`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`REGENERON PHARMACEUTICALS, INC.
`Petitioner,
`v.
`
`NOVARTIS PHARMA AG,
`NOVARTIS TECHNOLOGY LLC,
`NOVARTIS PHARMACEUTICALS CORPORATION,
`Patent Owner
`
`
`Case No. IPR2021-00816
`U.S. Patent No. 9,220,631
`
`
`
`DECLARATION OF JOEL M. COHEN
`
`
`
`
`
`
`
`
`Regeneron Exhibit 1108.001
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`TABLE OF CONTENTS
`
`Page
`I.
`INTRODUCTION ........................................................................................... 1
`QUALIFICATIONS AND COMPENSATION .............................................. 1
`II.
`PERSON OF ORDINARY SKILL IN THE ART .......................................... 5
`III.
`IV. BACKGROUND OF THE TECHNOLOGY .................................................. 6
`V.
`PRIOR ART ..................................................................................................... 7
`A.
`“Sigg” – WO 2011/006877 ................................................................... 7
`
`B.
`
`C.
`
`“Lam” – International Pat. Appl. Pub. No. WO 2008/077155 ............. 7
`
`“Boulange” – International Pat. Appl. Pub. No. WO
`
`2009/030976 .......................................................................................... 7
`
`VI. OVERVIEW OF PARYLENE C .................................................................... 8
`VII. A TOXICOLOGIST WOULD NOT HAVE BEEN DETERRED
`FROM USING PARYLENE C AS A STOPPER COATING IN A
`PRE-FILLED SYRINGE COMPRISING A VEGF ANTAGONIST .......... 11
`A.
`Protein Adsorption Would Not Deter One From Using
`
`Parylene C ................................................................................. 12
`
`B.
`
`Purported Cytotoxic Characteristics Would Not
`
`Discourage a Toxicologist from Supporting Use of
`
`Parylene C in a PFS .................................................................. 16
`
`VIII. DECLARATION ........................................................................................... 24
`
`
`
`i
`
`
`
`
`
`Regeneron Exhibit 1108.002
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`I.
`
`INTRODUCTION
`1.
`I have been retained by Petitioner Regeneron Pharmaceuticals, Inc.
`
`(“Petitioner” or “Regeneron”), as an independent expert witness in the above-
`
`captioned inter partes review (“IPR”), in which Regeneron has requested that the
`
`U.S. Patent and Trademark Office cancel as unpatentable all claims of U.S. Patent
`
`No. 9,220,631 (“the ’631 patent”).
`
`2.
`
`This declaration sets forth my analyses and opinions in response to the
`
`declaration of Dr. John E. Dillberger (Ex. 2202). As I explain below, it is my
`
`opinion that a toxicologist, as a member of a product development team, would not
`
`have considered Parylene C to be unsafe, toxic, or unacceptable to be used as a
`
`stopper coating in a prefilled syringe for intravitreal injection of a VEGF
`
`antagonist drug.
`
`II. QUALIFICATIONS AND COMPENSATION
`3.
`I have a Sc.D. in Environmental Health from Harvard School of
`
`Public Health, a B.A. in Anthropology, Environmental Science, and Public Health
`
`from Tufts University, and am a certified Diplomate of the American Board of
`
`Toxicology (DABT).
`
`4.
`
`In addition, I have worked in the field of toxicology since 2011,
`
`including 8 years working as a toxicology consultant at Gradient which provides
`
`1
`
`
`
`
`
`Regeneron Exhibit 1108.003
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`scientific consulting services specializing in toxicology, epidemiology, risk
`
`assessment, and product safety, among others. In 2021 I was promoted to
`
`Principal Scientist, and in 2022 I was promoted to Principal at the firm. My
`
`professional experience in toxicology includes work in the pharmaceutical,
`
`medical device, and product safety industries.
`
`5.
