Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 1 of 19 PageID #: 11769
`
`THE UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF DELAWARE
`
`GUARDANT HEALTH, INC.,
`
`Plaintiff,
`
`v.
`
`FOUNDATION MEDICINE, INC.,
`
`Defendant.
`
`)
`)
`)
`)
`)
`)
`)
`)
`)
`
`
`C.A. No. 20-1580 (LPS)
`
`FILED UNDER SEAL
`
`REPLY IN SUPPORT OF PLAINTIFF GUARDANT HEALTH, INC.’S
`MOTION FOR PRELIMINARY INJUNCTION
`
`Brian E. Farnan (Bar No. 4089)
`Michael J. Farnan (Bar No. 5165)
`FARNAN LLP
`919 N. Market Street, 12th Floor
`Wilmington, Delaware 19801
`Telephone: (302) 777-0300
`Facsimile: (302) 777-0301
`bfarnan@farnanlaw.com
`mfarnan@farnanlaw.com
`
`Attorneys for Plaintiff Guardant Health, Inc.
`
`Edward R. Reines (admitted pro hac vice)
`Derek C. Walter (admitted pro hac vice)
`WEIL, GOTSHAL &MANGES LLP
`201 Redwood Shores Parkway
`Redwood Shores, CA 94065
`Telephone: (650) 802-3000
`Facsimile: (650) 802-3001
`edward.reines@weil.com
`derek.walter@weil.com
`
`Garland T. Stephens (admitted pro hac vice)
`Justin Constant (admitted pro hac vice)
`WEIL, GOTSHAL & MANGES LLP
`700 Louisiana, Suite 1700
`Houston, TX 77002
`Telephone: (713) 546-5000
`Facsimile: (713) 224-9511
`garland.stephens@weil.com
`justin.constant@weil.com
`
`
`
`
`
`
`
`
`
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 2 of 19 PageID #: 11770
`
`
`
`I. 
`
`II. 
`
`TABLE OF CONTENTS
`
`Page
`
`INTRODUCTION ...............................................................................................................1 
`
`FMI’S OWN ARGUMENTS PROVE THAT THE ACCUSED PRODUCT
`INFRINGES .........................................................................................................................1 
`
`III. 
`
`FMI CANNOT SHOW THAT THE ASSERTED CLAIMS ARE INVALID ....................3 
`
`A. 
`
`B. 
`
`FMI’s Obviousness Arguments Are Meritless ........................................................3 
`
`FMI’s § 112 Arguments Are Meritless ....................................................................5 
`
`1. 
`
`2. 
`
`3. 
`
`“grouping…based on…a start base position”/“stop base position”
`......................................................................................................................5 
`
`“ligating a set of molecular barcodes…wherein a given molecular
`barcode is a member of a set of molecular barcodes comprising 2
`to 1,000,000 different molecular barcode sequences” (’085
`patent) ..........................................................................................................7 
`
`“wherein the cfDNA molecules that map to a mappable base
`position of a reference sequence are tagged with a number of
`different molecular barcodes ranging from at least 2 and fewer
`than a number of cfDNA molecules that map to the mappable
`base position” (’086 patent) .........................................................................8 
`
`IV. 
`
`V. 
`
`THE ASSERTED CLAIMS ARE ENFORCEABLE ..........................................................9 
`
`FMI’S EVIDENCE SHOWS IRREPARABLE HARM TO GUARDANT AND
`THAT THE BALANCE OF HARMS FAVORS AN INJUNCTION...............................12 
`
`VI. 
`
`THE PUBLIC WILL NOT BE HARMED BY AN INJUNCTION ..................................13 
`
`VII.  CONCLUSION ..................................................................................................................15 
`
`
`
`
`i
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 3 of 19 PageID #: 11771
`
`
`
`TABLE OF AUTHORITIES
`
`Page(s)
`
`Cases 
`Erfindergemeinschaft v. Eli Lilly,
`240 F.Supp 605 (E.D. Tex. 2017) .............................................................................................. 6
`
`Therasense, Inc. v. Becton, Dickinson & Co.,
`649 F.3d 1276 (Fed. Cir. 2011) ................................................................................................ 10
`
`Statutes and Regulations 
`35 U.S.C. § 112 ............................................................................................................................... 6
`
`
`
`
`ii
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 4 of 19 PageID #: 11772
`
`
`
`I.
