason to combine. The grounds were the absence of a record-supported reason or the failure of the combination to produce the claimed appearance.
Ex parte Thornburg, Appeal No. 2023-002969, Application No. 29/570,477 (air filter), June 21, 2024: “It may be that a sustainable obviousness rejection of Appellant’s claim over Crabtree could be articulated, applying LKQ, but such a rejection is not present in the record before us.”
Appeal No. 2023-001416, Application No. 29/724,186 (icon), August 30, 2024. The rejection used “so related.” In the Answer the examiner cited “multiple newly cited (but not explicitly applied) prior art references” showing the feature and called the change “a routine variation to a skilled designer.” The Board held that “such reasoning does not sufficiently address the above-noted deficiency as to whether a designer of ordinary skill would have been motivated to modify the design of Jasmadi in view of Wells-Papanek in the first instance.”
Appeal No. 2023-001595, Application No. 29/678,877 (interface), September 20, 2024: “we are not provided with any such record-supported reasoning for such a modification.”
Ex parte Argo, Appeal No. 2023-002547, Application No. 29/732,483 (interface), December 11, 2024: reversed because the overall appearance “differs substantially from the end result of the Examiner’s consideration” of the primary reference.
Ex parte Caro, Appeal No. 2024-002550, Application No. 29/841,229 (interface), August 7, 2025. The stated reason was that the references “are so related that the appearance of features shown in one would suggest the application of those features to the other.” In the Examiner’s Answer the examiner defended it with MRC Innovations. The Board held that the modification was “simply supported by the statement that this would have been done because Tarik and Sato each discloses certain design features, and that Tarik and Sato ‘are so related,'” that “The Examiner does not explain adequately why an ordinary designer would have been motivated (without hindsight) to modify Designer,” and that “Although there may be a reason or reasons why a designer of ordinary skill would have modified Designer in the manner proposed in the rejection in view of the designs disclosed in Tarik and Sato, the record does not appear to provide such a reason for us to consider under LKQ.”
Ex parte Jafarzadeh, Appeal No. 2025-002153, Application No. 29/671,792 (delivery robot), January 29, 2026. The examiner called the change “an obvious change in arrangement to a designer of ordinary skill in the art of robots” and cited disclosures about moving digital images closer together on a screen. The Board: “We, like the Appellant, are confused as to how this relates to the case at hand.”
Ex parte Dascola, February 10, 2026, is described in Part 2.
Figure 4. The seven Board decisions in the corpus, June 2024 to February 2026.
Whether these holdings are consistent with LKQ and KSR
The Board has not rejected any source on the court’s list. It has rejected the absence of a reason, and in each case on a ground that KSR or the Office’s manual already states. A recitation that the references are related is a statement about the prior art, not a finding about the designer (Caro, Chen, Argo, Appeal No. 2023-001595, and MPEP § 2143.01(IV)). Evidence that a feature is common, or that software can move it, shows that the change could be made, not that it would be made (Appeal No. 2023-001416, Jafarzadeh, MPEP § 2143.01(III), and Chen on “could be adapted”). Availability of a feature does not explain the claimed degree of it (Dascola, Next Step, A&A Global). A modification “as in the claim” is hindsight (Thornburg, and In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006)). Each holding applies KSR‘s requirement that “there must be some articulated reasoning with some rational underpinning to support the legal conclusion of obviousness.” 550 U.S. at 418, quoting Kahn.
Two further observations. Several of the decisions reviewed involve overlapping Board membership. Judges Song, Hoskins and Murphy decided Caro, Chen, Argo, Appeal No. 2023-001416 and Appeal No. 2023-001595. Judge Murphy also wrote Jafarzadeh and sat on Dascola, which was decided by Judges Fischetti, Staicovici and Murphy. The decisions reviewed are therefore not a cross-section of Board panels. And the Board has not had before it a rejection that offered one of the court’s sources with evidence behind it. It has therefore had no occasion to accept or reject any source.
6. Why the Rejections Have Not Held as Findings, What the Files Show About Amendment, and Whether the Court Chose the Right Sources
The record separates two results.
