VERINATA HEALTH, INC., Illumina, Inc., Plaintiffs-Appellants v. ARIOSA DIAGNOSTICS, INC, Roche Molecular Systems, Inc., Defendants-Cross-Appellants
After trial on the merits, a jury found two U.S. patents valid and infringed. Ariosa Diagnostics, Inc., and Roche Molecular Systems, Inc., moved for judgment as a matter of law on invalidity and noninfringement. Verinata Health, Inc., and Illumina, Inc., moved for a permanent injunction, supplemental damages, an accounting, and pre-and post-judgment interest. The district court denied the parties’ motions. Verinata and Illumina appeal the denial of the permanent injunction, supplemental damages, an accounting, and pre-judgment interest. Ariosa and Roche cross-appeal the denial of judgment as a matter of law on invalidity and noninfringement. We conclude that substantial evidence supports the district court's denial of Ariosa's motion for judgment as a matter of law on noninfringement and invalidity. We also conclude that the district court did not abuse its discretion by denying Verinata and Illumina's motion for a permanent injunction, supplemental damages, an accounting, and pre-judgment interest. We affirm.
Appellant Illumina, Inc., develops, manufactures, and markets integrated systems and tools for DNA analysis. Verinata Health, Inc., a wholly-owned subsidiary of Illumina (collectively “Illumina”), developed and offered a noninvasive prenatal test (“NIPT”) for the early identification of fetal chromosomal abnormalities. Appellee Ariosa Diagnostics, Inc., also conducts research and development in the field of NIPT for fetal chromosomal abnormalities. Roche Molecular Systems, Inc., acquired Ariosa in December 2014. In an effort to “streamline issues in the [l]itigation and avoid unnecessary discovery,” the parties stipulated that “Ariosa will be deemed the Defendant responsible for the conduct that Illumina has accused of infringing the asserted claims” and that Roche would be dismissed from the litigation and subsequently “deemed a party to any judgment to the same extent as Ariosa.” J.A. 11606-07.
Illumina owns U.S. Patent No. 7,955,794 (the “’794 patent”), which is directed to custom DNA assay optimization techniques. The ’794 patent identifies seven inventors, including Dr. John Stuelpnagel and Dr. Arnold Oliphant. Dr. Stuelpnagel was a co-founder of Illumina, and Dr. Oliphant served as Illumina's executive vice president of scientific operations.
The ’794 patent describes the detection of DNA target sequences by introducing probes with complementary sequences into a sample and observing whether hybridization occurs. An excerpt of claim 1 identifying the elements relevant to this appeal is set forth below:
A multiplex for determining whether a sample contains at least 100 different target sequences, comprising:
a) providing a sample which may contain at least 100 different single-stranded target sequences attached to a first solid support;
b) contacting said target sequences with a probe set comprising more than 100 different single-stranded probes, wherein each of said more than 100 different probes comprises:
i) a first universal priming site, wherein each of said more than 100 different probes has identical universal priming sites, and
ii) a target specific domain, such that different double-stranded hybridization complexes are formed, each of the different hybridization complexes comprising one of said more than 100 different single-stranded probes and one of the different single-stranded target sequences from the sample;
d) contacting said probes of the hybridization complexes with a first enzyme and forming different modified probes;
e) contacting said modified probes with:
i) at least a first primer that hybridizes to said universal priming site;
ii) NTPs; and
iii) an extension enzyme;
wherein said different modified probes are amplified and forming different amplicons;
f) immobilizing said different amplicons to a second solid support, and
g) detecting said different amplicons immobilized to said second solid support, thereby determining whether the sample contains at least 100 different target sequences.
’794 patent col. 68 ll. 46-67, col. 68 l. 65-col. 69 l. 12.
Verinata owns U.S. Patent No. 8,318,430 (the “’430 patent”), which is directed to methods for NIPT screening of fetal chromosomal abnormalities. An excerpt of claim 1 is appended below identifying the elements relevant to this appeal:
1. A method for determining a presence or absence of a fetal aneuploidy in a fetus for each of a plurality of maternal blood samples ․ comprising fetal and maternal cell-free genomic DNA, said method comprising:
(e) ․ enumerating sequence reads corresponding to enriched and indexed fetal and maternal non-random polynucleotide sequences ․; and
(f) ․ determining the presence or absence of a fetal aneuploidy comprising using a number of enumerated sequence reads corresponding to the first chromosome and a number of enumerated sequence reads corresponding to the reference chromosome of (e).