`
`INTRODUCTION
`
`FMI’s opposition brief and evidence fails to rebut Guardant’s showing that a preliminary
`
`injunction should issue in this case.
`
`II.
`
`FMI’S OWN ARGUMENTS PROVE THAT THE ACCUSED PRODUCT
`INFRINGES
`
`FMI argues that FoundationOne Liquid CDx (the Accused Product) does not infringe the
`
`’085 and ’086 Patents because, it claims, the Accused Product does not group sequencing reads
`
`“based on… at least (1) a start base position… at which the given mapped sequencing read is
`
`determined to start mapping to the reference sequence.” D.I. 26 at 4, ’086 Patent claim 1.
`
`However, FMI’s own arguments and the declaration of its expert show that the Accused Product
`
`does group sequences based on at least a start base position as claimed, demonstrating that the
`
`Accused Product infringes at least claim 1 of the ’086 Patent. FMI’s arguments and expert
`
`declarations also fail to rebut Guardant’s showing that the Accused Product infringes claim 1 of
`
`the ’085 Patent.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`1
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 5 of 19 PageID #: 11773
`
`Grouping is therefore based on the start base position as claimed.
`
`and the Accused Product infringes claim 1 of the ’086 Patent.
`
`See also Cooper Decl. W 12-15
`
`The claim recites grouping “based at least on. .. a start base position. . and-
`
`- FMI has offered no other argument that the Accused Product does not literally infringe
`
`claim 1 of the ’086 Patent as Dr. C ooper’s declaration established. Cooper Decl. W 11-20.
`
`1 Even if the Court were to acce t FMI’s aroument. the Accused Product still infringes.
`
`
`
`
`2 FMI cites Auburn Univ. v. [BA], 2012 WL 3151545 for the proposition “that X is a necessary
`precondition for Y does not necessarily mean that Y is ‘based on’ X.” Opp. Br. at 5. But in
`Auburn the “necessary precondition” (that a manufactured part was classified as good) was not
`used at all to make claimed determination an estimate of reliabili
`Auburn, at *14.
`
`
`
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 6 of 19 PageID #: 11774
`
`Therefore, the Accused Product
`
`meets this limitation literally even under FMI’s theory. Cooper Decl. 1] 22.
`
`III.
`
`FMI CANNOT SHOW THAT THE ASSERTED CLAIMS ARE INVALID
`
`A.
`
`FMI’s Obviousness Arguments Are Meritless
`
`Notably. FMI does not claim that the ’085 and ”086 Patents are anticipated by any prior art
`
`and instead does nothing more than rehash obviousrress arguments that it previously made to the
`
`PTAB and lost. FMI attempts to distract from this fact by characterizing the PTAB’s decisions
`
`as preliminary, but fails to note that the standard FMI failed to meet at this stage was even lower
`
`than What is required in a PTAB final decision and much lower than the “clear and convincing
`
`evidence” standard required by this Court. All of the art that FMI now asserts here was available
`
`to FMI when it filed its IPRs against the ’992 Patent. and the limitation the PTAB found rrrissing
`
`3
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 7 of 19 PageID #: 11775
`
`
`
`is present in the asserted claims. Because FMI failed to establish invalidity before under a much
`
`lower standard, there is no reason to believe it will succeed here.
`
`FMI, through its expert, Dr. Gabriel, argues that 1) Schmitt alone, 2) Schmitt in view of
`
`Hendricks, 3) Schmitt in view of Hendricks and Fan/Forshew, and 4) Rava in view of NEB Manual
`
`and Schmitt renders the claims obvious. These combinations have many flaws, but they all fail
`
`to teach at least “wherein at least 20% of the double-stranded cfDNA molecules from the
`
`population of cfDNA molecules are attached to molecular barcodes” (a 20% ligation efficiency)
`
`as required by the asserted claims.