As findings, the rejections have not held. In the 171 files, no reviewing tribunal sustained an examiner’s stated reason. Seven files went to the Board, and the Board reversed each of the seven. Where the applicant argued and the examiner withdrew, the examiner identified a visual difference in 53 files, gave no reason in 41, and conceded the motivation in 4. Examiners answer a motivation argument by finding a visual difference. A visual difference can be found in the drawings. A reason to combine that was not in the rejection can be supplied later only by reopening prosecution or by a new ground of rejection in the Examiner’s Answer, 37 C.F.R. § 41.39, and the files show that reasons supplied after argument have also failed on the merits, as in Appeal No. 2023-001416.
The rejections coincided with claim-narrowing amendments in a quarter of the files. In 43 files the file ended with a drawing or claim amendment rather than an appeal or an express examiner statement on motivation. The wrappers show the sequence. They do not show why the applicant chose to amend. Each amendment narrows the claim. Under Top Brand LLC v. Cozy Comfort Co., 143 F.4th 1349 (Fed. Cir. 2025), an amendment or argument that clearly distinguishes the prior art may limit the enforceable scope of the design patent.
Four causes appear in the files.
First, LKQ‘s sources are facts about the field. Market demand and industry custom are proved with catalogs, trade publications, dated product histories, and testimony from designers. The files reviewed rarely contain evidence of that kind, and ex parte examination gives an examiner limited means of developing it when the reason depends on sources outside the applied references. The deck names the sources and does not say how an examiner would establish them. In the files, examiners prove what the references show, which is that a feature exists, and add a sentence that asserts the rest.
Second, the deck’s sample sentences state categories, not reasons. The utility version of the simple-substitution rationale, MPEP § 2143(I)(B), requires articulated findings, including “a finding that the substituted components and their functions were known in the art” and a finding that “the results of the substitution would have been predictable,” and states: “If any of these findings cannot be made, then this rationale cannot be used.” The design version keeps the label and omits the findings. The omitted findings are the ones with no counterpart in a change of appearance.
Third, as an analytical matter, availability is not selection. Each of the Office’s rationales proves that the borrowed feature was available. None explains why the designer would have chosen it, or this version of it, over the alternatives the same field contains. The Board identified that gap in Next Step, A&A Global and Dascola. In a field defined by appearance, nearly every shape exists somewhere, so availability proves little. The claimed design answers the selection question for the examiner unless the examiner finds another answer. This article uses “selection” as a shorthand for the gap between evidence that a feature was available in the prior art and evidence that an ordinary designer would have chosen that feature, in that form, for the proposed modification. The term is analytical. No court has adopted it as a doctrinal label.
Fourth, the rationale is assembled with the claim in view. The examiner begins with the claimed design, selects a primary reference that resembles it, selects a secondary feature that resembles it, and states that the two could be joined. The direction of the modification comes from the claim. Thornburg states the rule: a modification “supported only by the statement that this would have been done in order to achieve ‘the claim'” is improper hindsight.
Were these the right sources?
The court’s three are facts about the field, they are proved with evidence that predates the claimed design, and Whitman Saddle shows one of them deciding a case. The record does not show a defect in the list. It shows that the Office adopted the list without a method of proof and substituted labels for proof.
What might be removed is clear from the authorities. The “so related” sentence is the abrogated test, and the deck says so. The “within the skill” sentence and the Antle/Nalbandian sentence are barred by MPEP § 2143.01(IV). The bare simple-substitution sentence does not satisfy MPEP § 2143(I)(B). Aesthetic appeal does not select among modifications. Analogousness offered as the reason merges the scope of the prior art with the motivation, which LKQ treated as separate inquiries. “Obvious to try” requires “a finite number of identified, predictable solutions,” KSR, 550 U.S. at 421, and the ways to vary an appearance are not finite.
What might be added is a method rather than a new source. The 24 files in which examiners cited background references show the corps using the right evidence for the wrong proposition. The references were cited to show that the feature exists. Whitman Saddle used the same kind of evidence to show that the trade customarily varied the article in that way. The difference is between “circles of this size appear in the art” and “designers of this article vary this element in this way as a matter of practice, and here is the record of them doing so.” The second statement concerns the designer’s conduct and does not depend on the claim. The first concerns the prior art and becomes a reason only when the claim supplies the direction. That distinction is analysis. No tribunal has drawn it.