’430 patent at col. 63.
In 2008, both Dr. Stuelpnagel and Dr. Oliphant left Illumina. By late 2009, Dr. Stuelpnagel launched Ariosa. Dr. Oliphant rejoined Dr. Stuelpnagel at Ariosa shortly thereafter. They sought to develop a NIPT for the detection of fetal aneuploidies, which can lead to conditions such as Down Syndrome. Between 2010 and 2011, Ariosa provided Illumina, as a prospective investor in Ariosa, technical information about its product proposals under development. In January 2012, seven months after the ’794 patent issued, Ariosa entered into a three-year sale and supply agreement (“SSA”) with Illumina. J.A. 4326, J.A. 4349-4350 (excerpts from SSA).
In March 2012, Ariosa launched a DNA-sequencing test called the Harmony Prenatal Test. The test consisted of materials supplied by Illumina. The Harmony Prenatal Test is a multiplex method that analyzes fetal cell-free DNA (or cfDNA). Ariosa designed two versions of the Harmony test—“Harmony V1” and “Harmony V2.” For purposes of this appeal, we focus our discussion of the relevant technology on Harmony V2.
Harmony V2 tests a sample of isolated fetal cfDNA for the presence of about 6800 gene sequences by using a laboratory robot to perform the steps summarized below in Figure 1.
J.A. 3100-3101; J.A. 2067-2068. First, the sample's double-stranded fetal cfDNA is separated, or “denatured,” into individual strands. Next, a molecule called biotin is added to the end of each cfDNA strand (represented by “B” in Figure 1). The robot then adds a solution containing a mixture of single-stranded oligonucleotides that are complementary to the 6800 sequences Harmony V2 detects (orange lines in Figure 1). The mixture contains three different oligonucleotides for each of the 6800 target sequences, corresponding to the beginning, middle, and end portions of the target sequence. The oligonucleotide for the beginning of each sequence contains a “readout cassette,” which is a short, artificial DNA segment that is uniquely assigned to each of the 6800 sequences tested in Harmony V2. If the cfDNA sample contains one of the 6800 target sequences, each of the three oligonucleotides corresponding to that target sequence will hybridize to it, creating a section of double-stranded DNA with two gaps (between the first and second and between the second and third oligonucleotides). If the cfDNA does not contain a certain target sequence, the oligonucleotides corresponding to that sequence will remain unbound in solution.
The test allows the oligonucleotides two hours to bind to target sequences. After the two hours elapse, the robot adds magnetic beads coated with a protein called streptavidin, which binds strongly with the biotin on the cfDNA and links it to the beads. The robot then immobilizes the magnetic beads (and therefore the sample DNA and any bound oligonucleotides) and washes away anything that is left in solution, including any unbound oligonucleotides.
Next, the robot adds an enzyme that ligates, i.e., connects, the three oligonucleotides, creating a single DNA strand. This only happens if all three oligonucleotides corresponding to the target sequence are bound to the sample cfDNA. The robot then denatures, i.e., separates, the newly-joined oligonucleotides from the sample cfDNA and amplifies the newly-joined oligonucleotides. Universal primer sequences on the first and third oligonucleotides enable this amplification.
During processing, the copies that result from the amplification step (termed “amplicons”) are purified and added to a mixture that cuts (“digests”) them into fragments. Then, detection begins by applying the digested reaction mixture, including the readout cassettes, to an array. An array is a chip (or device) containing thousands of short DNA sequences attached to a solid support. If a readout cassette corresponding to one of the 6800 target sequences is present, part of the readout cassette will bind to a DNA sequence on the array. The other part of the readout cassette remains unbound, hanging like a single-stranded tail off the double-stranded sequence attached to the solid support. Figure 2, below, shows how readout cassettes indicating target sequences on chromosomes 18 and 21 bind to the array while other readout cassettes remain unbound.
Any materials that do not bind to the array, e.g., chromosomes Y, X, and 13 in Figure 2, are washed away. Readout cassettes remain bound to the array. Fluorescently labeled oligonucleotides that are complementary to the readout cassettes’ free single-stranded tails are then added. After the labeled oligonucleotides are given time to bind to the single-stranded tails on the readout cassettes, they are chemically joined or ligated to the DNA strand attached to the chip. The array is then heated up to separate the readout cassettes from the fluorescently tagged chip. The readout cassettes are then washed away, leaving only the labeled oligonucleotides attached to the DNA strands.