`
`Dr. Gabriel admits that there are many factors that can affect ligation efficiency and a
`
`person of ordinary skill in the art cannot know what ligation efficiency will result unless they
`
`perform an experiment themselves. Ex. 2 [Gabriel Tr.] at 29:9-16, 33:5-12; see also id. at 25:23-
`
`26:2, 26:11-20, 27:6-20, 31:8-19, 33:14-34:25 (identifying relevant factors including temperature,
`
`duration, concentration of adapters, buffer composition, enzyme used, and the kit used). As Dr.
`
`Gabriel admits, Schmitt does not disclose what ligation efficiency was achieved by the described
`
`method. D.I. 31 ¶ 141. Dr. Gabriel relies instead on experiments performed by Dr. Mortimer
`
`and a Guardant employee, Alain Mir, which Dr. Gabriel admits are different from the Schmitt
`
`method. Ex. 2 [Gabriel Tr.] at 38:25-39:7. In deposition, Dr. Gabriel had no opinion as to how
`
`the differences between the methods would change the ligation efficiency that was achieved. Id.
`
`at 43:14-24, 44:14-45:12. Despite these admissions, Dr. Gabriel never attempted to perform the
`
`Schmitt method herself to determine the resulting ligation efficiency and could not identify any
`
`barriers to her doing so. Id. at 42:5-16.
`
`In an attempt to cure FMI’s failure to show a 20% ligation efficiency, Dr. Gabriel points
`
`to other references showing instances where a higher ligation efficiency occurred in certain
`
`4
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 8 of 19 PageID #: 11776
`
`
`
`controlled circumstances. D.I. 31 ¶ 162, 249. This fails for the same reason as Schmitt. Even
`
`if high ligation efficiencies are found in conditions not disclosed by Schmitt, FMI fails to identify
`
`any method that satisfies all of the claim limitations, including achieving a 20% ligation efficiency,
`
`that would have been performed by a person of ordinary skill in the art with a reasonable
`
`expectation of success. For example, Dr. Gabriel admits that Hendricks utilized different
`
`temperatures and for a different duration than Schmitt (Ex. 2 at 106:22-107:11), factors that she
`
`admits affect ligation efficiency. She, however, fails to identify what specific method a person
`
`of ordinary skill in the art would perform in view of Schmitt and Hendricks or whether it would
`
`meet all the limitations. For these reasons, FMI’s Schmitt combinations fail to raise a substantial
`
`question of patentability.
`
`The same issues plague Dr. Gabriel’s asserted combination of Rava and the NEB Manual.
`
`While she claims that the NEB Manual discloses ligation efficiencies of over 95% in certain
`
`circumstances, she fails to explain what specific experiment a person of ordinary skill in the art
`
`would have been motivated to perform in light of Rava and the NEB Manual, much less what
`
`ligation efficiency would be achieved in those circumstances. Thus, for the same reasons that the
`
`Schmitt combinations fail, Dr. Gabriel’s Rava combination fails as well.
`
`Because FMI has already tried, and failed, to demonstrate that these same limitations are
`
`taught when it attempted to invalidate the related ’992 Patent, with a lower burden of proof, before
`
`the PTAB, FMI has not demonstrated a substantial question of invalidity.
`
`B.
`
`FMI’s § 112 Arguments Are Meritless
`1.
`
`“grouping…based on…a start base position”/“stop base position”
`
`FMI argues that this limitation is indefinite because the determination of start and stop
`
`bases depends on the aligner used and because sequences can be oriented in different directions.
`
`D.I. 26 at 10. But, as Dr. Cooper explains, regardless of the aligner used, every alignment has
`5
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 9 of 19 PageID #: 11777
`
`
`
`two ends, and a person of ordinary skill in the art would understand that the start and stop base
`
`positions refer to the two ends of the alignment.3 Such a person would have no difficulty applying
`
`the limitation to any particular sequencing read aligned to a reference sequence, like those shown
`
`in Dr. Benson’s figures. Cooper Decl. ¶ 26. FMI’s argument that the limitation is indefinite
`
`because sequence reads can have multiple orientations is akin to arguing that the start and end of
`
`a ruler is indefinite because it can be flipped end-for-end. The ruler still has a beginning (start) and
`
`end (stop).