Two further sources are available to examiners without new tools. The primary reference’s own variants can supply a reason internal to the record. In In re Borden, 90 F.3d 1570, 1575-76 (Fed. Cir. 1996), the brochure for the primary reference did more than show alternatives. It presented custom variations, including changes to the necks and chamber sizes, which were the kinds of changes needed to reach the claimed design. The source itself offered those variations as options for the same article. In Application No. 29/856,912 the examiner wrote that “[t]he primary reference itself … shows further examples of a notification bar and time/date stamp being arranged higher or lower within the graphical user interface.” A documented constraint, such as a mating part, a standard, or a fixed component the design must accommodate, can supply a reason that narrows the available forms. That is the functional source in the form Kong would accept, provided the constraint reaches the specific claimed appearance.
7. Litigation
Litigation supplies what examination lacks: experts, discovery, declarations, and documents from the field. The decisions so far show challengers making the same showing the examiners make.
In Diode Dynamics, L.L.C. v. 5DLight, Inc., 2025 WL 2827023 (C.D. Cal. Sept. 15, 2025), the court granted the patentee summary judgment. It found the combination “tailored to replicate the D’648 design itself, not to solve a known design problem or respond to a real-world aesthetic trend,” and found that the expert cited no “functional constraint, market pressure, or trend.” By treating the absence of a constraint as a deficiency, the court treated a constraint as an available source.
In The Kong Co. v. Bounce Enterprises, LLC (D. Colo. Mar. 27, 2025), the functional evidence was that both inventors designed the channels for airflow. The court denied summary judgment of obviousness. Airflow motivated the channels in both designs, and it did not explain the curved corners or channel width that separated them.
In Gema USA, Inc. v. First in Finishing, Inc. (S.D. Ind. Mar. 27, 2025), the stated reason for combining was that the combination would create the overall impression of the patents in suit. The court granted the patentee summary judgment, holding that using the patents to supply the motivation is hindsight.
In Dynamite Marketing, Inc. v. The WowLine, Inc. (Fed. Cir. Sept. 12, 2025) (nonprecedential), the challenger never put a primary reference in evidence.
In the three Board institution denials (Next Step, A&A Global, and Arashi Vision v. GoPro, IPR2024-01434), the petitioners asserted that features were common or minimal and did not explain selection or show that the combination produced the claimed appearance.
The one source that has survived a dispositive motion is the one LKQ did not name. In LKQ Corp. v. General Motors Co., No. 1:21-cv-05854, ECF No. 353 (N.D. Ill. June 30, 2025), the court denied the patentee’s motion for summary judgment of no obviousness. The expert had identified primary and secondary references, all front fenders, that in his opinion together produced the claimed appearance. The court quoted Sealy and MRC Innovations for the proposition that “the mere similarity in appearance” between references “itself provides the suggestion that one should apply certain features to another design,” held in a footnote that those cases remain good law because LKQ “did not purport to make the standard for demonstrating motivation more stringent,” and noted that the patentee “has provided no argument that the primary and secondary references, all of which are front vehicle fenders, are not so related to themselves establish a motivation to combine.”
That ruling and Caro can both stand. The district court denied summary judgment on the record before it. The ruling is interlocutory, and it does not establish that similarity between references alone satisfies LKQ. The court decided that a jury could find motivation on an expert’s similarity showing that the patentee had not contested. The Board decided that an examiner’s recitation of relatedness, with no findings, does not make a prima facie case. Neither tribunal has decided whether similarity between references, explained on the record, satisfies LKQ. The Federal Circuit has not addressed it.
The litigation source with the most support in the authorities is the one raised and reserved in the fender case: background art offered to prove what the design community knew and demanded, which is the Whitman Saddle method with a modern record. It has one limit. The last sentence of § 103 provides that patentability shall not be negated by the manner in which the invention was made. The patentee’s own market research is often the best evidence of demand and is also a record of the inventor’s path. The fender court assumed without deciding that the patentee’s clinic evidence could not be used and reserved the question for a motion in limine.
The merits decisions to watch are the trial in the fender case and Top Glory Trading Group v. Cole Haan LLC, IPR2025-01395, instituted February 24, 2026, with a final written decision due by February 2027.
8. What Practitioners Might Consider
For an applicant responding to a rejection, the record suggests the following.
