UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD

Transcription

1 Filed on behalf of Petitioner Ariosa Diagnostics, Inc. by: Greg H.Gardella Reg. No. 46,045 Oblon, McClelland, Maier & Neustadt, LLP 1940 Duke Street Alexandria, VA David L. Cavanaugh Reg. No. 36,476 Wilmer Cutler Pickering Hale and Dorr LLP 1875 Pennsylvania Ave, N.W. Washington, DC Dianna L. DeVore Reg. No. 42,484 Convergent Law Group LLP 475 N. Whisman Road, Suite 400 Mountain View, CA UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD ARIOSA DIAGNOSTICS, INC., Petitioner, v. ILLUMINA, INC., Patent Owner. U.S. Patent No. 7,955,794 PETITIONER S REPLY IN SUPPORT OF THE PETITION

2 TABLE OF CONTENTS I. INTRODUCTION... 1 II. III. THE DISCLOSURE OF THE 810 APPLICATION INCORPORATED IN THE 946 PUB IS PRIOR ART... 4 THE 810 APPLICATION ANTICIPATES THE 794 PATENT CLAIMS... 5 A. The 810 Application Anticipates 794 Patent Claim Fig. 4 and Related Disclosures of the 810 Application Anticipate 794 Patent Claim Fig. 2b and Related Disclosures of the 810 Application Anticipate 794 Patent Claim Illumina s Remaining Arguments Are Unsupported and Do Not Overcome Anticipation of Claim B. The 810 Application is an Enabling Disclosure C. The 810 Application Anticipates the Other Challenged Claims The 810 Application Anticipates 794 Patent Claim The 810 Application Anticipates 794 Patent Claims 2, 5-8, and The 946 Pub Anticipates 794 Patent Claims 10, 12, 19, and IV. THE 810 APPLICATION REFLECTS THE CONTRIBUTIONS OF BOTH DR. FU AND DR. FAN AND IS SECTION 102(e) PRIOR ART V. ASSIGNOR ESTOPPEL IS A FACT-DEPENDENT EQUITABLE DEFENSE INAPPROPRIATE FOR INTER PARTES REVIEW VI. CONCLUSION i

3 I. INTRODUCTION Petitioner s Reply in Support of the Petition The Petition demonstrates that each limitation of claims 1-22 of U.S. Patent No. 7,955,794 (the 794 patent ) is disclosed in U.S. Patent Publication 2002/ (the 946 Pub ), which incorporates U.S. Provisional 60/180,810 (the 810 application ) by reference. (Paper 1 at 14.) Illumina s primary argument in response is that the Petition does not identify a single anticipatory disclosure because it cites disclosures from throughout the 810 application that are purportedly from unrelated and separate experiments. Illumina attempts to bolster this position by arguing that no single figure of the 810 application discloses each step of the claimed method on its own. These arguments are unsustainable. As an initial matter, the figures and description of generally applicable procedure in the 810 application would not be read in isolation even Dr. Fred Kramer, Illumina s expert, a well-respected and experienced molecular biologist, agreed that the figures of the 810 application and the accompanying text are substantively related. The 810 application discloses an experimental procedure for performing multiplexing assays using oligonucleotide ( oligo ) arrays, variations of which are described in figures of the application. Thus, Figs. 1a, 1b, and 3 disclose one-oligo approaches, Figs. 2a-c and 4 disclose two-oligo approaches, and Fig. 2d discloses a 1

4 three-oligo approach. (Ex at 9-14, 20, ) Phase 33, in turn, describes a scale up of these approaches to test genes simultaneously. (Id. at 35.) Figures from the 810 application are in fact incorporated and reproduced in priority applications to the 794 patent itself, demonstrating just how closely related the patented claims are to the prior art. (See, e.g., Ex (U.S. Provisional 60/297,609 (the 609 application )) at 1:2-4.) As disclosed in the Petition, when the figures are read with the related disclosures of the 810 application, the disclosure anticipates the claims of the 794 patent. Even apart from this interrelatedness, Illumina is wrong that no figure discloses each element of the claims on its own. Dr. Kramer does not contest that Fig. 4, for example, discloses every element of claim 1 but for the more than probes limitation of claim 1(b). But Dr. Kramer is wrong with respect to this supposedly missing limitation: the 810 application in fact discloses the use of the Fig. 4 approach in Phase R33 to test genes, which requires more than 100 probes. Similar anticipatory disclosure is present as well in Fig. 2b with respect to claim 1, and Fig. 2d with respect to claim 9. Illumina and Dr. Kramer are incorrect to argue that the routine step of washing necessary to perform a fundamental part of the assay process, removing unwanted elements prior to detection would be described in the 810 application (e.g., the text accompanying Fig. 4) but somehow 1 Citations to the 810 application in this Reply refer to page numbers of Ex