`
`FMI also argues that there is no written description of grouping on a start and stop base
`
`position, because in some places the specification describes start or stop regions, or sequencing
`
`data at the start or stop portions of the sequence read. This disclosure is clearly sufficient to
`
`support the claims, because a position at which a given sequencing read starts or stops aligning to
`
`a reference sequence is information about both position and base identity. Cooper Decl. ¶26 It
`
`is therefore sequencing data at the start or stop portion of the read as described in the specification.
`
`However, the specification also describes “grouping” as identifying sequencing reads that are
`
`derived from the same parent molecules, and explains that this can be done “by tagging parent
`
`molecules with a sufficient number of unique identifiers (e.g., the tag count) such that there is a
`
`likelihood that two duplicate molecules, i.e., molecules having the same start and stop positions,
`
`bear different unique identifiers so that sequence reads are traceable back to particular parent
`
`
`3 FMI cites Butamax v. Gevo, 117 F. Supp 3d 632 (D. Del. 2015) for the proposition that because
`the specific start and stop base positions may vary slightly depending on the aligner used, the claim
`is indefinite. But in Butamax, infringment could not be determined because it depended on the
`particular value for the measured parameter. Id. at 638-642. The particular base chosen for a
`start or stop position does not affect the claim meaning or whether the Asserted Claims are
`infringed. All that matters is that grouping is based on one or both of those positions. Butamax
`is therefore inapposite. Erfindergemeinschaft v. Eli Lilly, 240 F.Supp 605, 633-34 (E.D. Tex.
`2017).
`
`6
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 10 of 19 PageID #: 11778
`
`
`
`molecules.” D.I. 9-4 at 27:7-17-32, 41:3-9. FMI’s claim that there is no disclosure of grouping
`
`based on start and stop positions is simply wrong.
`
`2.
`
`“ligating a set of molecular barcodes…wherein a given molecular
`barcode is a member of a set of molecular barcodes comprising 2 to
`1,000,000 different molecular barcode sequences” (’085 patent)
`
`FMI argues that “this limitation lacks written description because the specific range of “2
`
`to 1,000,000” only appears in the specification in embodiments involving 2 to 1,000,000 “unique
`
`identifiers” which are different than barcodes. D.I. 26 at 12. Once again, FMI simply has its
`
`facts wrong. The specification clearly identifies barcodes as unique identifiers, and says their
`
`number can range between 2 and 1,000,000 in some embodiments:
`
`In some embodiments, the unique identifiers comprise nucleotide barcodes. The
`disclosure also provides for a method comprising: a) providing a sample comprising
`between 100 and 100,000 haploid human genome equivalents of cfDNA
`polynucleotides; and b) tagging the polynucleotides with between 2 and 1,000,000
`unique identifiers.
`
`D.I. 9-4 at 28:14-19 (emphasis added); see also id. at 30:37-41 (referring to “unique identifiers
`
`(e.g., barcodes),” and “a unique identifier such as a barcode.”)
`
`FMI also suggests that the specification does not contemplate numbers of different
`
`barcodes as large as a million. This is simply not the case. The specification states that
`
`In some embodiments, the barcode is a polynucleotide, which may further comprise
`random sequence or a fixed or semi-random set of oligonucleotides that in
`combination with the diversity of molecules sequenced from a select region enables
`identification of unique molecules and be at least a 3, 5, 10, 15, 20 25, 30, 35, 40,
`45, or 50 mer base pairs in length.
`
`D.I. 9-4 at 3:31-37. The number of distinct barcodes of length 10 is 410 = 1,048,076, and the
`
`number of length 50 is 450 = ~1.26 x 1030. Although the specification also describes using a small
`
`number of different barcodes, it clearly contemplates much larger numbers as well.