A machine then analyzes the array and detects the different colors of the fluorescent tags and their positions. From these data, and using algorithms and analyses, Ariosa can calculate the probability that each of the 6800 sequences was present in the cfDNA sample.
Starting late 2012, Illumina and Verinata filed several lawsuits against Ariosa and its parent company Roche accusing the Harmony V1 and V2 tests of infringing the ’794 patent and the ’430 patent. Verinata alleged Harmony V1 infringed the ’430 patent, and Illumina alleged both Harmony versions infringed the ’794 patent. Ariosa argued that the patents-in-suit were invalid and that it had an express license to the ’794 patent. Ariosa also asserted a counterclaim for breach of contract.
During claim construction, the parties disputed the construction of two terms of the ’794 patent: (a) “modified probes” and (b) “wherein said different modified probes are amplified and forming different amplicons.” The district court construed those claims as follows:
• “modified probe” means “an enzymatically altered polynucleotide which contains a universal priming site and is capable of substantially hybridizing to a target sequence.”
• “wherein said different modified probes are amplified and forming different amplicons” means “wherein the different modified probes are replicated, in whole or in part, to yield amplification products of each of the different modified probes.”
The district court held a jury trial from January 8 to January 25, 2018. The jury returned a verdict finding the ’430 patent not invalid and infringed by the Harmony V1 product and the ’794 patent not invalid and infringed by both the Harmony V1 and Harmony V2 products; that Ariosa did not have an express license to the Harmony V1 product under the SSA; and that Illumina did not breach the SSA by suing Ariosa. The jury awarded plaintiffs approximately $27 million in damages. Thereafter, the parties filed post-trial motions.
Ariosa moved for judgment as a matter of law (“JMOL”), under Fed. R. Civ. P. 50(b), on the jury's various infringement and validity determinations. Illumina moved for a permanent injunction, a Fed. R. Civ. P. 52 conclusion of law that Ariosa was estopped as an assignor from challenging the validity of the ’794 patent, and an accounting, supplemental damages, pre-judgment interest at the prime rate and post-judgment interest.
The district court denied Ariosa's motions for JMOL. The district court found that substantial evidence supported the jury's findings of no anticipation of the ’794 patent by U.S. Patent Application No. 2003/0228599 A1 to Straus (“Straus”); that the Harmony V2 product infringes the ’794 patent; that the ’430 patent meets the enablement requirement; and that the Harmony V2 product infringes the ’430 patent. The district court granted Illumina's motion for a Rule 52 conclusion of law and denied Illumina's motion for an accounting, and supplemental damages. The district court granted pre-judgment interest at the 52-week Treasury Bill rate and granted post-judgment interest at the statutory rate but deferred on calculating post-judgment interest until after appeal once the final amount of the judgment is known.
These appeals ensued. Illumina appeals the denial of a permanent injunction, supplemental damages, an accounting, and pre-judgment interest at the prime rate. Ariosa cross-appeals the denial of JMOL on validity of the ’430 patent and the ’794 patent and infringement of only the ’794 patent by Ariosa's Harmony V2 product.
We address Ariosa's cross-appeal in §§ A, B, and C below. Then, in § D, we address Illumina's appeal.
We begin by addressing the district court's denial of Ariosa's motion for JMOL of noninfringement of the ’794 patent. Ariosa argues that Harmony V2 does not literally infringe claim 1, steps (a) and (b). Ariosa also argues that Harmony V2 does not infringe claim 1, steps (f) and (g) literally or under the doctrine of equivalents. The district court's denial of Ariosa's motion for JMOL is supported by substantial evidence.
We review denials of JMOL under the law of the relevant regional circuit, in this case, the Ninth Circuit. A TEN Int'l Co., Ltd. v. Uniclass Tech. Co., Ltd., 932 F.3d 1364, 1367 (Fed. Cir. 2019). The Ninth Circuit reviews a denial of JMOL de novo. Harper v. City of Los Angeles, 533 F.3d 1010, 1021 (9th Cir. 2008). JMOL is proper when the evidence permits only one reasonable conclusion that itself is contrary to the jury's verdict. Id. But the jury's verdict must be upheld if it is supported by substantial evidence. Id. Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” TVIIM, LLC v. McAfee, Inc., 851 F.3d 1356, 1362 (Fed. Cir. 2017) (citation and quotation omitted).