5 not read to apply to Fig. 2b. As described further below, Figs. 4, 2b, and 2d make clear what is established by the Petition, that the 810 application anticipates the claims of the 794 patent. The remainder of Illumina s arguments in response are based on bare attorney argument that does not withstand scrutiny. The 946 Pub is clearly prior art because it incorporates the 810 application by reference and is entitled to the priority date of that application for the disclosures therein, In re Giacomini, 612 F.3d 1380, (Fed. Cir. 2010); and, Illumina does not and cannot establish that the 794 patent has priority before the 946 Pub dated February 7, 2001, because the earliest disclosure of an element of claim 1 is in a subsequent provisional application. Likewise, Illumina s attorney arguments that various claim terms of claim 1 (such as multiplex, single-stranded, and attach) should be given narrow, self-serving constructions that the Board has already rejected cannot be sustained. Under the broadest reasonable interpretation, the 810 application discloses each of these limitations. And Illumina s argument that the 810 application is not enabled is contradicted not only by Dr. Kramer s testimony that each step would be feasible to a person of ordinary skill, but by the fact that the 794 patent and its priority application have similar levels of detail. For the reasons described in the Petition and below, Ariosa respectfully submits that the 946 Pub (and incorporated 810 application) anticipates the claims 3

6 of the 794 patent. Accordingly, claims 1-22 should be cancelled. II. THE DISCLOSURE OF THE 810 APPLICATION INCORPORATED IN THE 946 PUB IS PRIOR ART Illumina erroneously argues that the 810 application is not prior art. (Paper 31 at ) In fact, the 810 application is explicitly incorporated by reference into the 946 Pub. (See Ex [0001].) As a result, the disclosures of the 810 application are carried forward to the 946 Pub and constitute Section 102(e) prior art as of the 810 application s filing date. Giacomini, 612 F.3d at 1383 ( [A]pplicant is not entitled to a patent if another s patent discloses the same invention, which was carried forward from an earlier U.S. provisional application.... ); see also Telemac Cellular Corp. v. Topp Telecom, Inc., 247 F.3d 1316, 1329 (Fed. Cir. 2001) (document incorporated by reference... becomes effectively part of the host document ); MPEP 2127(I). Illumina s reliance on Ex Parte Yamaguchi, 88 U.S.P.Q. 2d 1606 (P.T.O. Aug. 29, 2008), Giacomini, 612 F.3d at , and Dynamic Drinkware LLC v. National Graphics, Inc., IPR , Paper 42 at 5-6 (P.T.A.B. Sept. 12, 2014) is misplaced because none of those cases involved a provisional that was incorporated by reference. And, in any event, the Giacomini and Yamaguchi panels sustained Section 102(e) rejections with reference to the dates of provisional applications. See, e.g., Giacomini, 612 F.3d at The panel need not address Illumina s argument that the 810 application is significantly different than other 4

7 disclosures of the 946 Pub (Paper 31 at 12), because the 810 application, which is incorporated by reference, has a patent-defeating effect as of [its] filing date. Giacomini, 612 F.3d at III. THE 810 APPLICATION ANTICIPATES THE 794 PATENT CLAIMS In its response, Illumina only contests anticipation of claims 1, 2, 5-10, 12, and As identified in the Petition, however, these claims are anticipated by the teachings of the 946 Pub and 810 application. (Paper 1 at ) A. The 810 Application Anticipates 794 Patent Claim 1 The Petition cites disclosures in the 810 application that recite each and every element of 794 patent claim 1. (Paper 1 at ) In response, Illumina argues that passages cited by the Petition cannot be read together to anticipate because they come from unrelated parts of the application, and further argues that no one figure, or protocol, in the 810 application anticipates on its own. (Paper 31 at ) Neither argument is correct. In fact, Illumina s argument is contrary to Figs. 4 and 2b, each of which demonstrate that the procedures of the 810 application cited in the Petition are common across the entire disclosure and 2 Modine Mfg. Co. v. ITC, 75 F.3d 1545, 1553 (Fed. Cir. 1996), which discusses whether a disclosure in an application incorporated by reference should be accorded great weight for purposes of claim construction, is also inapposite. 5