`
`FMI next argues that “the claim does not indicate what constitutes ‘a set’ of molecular
`
`barcodes of which ‘a given molecular barcode is a member.’” D.I. 26 at 12-13. This ignores
`7
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 11 of 19 PageID #: 11779
`
`
`
`the plain language of the claim, which indicates that the claimed “set” are those barcodes used for
`
`ligating to the ends of the double-stranded cfDNA molecules as recited in step b: “(b) ligating a
`
`set of molecular barcodes to both ends of a plurality of the double-stranded cfDNA molecules….”
`
`D.I. 9-4 at claim 1. Dr. Cooper’s testimony is entirely consistent with this, and he specifically
`
`disagreed with FMI’s attempts to characterize the set as a “massive universe of alternate sequence
`
`configurations”:
`
`Q. Okay. So if I have a set of barcodes that I'm ligating to both ends of my cell-free
`DNA fragments and a given molecular barcode is a 12-nucleotide-long barcode,
`that barcode is a member of a set of 4-to-the-12th different possible molecular
`barcode sequences; correct?
`A. No, I completely disagree, because you didn't synthesis 4-to-the-12th different
`barcodes. You have a set -- you have -- the typical experiment is that you have a --
`here's a pool of barcodes that I've synthesized. The diversity within that pool is the
`thing that matters. So the fact that there is hypothetically this massive universe of
`alternate sequence configurations does not mean that I actually made those
`barcodes.
`Q. Does the claim require that a set of molecular barcodes have to be barcodes that
`are synthesized for use in that particular experiment?
`
`THE WITNESS: I think the claim requires that you're using a specific set of
`molecular barcodes, and that's what you're ligating on, and that that -- the diversity
`of that set be between 2 and 1 million different barcodes.
`
`Ex. 3[Cooper Tr.] at 317:20-319:2 (objections omitted).
`
`3.
`
`“wherein the cfDNA molecules that map to a mappable base position
`of a reference sequence are tagged with a number of different
`molecular barcodes ranging from at least 2 and fewer than a number
`of cfDNA molecules that map to the mappable base position” (’086
`patent)
`
`FMI first argues that “the cfDNA molecules that map to a mappable base position” has no
`
`antecedent basis or explanation of what these particular cfDNA molecules are or how they are
`
`determined. D.I. 26 at 13. Again, FMI ignores the plain language of the claim, which explains
`
`precisely where the cfDNA molecules come from in step (a): “(a) non-uniquely tagging a plurality
`
`of cfDNA molecules from a population of cfDNA molecules obtained from the bodily fluid
`8
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 12 of 19 PageID #: 11780
`
`
`
`sample.” D.I. 9-5 at claim 1. The cfDNA molecules that map to a mappable base position of a
`
`reference sequence are the tagged cfDNA molecules from the claimed population obtained from a
`
`sample that map to a mappable base position.
`
`FMI next argues that “nothing in the specification or the art informs what ‘a number of
`
`cfDNA molecules that map to the mappable base position’ means with respect to limiting the
`
`number of barcodes used in a particular sample.” D.I. 26 at 13. This is also wrong. The
`
`specification explains how to determine the number of cfDNA molecules that map to each
`
`mappable base position, called the read coverage:
`
`For an exemplary genome derived from cell free polynucleotide sequences, the next
`step comprises determining read coverage for each mappable base position. This
`may be performed using either reads with barcodes, or without barcodes. In cases
`without barcodes, the previous mapping steps will provide coverage of different
`base positions. Sequence reads that have sufficient mapping and quality scores may
`be counted. The number of coverage reads may be assigned a score per each
`mappable position. In cases involving barcodes, all sequences with the same
`barcode may be collapsed into one consensus read, as they are all derived from the
`sample parent molecule.
`
`D.I. 9-5 at 51:31-42. The specification also explains that there is a straightforward relationship
`
`between the input mass of DNA and the read coverage, “A sample of 100 ng of nucleic acid can
`
`contain about 30,000 human haploid genome equivalents, that is, molecules that, together, provide
`
`30,000-fold coverage of a human genome.” D.I. 9-5 at 44:33-35. Dr. Cooper’s testimony is
`
`consistent with this. Ex. 3 at 62:2-63:7.