A party asserting infringement under the doctrine of equivalents may prove its case by showing, on an element-by-element basis, that the accused product performs substantially the same function in substantially the same way with substantially the same result as each claim limitation of the patented product. See, e.g., Crown Packaging Tech., Inc. v. Rexam Beverage Can Co., 559 F.3d 1308, 1312 (Fed. Cir. 2009).
Ariosa argues that its Harmony V2 does not literally infringe the step (a) “providing” and the step (b) “contacting” processes of the ’794 patent. Cross-App. Br. 40-47. Ariosa argues that Dr. Cooper, Illumina's expert, offered no evidence that at least 100 different single-stranded target sequences remain completely unbound from any probe after the two-hour hybridization period. Ariosa further argued that Dr. Cooper presented no evidence that any unbound single-stranded target sequences would bind to all three probes during the short period between the addition of the streptavidin beads and the washing-away of the probes.
Dr. Cooper detailed the reaction conditions in Ariosa's Harmony V2 that practice the method recited in step (a). J.A. 1965-1968. He explained that Harmony V2's annealing reaction is less than 99% complete following the two-hour incubation time. Id. He explained that Harmony V2's hybridization would occur after step (a) as a function of the relative rates of the slower “annealing reaction” compared to the faster “hybridization reaction.” J.A. 1951-1952; J.A. 1955; J.A. 1964-1965; J.A. 2675-2676. Dr. Cooper concluded that, after annealing, at least 100,000 single-stranded target sequences attach to a solid support before hybridization takes place. J.A. 1967. Dr. Cooper testified that, given the reaction setup, the annealing reaction is “unlikely to complete or come close.” See J.A. 2676.
Dr. Cooper also testified regarding how the solid support first attaches to 100 different single-stranded target sequences and how the target sequences hybridize to the probes as recited in step (b). According to Dr. Cooper, after two hours, the solid support is added and the process is “allow[ed] continued time to proceed.” J.A. 1964-1965. Dr. Cooper explained that the solid support streptavidin beads quickly attach to the target sequences given the “extremely strong” covalent bond between streptavidin and biotin-coated cell-free DNA fragments. J.A. 1951-1952. Given the additional time and the strong bond between the solid support and the target sequences, Dr. Cooper testified that the reaction allows for the 100 single-stranded target sequences to “hybridize with their oligos.” J.A. 1964-1965. Dr. Cooper concluded, therefore, that Ariosa's Harmony V2 practices steps (a) and (b) of claim 1. Dr. Cooper's testimony constitutes substantial evidence supporting the verdict of infringement.
Ariosa argues that the Harmony V2 does not literally infringe claim 1, steps (f) and (g) of the ’794 patent because its readout cassettes do not meet claim 1's “amplicons” element. Cross-App. Op. Br. 28-31. Ariosa argues that after the amplification step performed in Harmony V2, the readout cassette is only a portion of each of the amplified DNA segments and not the complete “amplicon” that is required by the claims. Ariosa argues in the alternative that even if the readout cassettes are amplicons, Harmony V2 does not practice step (g)’s “detecting said different amplicons immobilized to said second solid support.” ’794 patent 69 ll. 10-12. Ariosa argues that because the readout cassettes are washed away from the array before the detection step takes place, the amplicons are not detected while attached to a second solid support.
Finally, Ariosa argues that Illumina failed to prove infringement of claim 1, steps (f) and (g) under the doctrine of equivalents. Cross-App. Br. at 31-35. Ariosa contends that the differences between the claimed amplicons and Ariosa's readout cassettes are substantial such that no evidence supports a doctrine of equivalents analysis. Ariosa further contends that Illumina failed to prove that immobilizing and detecting readout cassettes leads to insubstantially different results from immobilizing and detecting amplicons. We disagree.
Even were we to accept Ariosa's arguments for literal infringement, Ariosa fails to demonstrate that a reasonable jury could not find infringement under the doctrine of equivalents. Dr. Cooper testified that the readout cassettes and amplicons serve substantially the same function of “immobiliz[ing] onto a solid support”; in substantially the same way of “hybridization of [the] DNA molecule”; to achieve the same result of “detection of the target sequences that were in the original mixture.” J.A. 2683-2684, J.A. 1979-1985. That testimony constitutes evidence that a reasonable mind could accept as proving infringement under the doctrine of equivalents.