8 recite the same method as claim 1. Petitioner s Reply in Support of the Petition 1. Fig. 4 and Related Disclosures of the 810 Application Anticipate 794 Patent Claim 1 Fig. 4 and related disclosures of the 810 application recite a procedure that performs every element of 794 patent claim 1. 3 Dr. Kramer effectively concedes this with respect to all but one element of the claim. He admitted that Fig. 4 is enabling (Ex at 256:19-257:1), and contests only that the 810 application does not disclose the use in Fig. 4 of more than 100 different single stranded probes per claim 1(b). (Ex ; see also id 35; Paper 31 at ) As set forth in the Petition, the 810 application does disclose this limitation (Paper 1 at 30), a fact Dr. Kramer effectively admitted on cross-examination. The 810 application discloses examination of 50 to 100 genes at a time in the R33 phase, which discloses the use of more than 100 probes in the two oligo 3 As a threshold matter, Illumina has asserted in responding to the Petition that steps (a) and (b) of claim 1 need not be performed in the order set forth in the claim. (Paper 12 at 14; see also Ex at 83:5-25 ( if the same thing is accomplished the order of steps does not matter).) In view of this assertion, Fig. 4 of the 810 application which discloses hybridization of target mrna with a probe (claim 1(b)) before the mrna attaches to a bead (claim 1(a)) (Ex at 27) arguably becomes the clearest example that claim 1 is anticipated. 6

9 approach of Fig. 4. (Paper 1 at 30; Ex at 35.) To avoid this disclosure, Illumina responds that because 50 to 100 genes are disclosed for use in the R33 phase, they would not be used in the Fig. 4 approach. Illumina further argues that the disclosure of 50 to 100 genes would not necessarily require more than 100 probes. (Paper 31 at ) Neither contention is correct. First, the 810 application makes clear that the Fig. 4 approach is to be used to examine 50 to 100 genes at a time. The application describes that the R33 phase examines 50 to 100 genes in the same manner as in the R21 phase. (Paper 1 at 30; Ex at 35 (emphasis added).) The R21 and R33 phases differ with respect to scale, but apply to the same embodiments; phase R33 was designed to enlarge the scope of the R21 phase. (Ex at ) A person of ordinary skill would thus read the R33 phase to apply to the one oligo (Fig. 3) and two oligo (Fig. 4) approaches of the R21 phase. (Ex (Cantor Decl.) 33.) Second, Illumina s argument that the 810 application does not disclose testing of more than 100 isoforms (requiring at least one probe per isoform) (see Paper 31 at 36; Ex ), was completely undermined by Dr. Kramer. As he admitted, the focus of the R21 section of the 810 application is the detection of alternative splicing events occurring in a series of genes set forth in Table 1. (Ex at ) Dr. Kramer further conceded that alternative splicing is a natural mechanism by which more than one isoform is generated from the same 7

10 gene. (Ex at 38:13-39:4.) Dr. Kramer testified that each upstream oligo depicted in the 810 application is a probe as defined in claim 1 (Id. at 206:14-208:17), and that at least one probe is necessary to test for each isoform. (See id. at 279:18-280:2.) Thus, Dr. Kramer testified that testing 43 genes with multiple isoforms would certainly require more than 100 probes. (Id.) Dr. Kramer is correct. As explained by Dr. Cantor and Dr. Ward, a skilled artisan testing genes for multiple isoforms would necessarily use more than 100 upstream oligos with identical universal priming sites (probes) to test for the presence of each isoform. (Ex (Cantor Decl.) 36-38, 40; Ex ) That is, a two-oligo approach uses two upstream probes with the same universal priming site to test for two isoforms. (Ex at 354; Ex (Cantor Decl.) ) Thus, testing genes, each having at least two possible isoforms per gene, requires at least upstream probes with different target sequences and identical universal priming sites using the two-oligo approach. (Ex (Cantor Decl.) ) And, contrary to Dr. Kramer s suggestion (Ex ), a skilled artisan performing the 810 application method would not limit herself to testing for only one of multiple isoforms, as the purpose of the 810 application is to monitor different isoforms present in a gene. (See Ex at 32-33; Ex (Cantor Decl.) 36-38, 40.) Thus, as described in the Petition (Paper 1 at 16), disclosure of more than 100 probes is clear for the R33 phase of the grant proposal, 8

11 which describes testing of alternative splicing events (i.e., multiple isoforms), for 50 to 100 genes. (Ex at 35; Ex (Cantor Decl.) 36-38, 40.) 2. Fig. 2b and Related Disclosures of the 810 Application Anticipate 794 Patent Claim 1 Like Fig. 4, Fig. 2b describes a two-oligo approach to be performed according to the experimental procedures of the 810 application. Indeed, apart from the fact that mrna can be isolated from total RNA (i.e., by attachment to oligo(dt) beads) before probe hybridization, Fig. 2b describes an approach that is identical in all material respects to Fig. 4. (Ex at 12; Ex (Cantor Decl.) ) Thus, Fig. 2b discloses the additional element of attaching a single stranded target sequence (claim 1(a)) and then hybridizing it to a probe (claim 1(b)) in the same order as 794 patent claim 1, and anticipates claim 1 even if the 810 application must disclose an experiment with steps performed in the same order as Claim 1. (Ex at 12.) Dr. Kramer admits that Fig. 2b enables the performance of the disclosed assay to a person of ordinary skill. (Ex at 255:23-256:3.) Other than the issue of more than 100 probes, the only other element of claim 1 that Dr. Kramer alleges is missing from Fig. 2b is the step of removing the unhybridized probes recited in claim 1(c). (Ex ) 9