`
`IV.
`
`THE ASSERTED CLAIMS ARE ENFORCEABLE
`
`FMI argues that the claims of the ’085 and ’086 Patents are unenforceable due to
`
`inequitable conduct on the part of Dr. Eltoukhy to avoid assignment obligations to Illumina, his
`
`former employer. This makes little sense. As FMI notes, it must show a substantial question
`
`as to whether there was a “specific intent to deceive” that is “the single most reasonable inference
`
`able to be drawn from the evidence.” D.I. 26 at 15 (citing Therasense, Inc. v. Becton, Dickinson
`9
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 13 of 19 PageID #: 11781
`
`
`
`& Co., 649 F.3d 1276, 1290 (Fed. Cir. 2011)). However, FMI’s entire argument alleges that
`
`Guardant omitted Dr. Eltoukhy as an inventor of patents in 2012 when he was employed by
`
`Illumina. The ’085 and ’086 Patents’ earliest priority claim, however, is to an application filed
`
`in 2014 more than a year later. The patent office was well aware of both this 2014 priority claim
`
`and FMI’s inventorship allegations in the 1616 case, when the ’085 and ’086 patents issued on
`
`July 7, 2020. See e.g., Ex. 4 [’085 FH]. Thus, it would make no difference whatsoever to the
`
`ownership of the ’085 and ’086 Patents if Dr. Eltoukhy was listed, and the “single most reasonable
`
`inference” is that Guardant properly did not list Dr. Eltoukhy as an inventor. Indeed, within this
`
`patent family where Dr. Eltoukhy was actually an inventor, Guardant properly listed his name.
`
`See Ex. 5[’992 Patent].
`
`In recognition of this issue, FMI turns to an infectious unenforceability argument that also
`
`fails. As an initial matter, FMI’s entire argument focuses on the ’992 Patent which is not part of
`
`Guardant’s motion for a preliminary injunction. Further, ’992 Patent listed the 2012 application
`
`on its face. As the Court found in the 1616 Case, this fact gave FMI a “plausible allegation
`
`regarding the family history and the priority relationship of the respective patents” when denying
`
`Guardant’s motion to dismiss the allegations. Case No. 17-1616-LPS-CJB, D.I. 419 at 17.
`
`While Guardant continues to dispute whether the ’992 Patent is entitled to the 2012 priority date,
`
`such a “plausible allegation” is not present here because the ’085 and ’086 Patents did not claim
`
`priority to the 2012 application when they were issued by the PTO.
`
`FMI’s arguments about Dr. Eltoukhy being an inventor to the earlier 2012 application also
`
`lacks merit. As Guardant explained in its summary judgment briefing on the issue, Dr.
`
`Eltoukhy’s alleged contributions were nothing more than prior art concepts and FMI’s theory of a
`
`10
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 14 of 19 PageID #: 11782
`
`
`
`large group of Illumina employees contributing to the invention lacks merit. Case No. 17-1616-
`
`LPS-CJB, D.I. 293; 368. Guardant incorporates that briefing here.
`
`FMI also argues that Dr. Eltoukhy’s removal of emails from his personal laptop after his
`
`deposition and supposed deletion of files in October 2019 somehow renders the patents
`
`unenforceable due to “unclean hands.” This argument also fails. First, the issue of email
`
`removal by Dr. Eltoukhy relates only to Dr. Eltoukhy’s alleged activities in 2012. As discussed
`
`above, the ’085 and ’086 Patents’ earliest priority claim is to 2014, more than a year after Dr.
`
`Eltoukhy last worked for Illumina, and the patent office was fully aware of this priority date and
`
`FMI’s allegations of invalidity for the patents in the 1616 Case when the patents issued. See e.g.,
`
`Ex. 4at 2019-10-18 Request for corrected filing receipt, 2019-10-22 Notices of concurrent
`
`proceedings. FMI has identified no motivation—and there is none—for Guardant to improperly
`
`omit Dr. Eltoukhy as an inventor to the asserted patents as it would have no bearing on the
`
`ownership of the patents.