Next, we address the district court's denial of JMOL of invalidity of the ’794 patent. We conclude that the district court's denial of JMOL is supported by substantial evidence.
Ariosa appeals the district court's holding of assignor estoppel—that Ariosa is barred from challenging the validity of the ’794 patent because Drs. Stuelpnagel and Oliphant are inventors of the ’794 patent, they assigned their rights to the patent to Illumina, and they are in privity with Ariosa. See Verinata Health, Inc. v. Ariosa Diagnostics, Inc., 329 F. Supp. 3d 1070, 1113-18 (N.D. Cal. 2018). Despite its finding of assignor estoppel, the district court analyzed anticipation of the ’794 patent and found it invalid. Because we affirm the jury verdict of no invalidity, we need not reach the issue of assignor estoppel.
Ariosa contends that the district court improperly denied its motion for JMOL on anticipation of the ’794 patent based on the Straus prior art reference. Straus discloses multiplex methods for detecting more than 250 nucleic-acid sequences, such as the signature sequences of pathogens in a blood sample using DNA probes. See J.A. 5395-5441.
Ariosa argues that a skilled artisan reading Straus and the method depicted in Straus Figure 5 would understand that it discloses “ ‘numerous’ pathogens includ[ing] using at least 100 different target sequences and over 100 different single-stranded probes” as claimed in claim 1 of the ’794 patent. Ariosa further argues that Straus's disclosure of “a large number of distinct ID probes” anticipates the claimed universal priming sites because those probes disclose “substantial if not complete identity in the probes’ priming sites.” Finally, Ariosa argues that Straus need not disclose all the claimed limitations in a single disclosure or figure in order to anticipate.
Illumina disagrees, arguing that Dr. Cooper's testimony shows why Straus fails to anticipate the ’794 patent. Dr. Cooper focused on Straus's failure to disclose claim 1, step (b)(i) (“a first universal priming site, wherein each of said more than 100 different probes has identical universal priming sites”). Dr. Cooper testified that Straus discloses only forty-eight probes in Figure 5, well below the “level of multiplexing” required by the ’794 Patent, and that Straus is silent as to the actual number of primers that would be used. J.A. 2597-2598; see also J.A. 2602. Dr. Cooper further testified that Straus's references to ID probes confirms that there is no anticipation because ID probes “teach towards multiple different amplification sequences” and not a single universal primer as required by claim 1, step (b)(i). See J.A. 2600-2602. Dr. Cooper opined that even if some isolated disclosure in Straus did disclose or suggest a universal primer, that disclosure would fail to anticipate claim 1, step (b)(i), for it is unlinked to the disclosures on which Ariosa relies for anticipation, namely Figure 5. See J.A. 2654.
Ariosa's arguments are unavailing. Ariosa asks this court to reweigh the credibility of the parties’ respective expert witnesses. This court does not engage in fact finding, nor does it weigh the credibility of expert testimony. See Impax Labs. Inc. v. Lannett Holdings Inc., 893 F.3d 1372, 1382 (Fed. Cir. 2018). Our task is to review whether the jury's verdict is supported by substantial evidence.
Here, the jury heard conflicting expert testimony on whether Straus discloses a single universal primer. The jury was free to adopt Dr. Cooper's testimony over that of Dr. Cantor's in concluding that Straus did not disclose a single universal primer. See i4i Ltd. P'ship v. Microsoft Corp., 598 F.3d 831, 848 (Fed. Cir. 2010), aff'd, 564 U.S. 91, 131 S.Ct. 2238, 180 L.Ed.2d 131 (2011). We conclude that the jury verdict on invalidity is supported by substantial evidence.
We therefore affirm the jury's verdict of no invalidity and the district court's subsequent denial of Ariosa's motion for JMOL.
Next, we address whether substantial evidence supports the district court's denial of Ariosa's motion for JMOL of no enablement of the ’430 patent. We conclude that the jury's finding and the district court's denial of JMOL are supported by substantial evidence.
Enablement is a question of law reviewed de novo. Trustees of Bos. Univ. v. Everlight Elecs. Co., 896 F.3d 1357, 1361 (Fed. Cir. 2018). However, in the context of a jury trial, we review the factual underpinnings of enablement for substantial evidence. Id. The enablement requirement ensures that a patent contains a written description of the invention that enables “any person skilled in the art to which [the invention] pertains ․ to make and use the [invention]” without undue experimentation. 35 U.S.C. § 112(a); Storer v. Clark, 860 F.3d 1340, 1345 (Fed. Cir. 2017).