`
`As a Principal at Gradient, I provide scientific consulting services
`
`related to medical device biocompatibility and toxicological risk assessment,
`
`consumer product risk assessment, and evaluations of toxicology studies in
`
`connection with human health risks. I have worked on pharmaceuticals, medical
`
`devices, consumer product safety, and toxicological risk assessments.
`
`6. With respect to the pharmaceutical industry, I have evaluated the
`
`potential health risks from exposure to impurities in a drug product for juvenile
`
`patients. To determine the health risks, I evaluated the repeated dose toxicity,
`
`genotoxicity, and carcinogenicity of the components contained in the identified
`
`impurities, including those that had little toxicological data. I have also conducted
`
`human health hazard and risk assessments for potential impurities in cell-based
`
`cancer treatments. The evaluation was used as manufacturing guidance, as well as
`
`for regulatory applications. I have also evaluated the kinetics and biological effects
`
`of a new chemical platform for drug development. Specifically, I worked on
`
`understanding the potential off-target effects of the novel chemistry involved,
`
`2
`
`
`
`Regeneron Exhibit 1108.004
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`including known pharmacokinetics and potential for adverse health effects. Most
`
`recently I was hired to conduct a toxicological risk assessment on extractable and
`
`leachables from a pre-filled syringe, which involved chemical risk assessment for
`
`all compounds detected under aggressive extraction conditions (e.g. harsh polar
`
`solvents and at elevated temperatures), as well as chemicals detected having
`
`leached from the syringe into the drug formulation itself under more clinically
`
`relevant test conditions.
`
`7.
`
`I have also worked on a number of medical device projects. I have
`
`provided ISO-compliant toxicological risk assessments for chemicals identified in
`
`extracts of implantable medical ports and catheter systems, a permanent implant
`
`screw, dialysis equipment, and leached compounds from a dialysis machine. In
`
`particular, I identified toxicological data for relevant endpoints and used the data to
`
`derive chemical and device specific safety margins, in accordance with ISO10993-
`
`17, ICH M7, and US FDA guidance. The risk assessments were used to support
`
`safety evaluations of the medical devices, and specifically for addressing potential
`
`risks related to systemic toxicity, genetic toxicity, carcinogenicity, and
`
`reproductive and developmental toxicity.
`
`8.
`
`I have also established biocompatibility test plans involving
`
`experimental testing on medical devices for cytotoxicity, sensitization, irritation,
`
`genotoxicity, implantation, hemocompatibility, material mediated pyrogenicity,
`
`3
`
`
`
`Regeneron Exhibit 1108.005
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`and systemic toxicity, all in accordance with the ISO 10993 series of international
`
`standards and guidelines. I have then interpreted the results of experimental
`
`biocompatibility testing, synthesizing the information into an overall conclusion
`
`regarding biological safety of the final finished device.
`
`9.
`
`In addition to my professional experience at Gradient, I have also
`
`published over 30 papers, including on risk assessments of extractables and
`
`leachables in medical devices, predictive toxicology, and nanotoxicology.
`
`10.
`
`In addition to my professional experience, I have years of experience
`
`participating in professional organizations and scientific panels relating to
`
`toxicology. For example, I am the current President of the Northeast regional
`
`chapter of the Society of Toxicology. I have also served as Secretary of the
`
`Society of Toxicology Computational Toxicology Specialty Section, and Councilor
`
`of the Nanoscience and Advanced Materials Toxicology Specialty Section. In
`
`2019-2020, I served on United States Environmental Protection Agency (“US
`
`EPA”) Scientific Advisory Board Peer Review Panel, applying my expertise in
`
`physiologically based pharmacokinetic modelling to evaluate the US EPA All
`
`Ages Lead Model. My findings and those of my fellow panel members were
`
`summarized in a publicly available report submitted to the US EPA Administrator.
`
`11.
`
`In addition to the above, I have given numerous presentations at
`
`symposiums, conferences, workshops, and other professional organizational
`
`4
`
`
`
`Regeneron Exhibit 1108.006
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`meetings, including many presentations on toxicology, toxicological risk
`
`assessments, and biocompatibility for medical devices and pharmaceutical
`
`components.