`
`Second, even if the question of spoliation were relevant to these asserted patents (it is not),
`
`Guardant has located (and provided to FMI) two full and complete copies of Dr. Eltoukhy’s
`
`computer from March 2019, before any allegations of deletions by FMI or improper inventorship.
`
`Case No. 17-1616-LPS-CJB, D.I. 484 at 8, 11-13. Thus, no information has been lost or is
`
`otherwise unavailable to FMI that could possibly be relevant to the invention of the patents-in-
`
`suit. Further, as Dr. Eltoukhy explained and is supported by the evidence, Dr. Eltoukhy believed
`
`that all of the information removed from his computer was available to and was collected by
`
`counsel. Case No. 17-1616-LPS-CJB, D.I. 484 at 10-11, D.I. 485-1 at 789:5-15. It was only
`
`later discovered that the collection included corrupted emails. Thus, there was no bad faith on
`
`the part of Dr. Eltoukhy.
`
`11
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 15 of 19 PageID #: 11783
`
`
`
`V.
`
`FMI’S EVIDENCE SHOWS IRREPARABLE HARM TO GUARDANT AND THAT
`THE BALANCE OF HARMS FAVORS AN INJUNCTION
`
`FMI argues that Guardant’s showing of irreparable harm is general and speculative, but
`
`FMI’s own evidence supports Guardant’s position. FMI submitted the declarations of Dr. Shore
`
`and Dr. Dietrich claiming that physicians would have less confidence in relying on Guardant360®
`
`than the Accused Product for certain indications. See D.I. 35 ¶¶ 15, 17 (“As a result, patients and
`
`physicians may have less confidence in relying on Guardant360® CDx’s testing results to select
`
`patients for treatment with [rucaparib/olaparib]”); D.I. 36 ¶¶ 15 (“Thus, I and other physicians
`
`cannot rely on Guardant360® CDx's testing results to select patients for treatment with alectinib,
`
`gefitinib, erlotinib, olaparib, rucaparib, or alpelisib.”). However, there is no scientific basis for
`
`this belief. See Ex. 6 [Shore Tr.] at 18:16-21, 19:7-13 (objections omitted) (stating that he is not
`
`aware of any scientific reason why Guardant360® CDx could not be used to select patients for
`
`treatment with Lynparza® or Rubraca®). Importantly, FMI does not claim that there is. The
`
`use of FMI’s product in preference to Guardant’s based on companion diagnostic indications
`
`without any scientific basis—there is none—is precisely what is causing irreparable harm to
`
`Guardant.
`
`FMI also dubiously argues that Guardant’s claim of money damages in the 1616 litigation
`
`somehow precludes a showing of irreparable harm. D.I. 26 at 19. This is not the law. The
`
`fact that Guardant has shown that FMI has caused at least
`
` in damages in the earlier
`
`action does not mean that there is not additional irreparable harm being caused by FMI. If FMI’s
`
`position were taken to be true, then any patentee asking a Court for an injunction would need to
`
`forego all requests for monetary damages. This is obviously not the law and FMI identifies no
`
`cases supporting such a proposition.
`
`12
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 16 of 19 PageID #: 11784
`
`
`
`FMI further argues that Guardant has not established a nexus between the lost opportunities
`
`or relationships and FMI’s infringement. This is again false. While the cited reasons for
`
`choosing FMI over Guardant included FMI’s tissue biopsy and preexisting relationships, the only
`
`reason that FMI was in a position to capitalize on and take these opportunities was because it was
`
`offering a liquid biopsy product that implements Guardant’s patented method.
`
`FMI also argues that the balance of harms does not favor a preliminary injunction and
`
`claims that it would “fundamentally alter the status quo in its favor.” D.I. 26 at 22. This too is
`
`wrong. Guardant is seeking a preliminary injunction on FMI’s current FoundationOne® Liquid
`
`CDx test, not its previous tests that are the subject of the 1616 litigation. The “status quo” in this
`
`situation is the period before August 2020, when FMI received FDA-approval and began offering
`
`its FoundationOne Liquid CDx product to the market. D.I. 33 ¶13. Before that time, FMI had
`
`market share and was able to continue providing services to clinicians and biopharmaceutical
`
`companies. Nothing about the preliminary injunction Guardant is requesting would change that.