Ariosa argues that the ’430 patent does not meet the enablement requirement because the patent fails to disclose an algorithm for determining the presence or absence of a fetal aneuploidy in the context of a targeted sequencing approach as claimed in claim 1, step (f). Cross-App. Br. 55-58. Ariosa agrees that the ’430 patent incorporates by reference disclosures of “[m]ethods for determining fetal aneuploidy using random sequencing techniques.” Id. at 56 (citing J.A. 268 (12:49-55)). Ariosa contends, however, that a skilled artisan would not be able to adapt those random sequencing techniques into non-random sequencing data without undue experimentation. Ariosa relies on the testimony of Dr. Rava, a named inventor of the ’430 patent, and argues that Dr. Rava testified that a skilled artisan would be unable to use “random sequencing techniques ․ in a non-random method without modification.” Id. (citing J.A. 1344-1345). Ariosa argues that the ’430 patent discloses no such modification. Ariosa argues that even if the disclosures incorporated by reference could be modified for use in random sequencing techniques, their limited disclosure would not suffice to enable the full scope of the claimed invention.
In response, Illumina raises three main arguments. First, Illumina argues that Ariosa's expert, Dr. Cantor, testified that the Quake 1 and Craig 2 prior art references disclose the alleged missing enablement teachings of the ’430 patent and that a skilled artisan is presumed to be aware of all pertinent prior art. Appellant Reply and Resp. Br. 62 (citing J.A. 2490). Illumina argues that these references disclose methods for analyzing targeted regions of DNA sequences as claimed in the ’430 patent. Second, Illumina argues that Dr. Rava testified that “the algorithms for random ․ sequencing described in the publications referenced in the ’430 [p]atent can be ‘very similar to the ones that would be use[d] in a directed sequencing approach’ but ‘would have to be optimized.’ ” Id. at 64 (citing J.A. 1344-1345). Illumina further contends that Dr. Cooper confirmed that the references in the ’430 patent disclose numerous enabling techniques to determine fetal aneuploidy. Third, Illumina argues that, according to Dr. Cooper, “the exact statistical methods the ’430 Patent discloses based on Z-scores were in fact used by Roche scientists—and were ‘quite effective’ at determining fetal aneuploidy for the targeted approach.” Id. (citing J.A. 2619-2621).
We conclude that a reasonable mind might accept Dr. Cooper's testimony that Roche scientists used the same statistical methods disclosed in the ’430 patent to determine fetal aneuploidy in a targeted approach as evidence to support enablement of the ’430 patent. See Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384, (Fed. Cir. 1986) (finding specification was enabling where evidence showed the necessary screening and producing methods for making the monoclonal antibodies used in the claimed invention were known in the prior art). We therefore affirm the jury's verdict regarding enablement and the district court's subsequent denial of Ariosa's motion for JMOL.
Finally, we address whether the district court abused its discretion in denying Illumina's request for injunctive relief, supplemental damages, an accounting, and pre-judgment interest at the prime rate. We conclude that the district court did not abuse its discretion.
We review a district court's grant or denial of injunctive relief for abuse of discretion. Genband US LLC v. Metaswitch Networks Corp., 861 F.3d 1378, 1381 (Fed. Cir. 2017). A district court abuses its discretion if its ruling is based on an erroneous view of the law or on a clearly erroneous assessment of the evidence. Id. A plaintiff seeking a permanent injunction must satisfy a four-factor test before a court may grant such relief. eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388, 391, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006). A plaintiff must demonstrate that: (1) it has suffered an irreparable injury; (2) remedies available at law are inadequate to compensate for that injury; (3) considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) the public interest would not be disserved by a permanent injunction. Id. Because we affirm the district court's conclusion on irreparable injury and adequacy of monetary damages, we need not reach the district court's conclusions on balance of harms and public interest. See Nichia Corp. v. Everlight Americas, Inc., 855 F.3d 1328, 1340 (Fed. Cir. 2017).
Regarding irreparable injury, Illumina argues that the district court failed to recognize that Roche and Illumina are direct competitors and that Roche's infringement causes irreparable injury because each sale made by Roche is a sale forever lost by Illumina. Appellant Op. Br. 22-23. Illumina argues that the district court's understanding of ActiveVideo Networks, Inc. v. Verizon Communications, Inc., 694 F.3d 1312 (Fed. Cir. 2012), was too broad and caused it to err in its conclusion of no direct competition. Id. at 26-30. We disagree.