`
`12. Through my professional experience, I have gained extensive
`
`expertise in toxicological risk assessments in medical devices, including health
`
`risk, biocompatibility, and analysis of substances with respect to leachables and
`
`extractables.
`
`13. Gradient is being compensated at my standard rate of $430/hour. My
`
`compensation is in no way contingent upon my opinions or the outcome of the
`
`proceeding.
`
`14. My curriculum vitae is attached as Attachment A, and provides
`
`further information about my experience, expertise, and presentations.
`
`III. PERSON OF ORDINARY SKILL IN THE ART
`15.
`I have been informed that Novartis has offered the following
`
`definition of a person of ordinary skill in the art (POSA):
`
`A POSA would have an advanced degree (i.e., an M.S., a Ph.D., or
`equivalent) in mechanical engineering, biomedical engineering,
`materials science, chemistry, chemical engineering, or a related field,
`and at least 2–3 years of professional experience, including in the
`design of a PFS and/or the development of ophthalmologic drug
`products or drug delivery devices. Such a person would have been a
`member of a product development team and would have drawn upon
`5
`
`
`
`Regeneron Exhibit 1108.007
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`not only his or her own skills, but also the specialized skills of team
`members in complementary fields including ophthalmology,
`microbiology and toxicology.
`
`16. Under the definition of a POSA set forth by Novartis, I am qualified
`
`to offer opinions from the perspective of a member of a product development team
`
`with specialized skill of toxicology. In particular, I have provided toxicological
`
`assessments of medical devices and drug products that were utilized by product
`
`development teams to support safety evaluations and/or regulatory applications.
`
`17.
`
`I understand that the earliest claimed priority date for the ’631 patent
`
`is July 3, 2012, and my opinions offered herein are from the perspective of a
`
`member of a product development team with specialized skills in toxicology as of
`
`that date. I note that my opinions would be the same if offered from the
`
`perspective of a member of a product development team with specialized skills in
`
`toxicology as of October 2011, which I understand is the date by which Novartis
`
`has asserted that the named inventors conceived of the claims in the '631 patent.
`
`IV. BACKGROUND OF THE TECHNOLOGY
`18.
`I have reviewed the ’631 patent and its claims. I am not an expert in
`
`syringe design or sterilization techniques. I am, however, generally familiar with
`
`syringes and am able to apply my specialized skill of toxicology and review of the
`
`documents cited to syringes.
`
`6
`
`
`
`Regeneron Exhibit 1108.008
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`19.
`
`I am also currently conducting a toxicological risk assessment on
`
`extractable and leachables from a pre-filled syringe, which involved chemical risk
`
`assessment for all compounds detected under aggressive extraction conditions (e.g.
`
`harsh polar solvents and at elevated temperatures), as well as chemicals detected
`
`having leached from the syringe into the drug formulation itself under more
`
`clinically relevant test conditions.
`
`V.
`
`PRIOR ART
`A.
`“Sigg” – WO 2011/006877
`20.
`I understand from the declaration of Horst Koller that Sigg discloses a
`
`pre-filled syringe for intravitreal injection that contains the VEGF-antagonist
`
`ranibizumab (Lucentis). Ex. 1003, ¶ 123.
`
`B.
`21.
`
`“Lam” – International Pat. Appl. Pub. No. WO 2008/077155
`I understand from the declaration of Horst Koller that Lam also
`
`discloses a pre-filled syringe for intravitreal injection containing the VEGF-
`
`antagonist ranibizumab (Lucentis). Ex. 1003, ¶¶ 130-131.
`
`C.
`22.
`
`“Boulange” – International Pat. Appl. Pub. No. WO 2009/030976
`I understand from the declaration of Horst Koller that Boulange
`
`discloses a pre-filled syringe. Ex. 1003, ¶ 140. I further understand from Mr.