12 a) More than Probes (Claim 1(b)) As discussed in Section III.B.1.a, supra, the 810 application discloses the use of more than 100 probes. This disclosure applies equally to Figs. 4 (which appears in the grant application) and 2b (which appears in the invention disclosure). (See Ex at 272:25-273:7; notably, Fig. 2b uses splice junction sequences like Fig. 4.) Illumina argues that the grant application and invention disclosure portions of the 810 application are distinct and cannot be considered together (Paper 31 at 35, 46-47; Ex ), but there is no reason why they should be artificially separated in this way. In fact, the invention disclosure discloses a single common experimental strategy for multiple embodiments. (Ex at 3.) Dr. Kramer even conceded that the grant proposal was intended to be a part of the invention disclosure, and that these two portions of the 810 application are substantively related. (Ex at 116:18-117:7, 117:25-118:9.) The grant proposal includes two phases, referred to as R21 and R33, and, like the invention disclosure, describes one-oligo (Fig. 3) and multi-oligo (Fig. 4) embodiments. As identified in the Petition, the R33 phase of the 810 application refers to examination of 50 to 100 genes at a time in the same manner as in the R21 phase, i.e., in both one- and two-oligo approaches. The person of ordinary skill would read this discussion of the examination of genes in the context of the 10

13 full, substantively related, 810 application, and understand she could utilize the two oligo (Figs. 4, 2b, 2c) or three oligo approaches (Fig. 2d) in performing the R33 phase. (Ex (Cantor Decl.) 34.) Thus a person of skill in the art would understand the examination of 50 to 100 genes at a time as disclosed for Fig. 4 of the 810 application would be equally applicable to Figs. 2b and 2d, although they appear in different sections of the disclosure. b) Washing (Claim 1(c)) The 810 application discloses washing in multiple places, including referring to extensive washing with respect to the experimental procedures of the grant application (including Fig. 4 (two-oligo approach)), and with respect to poly(a) selection in both Fig. 4 and Part I of the invention disclosure section. (Ex at 5, 24, 27.) Dr. Kramer does not dispute that washing is disclosed, or that it is disclosed for a two-oligo approach. (Ex at 173:24-176:1.) Rather, Illumina argues that the washing step identified in the Petition and disclosed in the procedure sections would not be read to apply to the approach of Fig. 2b because it is in a different section. (Paper 31 at 42-43, ) This is incorrect. It would be clear to a person of skill in the art that washing is a common step that is equally applicable to all disclosed embodiments. The language of the 810 application makes clear that it discloses a common experimental approach to be applied to multiple embodiments, not siloed, unrelated protocols. Thus, 11

14 description of the experimental procedure to effectuate the claimed method in two parts of the invention disclosure section (Parts I and II), are applicable to both a single-oligo approach (Figs. 1a-b) and an alternative multiple-oligo approach (Figs. 2a-d). (Ex at 5-7.) By Illumina s own admission, Figs. 1a-2d, which are reproduced exactly in the 609 provisional as Figs. 1-6, are all preferred embodiments of the invention, (Ex at p. 3 l. 4) a fact Dr. Kramer was unaware of when he opined that these same figures were unrelated in the 810 application. (Ex at 226:23-227:11, 227:25-228:10.) Accordingly, extensive washing for a two-oligo approach identified in the Petition, applies equally to multiple-oligo approaches of Figures 2c and 2d. (Paper 1 at 17.) Dr. Kramer admitted at his deposition that the protocols disclosed in the 810 application have a substantive relationship to each other. (Ex at 113:5-113:12.) Part I and Fig. 4 each describe that part of the procedure involves washing, referred to as poly(a) selection or extensive washing. (Ex at 5, 24, 27.) This washing step is routine, and is done to remove excess or nonannealed oligos. (Ex (Cantor Decl.) 50-53; Ex at 161:18-162:5.) Indeed, the very fact that each figure of the 810 application relies on immobilization and attachment would indicate to a person of ordinary skill that washing to remove unwanted elements in the reaction solution is required, as the purpose of attachment or immobilization is to isolate the elements for study so that 12