Locate the “because” clause and classify it. Each category has a matching answer. The “so related” sentence meets LKQ and the four Board reversals. The “within the skill” sentence and the Antle/Nalbandian sentence meet MPEP § 2143.01(IV) and Levengood. A bare simple-substitution sentence meets MPEP § 2143(I)(B) and Dascola. A “commonplace” finding meets the selection point of Next Step and A&A Global, and where the examiner asserts prevalence without evidence, a request under MPEP § 2144.03 that the assertion be supported on the record. A market-custom statement without a source meets the same request. A functional reason meets Kong‘s two questions: whether the function distinguishes the designs or is shared by them, and whether it reaches the appearance or only the presence of the feature.
Pair the motivation argument with an overall-appearance argument. The overall-appearance argument is the one examiners act on: 53 withdrawals rested on a visual difference and 4 on motivation. The pairing has a cost under Top Brand, since a feature clearly identified as distinguishing the prior art may be discounted in a later infringement comparison. An overlay of the whole combination on the claimed design argues the impression without a list. Where a list is unavoidable, the fewest features that carry the impression, chosen with the likely accused products in mind, limit the disclaimer.
Treat the selection argument as an appeal argument. It was made in 20 files. No examiner adopted it. In the Board decisions reviewed that reached it, the Board adopted it. The seven post-LKQ obviousness appeals in the corpus were each reversed, and in Application No. 29/841,323 the pre-appeal conference conceded motivation before the brief was filed. An applicant weighing a continued prosecution application with amended drawings against an appeal might weigh the estoppel cost of the amendment against the delay of the appeal on that record.
Request an interview before the response. In the corpus the examiner’s reason was often stated for the first time in the answer to the applicant’s arguments.
Preserve the analogous art argument. Examiners accepted it as the ground of withdrawal in one of 22 files. It remains available on appeal, and it pairs with the field-definition point where the examiner has defined the field by the claimed features.
For a challenger in a petition or in court, the record suggests the reverse of each examiner error. A primary reference in evidence. Each Graham inquiry addressed, including who the ordinary designer is. A reason stated in the words of the field and provable without the claimed design in view, built from dated evidence of what designers of the article did, from the primary reference’s own variants, or from a documented constraint. An explanation of why that reason leads to the specific configuration claimed rather than to the alternatives. A demonstration that the modification produces the claimed overall appearance, with the remaining differences accounted for. And, where market demand is the reason, evidence from outside the patentee’s own files. An expert who states that the references could be combined has stated what the examiners state. An expert who documents what the trade did has supplied what Whitman Saddle required.
Conclusion
Under Rosen-Durling the relationship between the references served as the reason to combine. LKQ eliminated that rule and required a record-supported reason, and it named three sources that are facts about the field. The Office adopted the requirement and supplied examiners with sentences drawn from utility rationales, without a method for proving the facts the court named. Examiners prove that a feature exists and assert the rest. In the seven appeal decisions in the files reviewed, the Board, applying the rule stated in KSR and in the Office’s own manual, reversed each rejection for want of a finding about why the designer would have made the modification. A quarter of the files ended in a claim-narrowing amendment. In litigation the tools for proving facts about the field exist, and the challengers whose cases have been decided did not use them for that purpose. Among the authorities reviewed here, none is a merits decision sustaining a design patent combination on a developed LKQ record of field practice, documented design options, market evidence, or a constraint that directs the particular modification. That absence does not establish that such a record cannot be made. Whitman Saddle shows one. The decisions that may first address such a record are the fender trial in the Northern District of Illinois and Top Glory at the Board. On the authorities reviewed, a sustained reason would resemble Whitman Saddle: a documented practice in the trade of varying the article in the claimed way, stated without reference to the claimed design, and tied to the specific configuration claimed.
Limits of the Record
The 171 files are issued patents. Applications abandoned after the rejection are not in the set, so it shows how rejections were overcome, not how often they held. The search term captures files that quote LKQ‘s “record-supported” phrase. Files that answered a rejection without quoting it are outside the set, and a few files known to cite LKQ on obviousness were missed, apparently because the phrase broke across a line in the PDF. Files cited above as outside the corpus are marked as such. Quotations are from the text layer of the wrappers. The distinction in Part 6 between evidence that a feature exists and evidence of a trade practice is analysis, not a holding of any tribunal. Design Patent Pro subscribers can open each file wrapper from the application number.