`
`Koller’s declaration that Boulange discloses an embodiment in which the stopper
`
`of the pre-filled syringe includes a Parylene C coating. Id. at ¶ 172. Boulange
`
`describes that Parylene materials “have various properties, for example
`7
`
`
`
`Regeneron Exhibit 1108.009
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`imperviousness to gases, for example oxygen, and to dry-lubricating liquids, for
`
`example water, which make them particularly attractive for use in numerous
`
`biomedical applications, particularly for certain medical devices.” Ex. 1008 at
`
`2:21-24. Boulange further describes that Parylene materials are manufactured and
`
`sold by companies such as Union Carbide Corporation and Specialty Coating
`
`Systems. Id. at 2:10-13.
`
`23. The ’631 Patent describes that the pre-filled syringe can include a
`
`stopper that “may be made from rubber, silicone or other suitable resiliently
`
`deformable material.” Ex. 1001 at 2:33-34. Based on my review of the ’631
`
`Patent, it does not appear to describe what coatings are used on the stopper. I
`
`understand from the declaration of Horst Koller, however, that it was common for
`
`stoppers to be comprised of rubber and coated with silicone oil.
`
`VI. OVERVIEW OF PARYLENE C
`24. Parylene C is a polymer material that is part of a class of materials
`
`generally referred to as Parylene, which also includes Parylene N, Parylene D, and
`
`Parylene HT. Ex. 1008 at 2:7-10; Ex. 1074.002. The chemical structure of
`
`Parylene C is illustrated below:
`
`8
`
`
`
`Regeneron Exhibit 1108.010
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`Ex. 1074.002
`
`
`
`Parylene polymers are usually applied to components by a direct dry vacuum
`
`deposition process. Ex. 1008 at 2:27-28; see also Ex. 1074.003.
`
`25. A 2007 publication from one of the manufacturers of Parylene
`
`coatings, Specialty Coating Systems (SCS), describes that Parylene C has been
`
`used in a number of different applications, including automotive, electronics,
`
`military/aerospace, and medical. Ex. 1074.010. With respect to medical devices,
`
`the publication further describes that Parylene coatings are “listed in the FDA’s
`
`Biomaterials Compendium,” and “provide an ideal surface modification for
`
`implantable and non-implantable devices such as catheters, seals, stents, cochlear
`
`implants, surgical tools, pacemakers and components. The coatings protect devices
`
`and components from moisture, biofluids and biogases and serves as a
`
`biocompatible surface for tissue contact.” Id. Another prior art publication from
`
`SCS dated August 2002 describes that Parylene coatings, including Parylene C,
`
`have been used in a wide range of medical devices since the 1970s, including
`
`9
`
`
`
`Regeneron Exhibit 1108.011
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`“catheters and mandrels, stents, needles, cannulae, cardiac assist devices, [and]
`
`prosthetics….” Ex. 1075.002.
`
`26. The prior art further describes that Parylene C has “been tested
`
`according to the Biological Evaluations requirements of ISO 10993.” Ex.
`
`1074.010. A toxicologist would have understood that ISO 10993 is a standard
`
`established by the International Organization of Standardization (ISO) that sets
`
`forth tests and evaluations for the Biological Evaluation of Medical Devices. See
`
`e.g., Ex. 2042. Companies use the testing and evaluation protocols set forth in ISO
`
`10993 as a means of demonstrating that a particular material is safe for use with
`
`the human body. With respect to each medical device application that Parylene C
`
`had been used in, a toxicologist would expect that Parylene C and all other
`
`materials would have been evaluated via testing on the final finished device (e.g.,
`
`in accordance with ISO 10993) given that thermal, mechanical, and chemical
`
`processes used to manufacture, assemble, package and sterilize products can
`
`influence the quantity and type of chemicals that could be released from the
`
`product. Accordingly, a toxicologist would have expected that Parylene C had met
`
`the ISO 10993 standard or was otherwise determined to be safe for human use with
`
`respect to the medical device applications.