15 the rest can be removed. (Ex (Cantor Decl.) 31.) Thus, in view of the commonalities across the embodiments, a person of ordinary skill in the art would read the experimental procedure of the invention disclosure and the experimental procedures of the grant application, and would apply them to all 810 application embodiments. 4 (Ex at 7, 24; Ex (Cantor Decl.) 54.) 5 Illumina attempts to justify its strained argument by suggesting that because Fig. 2b discloses ligation, it does not employ washing. (Ex at 137:7-21.) This is illogical. Ligation and washing are complementary, not exclusive: the former joins together oligos that have hybridized while the latter removes non- 4 The 946 Pub reproduces Fig. 2b (as Fig. 4), and describes washing as a preferred method of removing unhybridized probes, confirming washing would apply to the Fig. 2b protocol. (Ex at Fig. 4, [0018]; Ex (Cantor Decl.) 54.) 5 Illumina s reliance on Net MoneyIN, Inc. v. VeriSign, Inc., 545 F.3d 1359, 1370 (Fed. Cir. 2008) is misplaced. In Net MoneyIN, the court held that anticipation could not be established based on a combination of two protocols in a prior art reference that were not directly related to each other. Id. at As the Federal Circuit held in Wm. Wrigley Jr. Co. v. Cadbury Adams USA LLC, 683 F.3d 1356, 1361 (Fed. Cir. 2012), a prior art reference with multiple disclosures may anticipate where that reference envisions the combination of the claim elements. 13

16 hybridized oligos. Indeed, Fig. 4 in the grant portion of the 810 application discloses a protocol that includes both washing and ligation. (Ex at 27; Ex (Cantor Decl.) 28-30; Ex at 73 (claim with wash and ligation steps).) A core purpose of the 810 application is detecting amplified probes, which is done by removing non-hybridized primers. (Ex at 7; Ex (Cantor Decl.) 31.) A person of ordinary skill would thus understand that a multiple-oligo approach includes the routine washing step used to remove non-hybridized probes, regardless of whether or not it includes ligation. (Ex (Cantor Decl.) 31; see also Ex at 159:19-25 (washing can be used in Fig. 2b assay).) 3. Illumina s Remaining Arguments Are Unsupported and Do Not Overcome Anticipation of Claim 1 Illumina s response includes attorney argument to the effect that other limitations of claim 1 are supposedly not disclosed by the 810 application. None of these arguments has any merit. a) Multiplex Method (Preamble) Illumina argues that claim 1 requires a single multiplex reaction. (Paper 31 at ) Even if this were correct, 6 the 810 application discloses such a reaction, describing examination of genes at a time. (Paper 1 at 15; Ex at 6 Dr. Kramer explained multiplex [can be] used when there are multiple reactions, the sum of which look at many different targets. (Ex at 45:25-46:9.) 14

17 35 (emphasis added).) The fact that this refers to a single reaction is confirmed by the description that [m]any different oligos... can be pooled for specific applications. (Ex at 5, 20 (emphasis added); Ex (Cantor Decl.) 55.) Moreover, the 810 application refers to examination of zip-code sequences that allow for multiplex detection in a single large array. (Ex at 6, 11-15, 27; Ex (Cantor Decl.) 55.) Thus, the 810 application states that [u]sing this zipcode system, large numbers of alternative splicing events can be simultaneously detected with pooled oligos... on a universal array. (Ex at 15 (emphasis added); see also Ex at 94:10-15; Ex (Cantor Decl.) 55.) b) Single-Stranded Target Sequence Attached to a Solid Surface (Claim 1a) The 810 application discloses attachment of single-stranded target sequences to a solid surface through a oligo(dt) to poly(a) hybridization. (Paper 1 at ) Each of Illumina s arguments to the contrary is premised on an unduly narrow claim construction that the Board has already rejected. (See Paper 12 at 12-13; Paper 14 at 6.) First, Illumina argues that a target sequence may only be single stranded if it (and any connected part of it) has not been hybridized in any way. (Paper 31 at ) The Board correctly declined to adopt such a construction and instead found that the term single stranded should be given its broadest reasonable construction. (Paper 14 at 6.) Under this construction, the 810 application 15

18 discloses a single-stranded target sequence. (Paper 1 at ) Illumina itself cites a passage of the 794 patent in its Patent Owner Preliminary Response that [e]ven after the hybridization between the probes and the target sequences, the target sequences can remain single-stranded, at least partially. (Paper 12 at 14 (citing Ex at 13:58-14:6) (emphasis added).) Consistent with this, in the R21 phase of the 810 application cited in the Petition, where probes are hybridized to target sequences before attachment to the oligo(dt) bead, single-stranded portions still exist on the mrna. (Paper 1 at 15-17; Ex at 27, Fig. 4; Ex (Cantor Decl.) ) And, likewise, in Fig. 2b of the 810 application, where probes can first be attached to the bead through hybridization of their poly(a) tail, the separate mrna target sequence is entirely single stranded at the time of attachment, and remains partially single stranded after hybridization. (Ex at 12; Ex (Cantor Decl.) 57.) Second, Illumina argues that the hybridization between the mrna poly(a) tail and oligo(dt) bead disclosed in the 810 application is insufficient to disclose the claimed attachment. This argument is premised on a claim construction of attachment that requires sufficient affinity to withstand denaturation.... (Paper 31 at 23.) In its Decision on Institution, the Board declined to adopt this narrow construction and found instead that the term attachment should be given its broadest reasonable construction. (Paper 14 at 6.) This was entirely correct. 16