In ActiveVideo Networks, we held a lack of direct competition is a substantial basis for finding no irreparable harm. 694 F.3d. at 1338. We reversed the injunction because the defendant (Verizon) competed with ActiveVideo's third-party licensees but not with the patentee (ActiveVideo). Id. The harm to ActiveVideo was therefore indirect, and ActiveVideo's loss was a “[s]traight-forward monetary harm” and “certainly not irreparable.” Id. Here, the district court found that Illumina licenses its patents and products under the SSA, allowing third party laboratories to conduct their own tests. J.A. 58 (citing J.A. 2109:9-15). The district court also found that Ariosa does not utilize a licensing model but instead sells its Harmony V2 test directly. Id. Relying on ActiveVideo, the district court found that the different sales models evidenced a lack of direct competition because defendants compete with Illumina's licensees. Id. The district court concluded that defendants’ losses would be quantifiable based at least on licensing fees per lost subscriber. J.A. 59. As we find no reason to disturb the district court's findings on irreparably injury, we turn to the next eBay factor, available remedies.
Illumina argues that the district court erred by finding that monetary relief would be adequate. Illumina reasserts that the district court erred in its reliance on ActiveVideo and its reasoning that, where licensees compete with the infringer, royalties are adequate forms of compensation. See J.A. 60 (citing ActiveVideo, 694 F.3d at 1338). As noted above, the district court's reliance on ActiveVideo does not constitute an abuse of discretion. And Illumina does not challenge the district court's finding that third-party licensees compete with Ariosa. See J.A. 58-59. Because Illumina failed to establish irreparable injury and inadequacy of monetary relief, the district court did not abuse its discretion in denying Illumina's request for a permanent injunction.
Regarding Illumina's request for supplemental damages, and an accounting, Illumina argues that the district court's order deferring its request until after the resolution of this appeal created confusion regarding whether it is entitled to supplemental damages and an accounting. We decline to decide, in the first instance, whether Illumina is entitled to the supplemental damages it seeks. See La Van v. United States, 382 F.3d 1340, 1350 (Fed. Cir. 2004) (declining to award damages in the first instance on appeal). And we do not fault the district court's decision to defer this issue. Cf., Sanofi-Aventis Deutschland GmbH v. Glenmark Pharm. Inc., USA, 748 F.3d 1354, 1357 (Fed. Cir. 2014) (explaining that district court's provision for an accounting of any additional damages that may accrue if the decision is affirmed on appeal did not negate finality of the judgment).
Regarding the district court's granting of pre-judgment interest at the 52-week Treasury Bill rate, Illumina re-quests we reverse and remand with an order to award prejudgment interest at the prime rate. Appellant Op. Br. 50-51. But Illumina articulates no reason in its opening brief for why a higher rate is appropriate. District courts have wide latitude in the selection of interest rates, Uniroyal, Inc. v. Rudkin-Wiley Corp., 939 F.2d 1540, 1545 (Fed. Cir. 1991), and prejudgment interest awards at the Treasury Bill rate are well within the court's discretion. See Laitram v. NEC Corp., 115 F.3d 947, 955 (Fed. Cir. 1997). The district court's decision does not constitute an abuse of discretion.
We have considered the parties’ remaining arguments and find them unpersuasive. For the foregoing reasons, we conclude that substantial evidence supports the district court's denial of Ariosa's motion for JMOL of noninfringement and invalidity. We also conclude that substantial evidence supports the district court's denial of Ariosa's motion for JMOL of no enablement of the ’430 patent.
We conclude that the district court did not abuse its discretion in denying Illumina's motion for a permanent injunction. We conclude that the district court did not abuse its discretion in denying Illumina's request for supplemental damages and an accounting. Finally, we conclude that the district court did not abuse its discretion in awarding pre-judgment interest at the 52-week Treasury Bill rate.
The parties shall bear their own costs.
1. U.S. Patent App. Pub. No. 2007/0202525 (published August 30, 2007, filed February 2, 2007).
2. Craig, et al., “Identification of genetic variants using bar-coded multiplexed sequencing,” Nature Methods, 5(10):887-93 (2008).
Reyna, Circuit Judge.