`
`27. Additional prior art publications confirm that Parylene C had been
`
`used in medical device applications. For example, a 2007 publication cited by Dr.
`
`10
`
`
`
`Regeneron Exhibit 1108.012
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`Dillberger (Chang et al., 2007) states that Parylene C is “extensively used as a
`
`coating for insulating implantable devices.” Ex. 2030.002. Chang et al. further
`
`describes that Parylene C was known to be “chemically inert and
`
`nonbiodegradable,” “highly resistant to most chemicals, as well as to fungal and
`
`bacterial growth,” and that “new implementations of Parylene-C will likely lead to
`
`new technologies and devices for biomedical applications.” Id.; see also id at .008.
`
`Similarly, a 2009 publication cited by Dr. Dillberger (Kaminska et al., 2009)
`
`describes Parylene C as “a promising candidate for metallic implant coatings
`
`separating an implant body from the surrounding tissues,” and concludes that
`
`“parylene C is a material worth considering for biomedical use.” Ex. 2031.001-
`
`.002. And as explained above, Boulange describes using Parylene C as a coating
`
`for a stopper in a pre-filled syringe, and explains that it is “particularly attractive
`
`for use in numerous biomedical applications, particularly for certain medical
`
`devices.” Ex. 1008.004.
`
`VII. A TOXICOLOGIST WOULD NOT HAVE BEEN DETERRED FROM
`USING PARYLENE C AS A STOPPER COATING IN A PRE-
`FILLED SYRINGE COMPRISING A VEGF ANTAGONIST
`28. Dr. Dillberger identifies two purported issues that he contends would
`
`have discouraged a toxicologist from using Parylene C as a coating in a PFS
`
`comprising a protein-based drug. First, Dr. Dillberger contends that “high protein
`
`adsorption” would have discouraged a toxicologist from using it. Ex. 2202.0038.
`
`11
`
`
`
`Regeneron Exhibit 1108.013
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`Second, Dr. Dillberger contends that “Parylene C was known to potentially
`
`generate harmful leachables.” Ex. 2202.0041. I address both of those issues
`
`below.
`
`A. Protein Adsorption Would Not Deter One From Using
`Parylene C
`I disagree that the evidence cited by Dr. Dillberger (Chang et al., 2007
`
`29.
`
`[Ex 2030] and Kaminska et al., 2009 [Ex 2031]) demonstrates that “high protein
`
`adsorption” would have discouraged a toxicologist from using Parylene C. Chang
`
`et al., for example, describe that the application of Parylene C can be “easily
`
`modified” for different biomedical applications depending on the desired result.
`
`Ex. 2030.008. In particular, Chang et al. disclose that different surface treatments
`
`on Parylene C membranes (treatment with air plasma, and also coating with
`
`fibronectin) lead to differences in the level of protein adsorption and cell adhesion.
`
`Id. at .004, .006 (“The ability to modify the level of protein adsorption on the
`
`parylene-C substrates is of potential value for various biomedical applications and
`
`microfabrication techniques.”). Similarly, the findings in Kaminska et al. indicate
`
`that the “parylene coating very well reproduces the surface texture of substrate
`
`material and is an important factor improving the platelet adhesion.” Ex. 2031.006.
`
`A toxicologist would understand from these references that Parylene C can be
`
`combined with various substrates in a manner that enhances or diminishes the level
`
`of protein adsorption. In other words, one would understand from these references
`12
`
`
`
`Regeneron Exhibit 1108.014
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`that Parylene C can be applied in a manner such that protein adsorption would be
`
`minimized.
`
`30.
`
`In addition, although Dr. Dillberger contends that Chang et al. and
`
`Kaminska et al. report that Parylene C’s protein adsorption is “high,” Dr.