19 The claim broadly recites attach[ment] and does not distinguish between covalent or permanent binding, on the one hand, and non-covalent binding (such as hybridization), on the other. Thus, Dr. Kramer conceded that the claim itself does not require any specific degree of attachment. (See, e.g., Ex at 303:18-25 (claim 1 does not require a covalent bond).) Though Illumina argues that attach must mean something different than immobilize (claim 1(f)) (Paper 31 at 24), it offers no construction for immobilize and no reason why attach would require the specific affinity limitation it proposes. Indeed, the 794 patent specification describes a number of embodiments that attach target sequences without the affinity that Illumina would read into the claim. (See Ex at 15:48-65; Ex (Cantor Decl.) 58.) 7 Under the broadest reasonable construction, hybridization of the poly(a) tail to the oligo(dt) bead attaches the mrna to the bead. (Ex (Cantor Decl.) 59.) This is evident from, inter alia, the fact that the 810 application describes washing that would not detach the mrna from the bead, consistent with the 7 The 794 patent incorporates by reference U.S. App. No. 09/931,285 (U.S. Pat. No. 6,931,884), which teaches affinity capture utilizing hybridization can be used to attach nucleic acids to surface or bead and refers to hybridization with Poly(A) as an example. (Ex. 1047, 884 Patent at 10:33-38 (emphasis added).) 17

20 passages of the 794 patent cited by Illumina. (See Paper 31 at 23 (citing Ex at 21:16-28, 14:11-16, 17:36-38, 68:64; Ex at 27, Fig. 4; Ex (Cantor Decl.) 59; see also Ex at 96:9-16; Ex at 72:21-73:13.) Indeed, Dr. Kramer conceded that a skilled artisan would understand that hybridization was one way to attach the probe to the solid support and that such attachment would be sufficient to withstand the washing recited in the claim. (Ex at 73:25-76:6; Ex (Cantor Decl.) 60.) c) Each of Said... Probes Has Identical Universal Priming Sites (Claim 1b) Illumina argues that the 810 application does not disclose probes with universal priming sites that are identical as required by claim 1. But Dr. Kramer admitted that the upstream oligos disclosed in the 810 application meet the definition of probe in claim 1 of the 810 application. (Ex at 206:14-208:23.) This is correct. A skilled artisan would understand that the method of the 810 application would employ upstream oligos with identical universal priming sites. (Ex (Cantor Decl.) ) This is consistent with the disclosure of the 810 application that universal priming sites are used to avoid biased signal amplification in PCR. (Ex at 6; Paper 1 at 16-17; Ex (Cantor Decl.) 61-62; Ex at 111:19-21.) Illumina also argues that the 810 application does not show as complete details as is contained in the... claim, because the 810 application only refers to 18

21 universal priming sites in generalized terms. (Paper 31 at 39.) This argument is unavailing. The 810 application discloses the use of a universal site throughout as part of its experimental strategy (see, e.g., Ex at 6-7, 21, 23, 27, Figs. 2a, 2b, 2d, 4), and as Drs. Ward and Kramer testified, the selection of a universal primer site would be well known by a person of ordinary skill in the art. (See Ex at 53:13-24; Ex at 74:5-23.) Use of identical universal priming sites therefore applies to every protocol of the 810 application. d) Claim 1(d)-1(g) Illumina does not dispute that the 810 application discloses the limitations of claim 1(d)-1(g). Each is disclosed in the 810 application. (Paper 1 at 17-19; Ex (Cantor Decl.) 63; Ex at 71:20-72:1, 72:3-13.) B. The 810 Application is an Enabling Disclosure Illumina argues that the 810 application is not enabling because it does not contain sufficient experimental details. (Paper 31 at 40.) In fact, the 810 application discloses each step of the claims and discusses a common set of experimental procedures that can be used to apply those steps. (Ex at 24, 27; see also supra Section III.) As noted above, Dr. Kramer admitted that each step in the experimental approach (e.g., of Figs. 2b, 2d, and 4) of the 810 application was known, and only required routine skill for the person of ordinary skill in 2000, and therefore each experimental scheme shown in the figures of the 810 application 19