`
`
`
` FMI
`
`identifies no reason that it would need to “break” any such contract when it could offer its earlier
`
`test. For these reasons, it is FMI that has changed the “status quo” by bringing FoundationOne
`
`Liquid CDx to market, and as such, an injunction should be issued.
`
`VI.
`
`THE PUBLIC WILL NOT BE HARMED BY AN INJUNCTION
`
`Every gene for which alterations are reported by FoundationOne Liquid CDx that has an
`
`associated FDA-approved therapy, is also interrogated by Guardant’s tests. Odegaard Decl. ¶ 16.
`
`FMI’s arguments to the contrary are contradicted by its own witnesses’ testimony and other
`
`evidence. Guardant first points to the CUL4A and CYP17A1 mutations. D.I. 26 at fn. 45 But
`
`13
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 17 of 19 PageID #: 11785
`
`
`
`FMI’s Vice President of Regulatory Affairs, Dr. Mansfield, who provided FMI’s only evidence on
`
`these mutations, admitted that she did not know if these genes have any FDA-approved therapy:
`
`Q. You identify several genes that Guardant does not test for in any of its assays
`but FoundationOne Liquid CDx does test for which have significant known clinical
`outcomes, right?
`A. Yes.
`Q. You don't say that those genes have any corresponding FDA-approved therapy,
`right?
`A. No, not in my deposition, I don't say that – in my declaration.
`Q. And they don't have any corresponding FDA-approved therapy, right?
`A. I would have to check the list.
`Ex. 7 [MansfieldTr.]at 6:14-25. She also admitted she did not know if any physicians use them
`
`for any prognostic purposes, and that FMI does not include such information in its CDx report.
`
`Id. at 8:5-9:24 Dr. Mansfield also identified a few other mutations, but FMI does not claim that
`
`they have any clinical significance. D.I. 26 at 24 n. 48.
`
`FMI next argues that “[r]emoving access to F1LCDx [] would deprive prescribers of the
`
`only comprehensive liquid biopsy option for reporting ATM alterations and the only approved
`
`comprehensive CDx for Lynparza® [olaparib] in prostate cancer.” D.I. 26 at 24-24. This is not
`
`true. Guardant does report ATM alterations (including somatic alterations) and their associated
`
`FDA-approved therapy olaparib [Lynparza] to physicians who order Guardant360 CDx for their
`
`patients. Odegaard Decl. ¶¶ 8-13, Ex. A (example report).
`
`In fact, every gene for which alterations are reported by FoundationOne Liquid CDx and
`
`has an associated FDA-approved therapy, is also interrogated by Guardant360® CDx or
`
`Guardant360® LDT, and most are covered by both. Odegaard Decl. ¶ 16.
`
`There is no medical or scientific reason why physicians cannot rely on the genomic
`
`information reported by Guardant360 CDx and/or LDT to treat patients, regardless of whether
`
`there is a companion diagnostic indication from the FDA for the prescribed treatment. Odegaard
`14
`
`

`

`Case 1:20-cv-01580-LPS Document 54 Filed 04/21/21 Page 18 of 19 PageID #: 11786
`
`
`
`Decl. ¶ 20. Guardant360 was used by more than 7,000 oncologists to select the best treatment
`
`for more than 100,000 patients before it received FDA approval. Odegaard Decl. ¶¶ 18, 20.
`
`FMI’s own declarant Dr. Shore admitted that there is no scientific reason why the Guardant360
`
`CDx report would be insufficient to select patients for treatment with Lynparza:
`
`Q. So just to be clear then, you're not aware of any scientific reason why the
`Guardant360 CDx would be insufficient to select patients for treatment with
`Linparza [sic], right?
`A. Correct.
`
`Ex. 6 at 18:16-21 (objection omitted).
`
`Oncologists regularly use both of these Guardant tests to diagnose and treat ca