`
`Dillberger does not address the context of that finding. Ex. 2202 ¶ 66. In
`
`particular, Chang et al. describes that Parylene C can adsorb the protein bovine
`
`serum albumin (BSA), a serum albumin protein collected from cows often used in
`
`laboratory experiments and that has similar function and physicochemical
`
`properties to human serum albumin, at a rate three times higher than glass. Ex.
`
`2030.005. Dr. Dillberger, however, has not explained why comparing Parylene
`
`C’s protein adsorption to glass would be relevant in the context of evaluating the
`
`performance and function of a stopper coating for a pre-filled syringe. Instead, a
`
`toxicologist would evaluate Parylene C’s level of protein adsorption relative to
`
`other lubricants that were typically and historically used in a prefilled syringe
`
`system.
`
`31.
`
`I understand from Mr. Koller that a syringe designer would have
`
`known that silicone oil was a known stopper coating for a PFS, and that poly-
`
`(dimethylsiloxane) (“PDMS”) is composed of the same material as silicone oil.
`
`Ex. 1003, ¶ 55. Chang et al., in turn, report a comparison between glass, Parylene
`
`C and PDMS with respect to protein adsorption of BSA. As shown below, protein
`
`13
`
`
`
`Regeneron Exhibit 1108.015
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`adsorption level of BSA on Parylene C is similar to adsorption on PDMS, both of
`
`which exceed BSA adsorption levels observed for glass.
`
`Ex. 2030 at Fig. 1A (annotated)
`
`
`
`32. Chang et al. also describe that protein adsorption is protein-
`
`dependent. For example, Chang et al. also tested IgG with respect to protein
`
`adsorption on glass, PDMS, and Parylene C. As shown below, Chang et al. show
`
`that protein absorption for IgG is similar for Parylene C, glass, and PDMS. Id. at
`
`.005. Thus, a toxicologist would discern from Chang that a VEGF-antagonist has
`
`potential to adsorb to Parylene C at approximately the same rate as other stopper
`
`coatings known in the art (e.g., silicone oil or PDMS).
`
`14
`
`
`
`Regeneron Exhibit 1108.016
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`Ex. 2030 at Fig. 1B (annotated)
`
`
`
`33. Dr. Dillberger also relies on Kaminska et al., a publication examining
`
`the interactions of Parylene C on medical steel substrates, noting that Parylene C
`
`has potential for high protein adsorption. Ex. 2202.0038-39. However, the material
`
`substrate that Parylene C is applied to has a significant effect on its protein
`
`adsorption. To this point, Kaminska et al. report differences in platelet adsorption
`
`for Parylene C coated polished medical steel as compared to Parylene C coated
`
`machined medical steel, following one-hour contact of test samples with citrated
`
`whole blood. Ex. 2031.004. Specifically, Parylene C coated polished medical steel
`
`exhibited significantly lower platelet absorption compared with Parylene C coated
`
`15
`
`
`
`Regeneron Exhibit 1108.017
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`machined medical steel. Furthermore, Kaminska assessed platelet adsorption using
`
`uncoated medical steel (both polished and machined) as control samples. The
`
`study authors reported a statistically significant decrease in platelet adsorption for
`
`the Parylene C coated polished medical steel when compared to the uncoated
`
`polished medical steel control sample. Altogether, none of these findings
`
`demonstrate how Parylene C would adsorb to protein when coating a rubber
`
`stopper in a pre-filled syringe.
`
`34. Based on the results reported by Kaminska et al. and Chang et al.,
`
`protein adsorption would not discourage a toxicologist from supporting use of
`
`Parylene C in a PFS. As I explained above, these references describe that Parylene
`
`C coatings exhibit protein adsorption potential that is comparable to silicone
`
`oil/PDMS, which was a commonly used coating for stoppers in the art. Moreover,
`
`neither Kaminska et al. nor Chang et al. suggest that Parylene C would have high
`
`protein adsorption when coating a rubber stopper, which I understand would be the
`
`intended application in Boulange.
`
`B.