22 was feasible to the skilled artisan in (Ex at 51:4-12, 70:23-74:4, 134:12-136:9, 159:19-25, 254:22-257:1; Ex (Cantor Decl.) ) Ilumina s argument that the 810 application is not an enabling disclosure is further undercut by the 794 patent and the 609 provisional, neither of which describes every step of the claims in detail or establishes that every step was in fact carried out. (Ex. 1001; Ex. 1038; Ex at 260:17-25, 263:14-264:20.) Illumina s assertion that Dr. Ward confirmed that one of ordinary skill would not be able to perform the experiment described in the R33 phase is wholly unsupported. In the passages Illumina cites, Dr. Ward (a) stated that the R33 procedure could be based on either a one- or two-oligo variation, but that no additional information would be needed because the two prior variations are self explanatory (Ex at 51:20-52:13), (b) that a specific universal priming site is not identified, but that it can be selected among ordinary options as long as it s different from the tag (id. at 53:8-23), and (c) that the experiment does not identify an exact number of probes to be used. (Id. at 116:25-117:14.) 8 The 810 application is enabled for a person of ordinary skill in the art as defined by Ariosa (see Ex (Cantor Decl.) 23 (bachelor s degree and relevant laboratory experience)), and also by such a person with more experience. (See Ex at 70:10-21 (Master s degree or Ph.D.).) 20

23 None of these statements suggests, much less confirms that the R33 experiment is inoperable or requires non-routine experimentation. They instead clarify that there are steps of the disclosed experiment that a person of ordinary skill would know how to perform without undue experimentation. (See Ex at 75:9-15, 79:7-25, 210:2-22.) C. The 810 Application Anticipates the Other Challenged Claims 1. The 810 Application Anticipates 794 Patent Claim 9 Claim 9 of the 794 patent is anticipated by Fig. 2d of the 810 application. (Paper 1 at ) Fig. 2d differs from Fig. 4 or 2b only in that Fig. 2d depicts the use of three oligos bound to a target, and one of skill in the art would understand the target could be the mrna. For the same reasons discussed above, Fig. 2d thus discloses each element of claim 1. This disclosure of three oligos also meets the additional limitations of claim 9: 1) a plurality of other probes and 2) a second enzymatic step. (Ex (Cantor Decl.) 66.) Indeed, Dr. Kramer s declaration states that Fig. 2d discloses two enzymatic modifications. (Ex ) In response, Illumina claims that the Petition improper[ly] cites multiple sections of the 810 application. (See Paper 31 at ) As discussed, Illumina s attempt to artificially divide the 810 application is unwarranted. Fig. 2d reflects steps to be performed pursuant to the common set of experimental procedures disclosed in the application. (Ex at ) A person of ordinary skill in the 21

24 art would thus understand that a three-oligo approach would be performed according to the experimental procedures of related multi-oligo approaches (e.g., Figs. 2b and 4), and would be scaled up according to the R33 phase, including to test genes with more than 100 probes. (Ex (Cantor Decl.) ) 2. The 810 Application Anticipates 794 Patent Claims 2, 5-8, and 21 Illumina claims that the Petition does not identify a relevant disclosure for claims 2, 5-8 and 21. (Paper 31 at ) This is incorrect. First, with respect to claim 2, the disclosure of the claimed modif[ication] of said probes in the 810 application (Paper 1 at 19) is present in any procedure that uses ligation of oligos, including every one of Figs. 2a-2d and 4. (Ex (Cantor Decl.) ) Likewise, the unique adapter sequences of claims 5-8 of the 794 patent (Paper 1 at 20-21), are present in each of Fig. 4 and Fig. 2b, and they, with their associated disclosures, anticipate the claims. (Ex (Cantor Decl.) 72.) Finally, the 810 application clearly discloses the use of beads in an array as required by claim 21 as a method of detecting amplicons. (Ex. 1012, at 18-19; Ex (Cantor Decl.) 73.) While this bead array is mentioned in the context of the single oligo variation (Paper 1 at 25), one of skill in the art, reading the 810 application as a whole, would understand it to be an option for each variation in the application. (Ex (Cantor Decl.) 73.) Thus, the person of ordinary skill would understand that the zip-code array in each figure may be the bead array. 22