`
`Purported Cytotoxic Characteristics Would Not Discourage a
`Toxicologist from Supporting Use of Parylene C in a PFS
`35. A toxicologist would not expect Parylene C to have cytotoxic
`
`characteristics based on its prior use in implantable devices and interactions with
`
`biological substances in the body. As described above in section VI, Parylene C
`
`was a known biocompatible material that had been used in implantable devices for
`16
`
`
`
`Regeneron Exhibit 1108.018
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`decades. The literature as a whole describes Parylene C as biocompatible,
`
`including meeting ISO guidelines such as ISO-10993. Dr. Dillberger cites to ISO-
`
`10993, stating that a toxicologist would understand that syringe components would
`
`be evaluated for biocompatibility risks. Ex. 2202 ¶ 40. ISO-10993 provides a
`
`general framework for the biological evaluation of medical devices, including tests
`
`for in vitro cytotoxicity. Ex. 2181.004. ISO-10993-1 explains that it gives
`
`“preference to chemical constituent testing and in vitro models, in situations where
`
`these methods yield equally relevant information to that obtained from in vivo
`
`models.” Id. at .006. ISO-10993 Part 5 describes various testing methods to assess
`
`the in vitro cytotoxicity of medical devices. Thus, a toxicologist would have
`
`understood, upon reviewing the prior art as a whole (including its use in a variety
`
`of implantable medical devices), that Parylene C would not pose a toxicity risk
`
`given that it had been tested in accordance with ISO 10993.
`
`36. Dr. Dillberger incorrectly dismisses the fact that Parylene C has been
`
`tested according to the Biological Evaluation requirements of ISO-10993 simply
`
`because the statement is found in an article authored by a manufacturer of Parylene
`
`C—SCS. Id. at ¶ 71. Dr. Dillberger, however, does not consider the fact that
`
`Parylene C had been used in numerous medical applications, including implantable
`
`devices, since the 1970s. Accordingly, a toxicologist would have understood that
`
`Parylene C had not raised toxicity concerns for human implantables, and therefore
`
`17
`
`
`
`Regeneron Exhibit 1108.019
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`would not expect it to have toxicity concerns if used as a coating in a PFS storing a
`
`protein drug product.
`
`37. As noted above, a toxicologist would conduct ISO 10993 testing on
`
`the final finished syringe to confirm patient safety given that processes used during
`
`manufacturing, assembly, packaging and sterilization can impact chemical
`
`exposures from the final finished device with implications for patient safety. The
`
`need to conduct such testing to confirm patient safety would not deter a
`
`toxicologist from supporting use of Parylene C as a stopper coating in a pre-filled
`
`syringe, however, because a toxicologist would expect to conduct such testing
`
`regardless of what materials were used. As explained above, a toxicologist would
`
`have understood that Parylene C had passed ISO 10993 testing with respect to its
`
`prior uses in medical applications. And as explained below, the evidence cited by
`
`Dr. Dillberger would not give a toxicologist an expectation of a different result
`
`when using Parylene C as a stopper coating in a pre-filled syringe comprising a
`
`VEGF-antagonist.
`
`38. The two publications cited by Dr. Dillberger (Ex. 2030 and Ex. 2031)
`
`fail to support his opinion that Parylene C would have been of concern to a
`
`toxicologist for use as a coating in a pre-filled syringe that stores a protein drug
`
`product. Notably, neither publication provides evidence of in vivo cytotoxicity.
`
`Chang et al. tested Parylene C in cell culture systems, while Kaminska et al. tested
`
`18
`
`
`
`Regeneron Exhibit 1108.020
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`Parylene C for use in medical steel implants. Ex. 2030.002; Ex. 2031.002. As Dr.
`
`Dillberger states in his report, “[t]he safety, suitability, and compatibility of a
`
`material is dependent on the context in which that material is used.” Ex. 2202 ¶ 54.
`
`Here, however, neither publication identified by Dr. Dillberger relates to coating a
`
`stopper with Parylene C in a pre-filled syringe comprising a protein drug product.
`
`And to the extent that Chang et al and Kaminska et al. describe using Parylene C
`
`in contexts outside those tested in the references, both ultimately recommend its
`
`use for biological applications. Ex. 2030.008 (“Given that parylene-C has already
`
`been shown to be well suited for microfabrication, and that it can be made into
`
`flexible and robust devices, the data presented here would be useful for the
`
`implementations that tailor to the biocompatibility of parylene-C. The new
`
`implementations of parylene-C will likely lead to new technologies and devices for
`
`biomedical applications.”); Ex. 2031.006 (“The results presented strongly support
`
`the thesis that parylene C is worth considering for biomedical use.”).
`
`39. Dr. Dillberger also does not address the findings by Chang et al. (Ex.
`
`2030) indicating a lack of significant difference in cellular health (as indicated by
`
`adhesion and morphology) between PDMS and Parylene C, with PDMS being the
`
`closest appropriate comparator to Parylene C. Ex. 2030.008 (“The cell adhesion
`
`and morphology on as-deposited parylene-C and plain PDMS substrates were not
`
`significantly different (Figures 2 and 3), despite a clear difference in surface
`
`19
`
`
`
`Regeneron Exhibit 1108.021
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`roughness between the two materials (Table 2).”). This would teach a toxicologist
`
`that the characteristics of Parylene C are approximately the same as silicone oil
`
`with respect to potential cytotoxicity. And given that silicone oil was a known and
`
`accepted coating for rubber stoppers in syringes, a toxicologist would not be
`
`deterred from recommending Parylene C given that its characteristics were
`
`comparable to silicone oil.
`
`40. Moreover, Kaminska et al. (Ex. 2031) does not explicitly conclude
`
`that there is a cytotoxicity concern associated with Parylene C. Dr. Dillberger
`
`suggests that the lack of cell growth and changes in gene expression on Parylene C
`
`coated medical steel suggests generation of harmful leachable substances. Ex.
`
`2202, ¶ 68. However, the study authors make no such suggestion, and a
`
`toxicologist would not infer from the article that Parylene C would negatively
`
`impact a biologic drug product or a patient, as I explain in more detail below.
`
`41. Dr. Dillberger refers to two different assays that were reported by
`
`Kaminska et al. (Ex. 2031). The first assay measured cell proliferation and cell
`
`death, where an immortalized line of human umbilical vein endothelial cells (Ea.hy
`
`926) were cultured on a Parylene C coated medical steel surface. The control
`
`surface in this assay was a standard cell culture plate, which is optimized for cell
`
`growth and proliferation, and is thus an improper control in the context of a PFS.
`
`Instead, a toxicologist would have wanted to compare the Parylene C surface to an
`
`20
`
`
`
`Regeneron Exhibit 1108.022
`Regeneron v. Novartis
`IPR2021-00816
`
`
`
`
`
`uncoated medical steel surface to determine whether Parylene C, as opposed to
`
`some other variable, was responsible for the cell death. A toxicologist also would
`
`have needed a comparison between a Parylene C coated surface and a silicone oil
`
`coated surface to assess how Parylene C performed relative to other coatings used
`
`for stoppers (e.g., silicone oil). Because Kaminska et al. (Ex. 2031) did not
`
`provide either of these controls, a toxicologist would not conclude that Parylene C
`
`would pose a cytotoxicity concern when used as a stopper coating in a PFS
`
`comprising a protein drug product.
`
`42. Moreover, Dr. Dillberger did not consider that Kaminska et al. was
`
`ultimately inconclusive as to Parylene C’s potential cytotoxicity. In particular, the
`
`authors describe another study where cells were observed in proximity to a
`
`Parylene C coated device but not on the device surface, suggesting a lack of
`
`cytotoxicity associated with potential release or leaching of parylene C from the
`
`device surfa