25 (Ex (Cantor Decl.) 73.) In this context, the disclosed bead array with either the Fig. 4 or Fig. 2b procedures anticipates claim The 946 Pub Anticipates 794 Patent Claims 10, 12, 19, and 20 As disclosed in the Petition, each element of each of claims 10, 12, 19, and 20 of the 794 patent is disclosed in the 946 Pub. (Paper 1 at 22-23, ) Illumina suggests that the 946 Pub does not anticipate these claims because it does not have a priority date earlier than claims 10, 12, 19, and 20. The 946 Pub is prior art to the 794 patent as a matter of law because it was filed by a different inventive entity (Fan and Fu) on February 7, 2001, before the earliest date to which the 794 patent could validly claim benefit under 35 U.S.C. 119(e). 9 Moreover, claims 10, 12, 19, and 20, which depend from claim 1, could at most be entitled to a priority date of August 9, 2001, the filing date of U.S. Provisional App. 60/311,271, because that application contains the first priority disclosure of more than probes recited in claim 1(b). (Ex at 17.) Illumina attempts to satisfy its burden to establish priority, see PowerOasis, Inc. v. 9 The 794 patent can only claim priority to a provisional application filed within one year of its filing date. See 35 U.S.C. 119(e). The applicant s attempt to make a belated priority claim to an intervening utility application was rejected and no petition under 37 C.F.R. 1.78(e) was filed. (See Ex ) 23

26 T-Mobile USA, Inc., 522 F.3d 1299, (Fed. Cir. 2008), by arguing, e.g., that because the Petition cites 43 of the 946 Pub in a claim chart for claim 1(b) and that paragraph is repeated in an earlier priority application, support for claim 1(b) must be present in the earlier priority application. (Paper 31 at 7-9.) This ignores 43 and the purpose for which it was included in the Petition: it discloses the probe set limitation of claim 1(b), not the more than 100 probes limitation, which is why the Petition also includes additional disclosures in the claim chart. (See Paper 1 at ) Absent this faulty logic, Illumina has done nothing to meet its burden to establish priority before the 946 Pub, and the 946 Pub is prior art. IV. THE 810 APPLICATION REFLECTS THE CONTRIBUTIONS OF BOTH DR. FU AND DR. FAN AND IS SECTION 102(e) PRIOR ART Illumina erroneously argues that the 810 application and 946 Pub are not Section 102(e) prior art because the relevant disclosures were invented by Dr. Fan and not Dr. Fu. Dr. Fu is a named inventor on the 946 Pub. (Ex at 1.) Although Dr. Fan s name is listed on the 810 application cover page, Dr. Fu s name is listed on the top corner of every page of the grant application. (Ex at ) Dr. Fu confirmed that he prepared this grant application and collaborated with Dr. Fan on the remaining invention, including the invention disclosure. (Ex at 58:12-13, 60:7-22; Ex at 3.) Indeed, statements reflecting the collaboration of Drs. Fu and Fan throughout the 810 application (in both the invention disclosure and grant 24

27 application) (Ex at 3, 15, 20-21, 28-29, 31, 38-39, 44; Ex (Cantor Decl.) 35; Ex at 116:18-117:7) confirm that they together provided a common experimental strategy to be applied to multiple embodiments. Illumina contends Dr. Fu admitted the three oligo embodiment of claim 9 was invented solely by Dr. Fan (Paper 31 at 19, 51), but this again mischaracterizes Dr. Fu s testimony. Dr. Fu testified only that Dr. Fan brought the use of three oligos to his attention (Ex at 91:16-17), not that he failed to contribute to the ultimate conception, or that Dr. Fan alone did so. Critically, Illumina has made no attempt to remove the 946 Pub as 35 U.S.C. 102(e) prior art by submitting an affidavit or declaration under 37 C.F.R establishing that the relevant disclosure is applicant s own work. MPEP (citing In re Mathews, 408 F.2d 1393 (C.C.P.A. 1969).) V. ASSIGNOR ESTOPPEL IS A FACT-DEPENDENT EQUITABLE DEFENSE INAPPROPRIATE FOR INTER PARTES REVIEW The Board rejected Illumina s attempt to invoke an assignor estoppel defense in this inter partes review. (Paper 14 at 12.) Moreover, as confirmed by the Board in a separate proceeding, assignor estoppel is not appropriate in inter partes review. See Redline Detection, IPR , Paper 31 at 4-5. VI. CONCLUSION Based on the foregoing, claims 1-22 of the ʼ794 patent should be cancelled. By: /Greg H. Gardella/ Reg. No. 46,045 25

28 CERTIFICATE OF SERVICE Pursuant to 37 C.F.R. 42.6(e), the undersigned certifies service of PETITIONER S REPLY IN SUPPORT OF THE PETITION and EXHIBITS on the counsel of record for the Patent Owner by filing this document through the Patent Review Processing System as well as delivering a copy via electronic mail to the following address: 2kst@knobbe.com 2mlf@knobbe.com Derek.Walter@weil.com Edward.Reines@weil.com Adrian.Percer@weil.com BoxIllumina@knobbe.com Dated: June 24, 2015 /Greg H. Gardella/ Greg H. Gardella (Reg. No. 46,045)