Petitioner, Celltrion, Inc., filed IPRs challenging each claim of U.S. Patent No. 8,580,264 (“the ’264 patent”) and U.S. Patent No. 10,874,677 (“the ’677 patent”) as anticipated under 35 U.S.C. § 102 and obvious under 35 U.S.C. § 103. IPR2022-00578, Paper 78 at *6, IPR2002-00579, Paper 72 at *5.

The Patent Trial and Appeal Board (“the Board”) found all claims of the challenged patents covering methods and devices for treating rheumatoid arthritis unpatentable.

The ’264 Patent


The ’264 patent is directed to a method of treating rheumatoid arthritis (RA) with an anti-IL-6 receptor (IL-6R) antibody. IPR2022-00578, Paper 78 at *4. Claims 1, 10, and 12 are independent. Claims 1 and 10 recite, in relevant part:

A method of treating rheumatoid arthritis (RA) in a patient comprising subcutaneously administering an anti-IL-6 receptor (IL-6R) antibody [tocilizumab[1]] to the patient … as a fixed [dose] of 162 mg per dose every week or every [two] weeks...

Independent claim 12 further recited, in relevant part, “a method of inhibiting the progression of structural joint damage” in an RA patient, “wherein structural joint damage at week 24 or week 48 is found to be inhibited.”

Claim Construction

The parties disputed (1) whether the term “fixed dose of 162 mg” necessitates a single injection, (2) whether the preamble of claims 1 and 10 was limiting, and (3) when joint damage “is found to be inhibited” as recited in claim 12.

First, the Board determined that the “fixed dose of 162 mg per dose” does not require the fixed amount to be delivered in a “single injection.” Id. at *12. Patent Owner argued that a construction allowing multiple injections would render the claim term “per dose” superfluous. But the Board found that the specification distinguishes “per dose” and “per injection” and that a “single injection,” while preferred in an embodiment, is not required. Id. at *14-15. Accordingly, the Board construed “a fixed dose of 162 mg per dose every week or every two weeks” to mean “a dose of 162 mg administered without regard to the patient’s weight or body surface area and administered every week or every two weeks.”

Next, the Board determined that the preamble of claims 1 and 10, while limiting, does not require the “treatment” to be safe and effective. Id. at *19-20 (citing United Therapeutics Corp. v. Liquidia Technologies, Inc., 74 F.4th. 1360 (Fed. Cir. 2023)). The Board noted that its determination was not only supported by United Therapeutics, but consistent with the “specification’s definition of ‘treatment’” which encompasses prophylactic use. IPR2022-00578, Paper 78 at *20.

Regarding claim 12, the Board found that it is “ambiguous as to when the measurement is actually made” but that the specification “equates assessing at 6 months with 24 weeks, and assessing at 1 year with 48 weeks.” Id. at *22. Accordingly, the Board determined that “wherein structural joint damage at week 24 or week 48 is found to be inhibited” encompasses a method where any degree of structural joint damage inhibition is found through actual assessment of structural joint damage at approximately week 24 (e.g., 6 months) or approximately week 48 (e.g., 1 year). IPR2022-00578, Paper 78 at *23.


Applying its claim construction, the Board found claims 1-3 and 6-11 of the ’264 patent anticipated by NCT 00965653 (NCT 653), which described a study of subcutaneously administering 162 mg tocilizumab to patients with rheumatoid arthritis either every week or every other week regardless of body weight or body surface area. Id. at *31. The Board credited Petitioner’s expert in determining that a POSA would understand that the tocilizumab was administered to the patients in NCT ’653 with the intent to treat RA and that the heavy and light chains of tocilizumab have the amino acid sequences recited in the challenged claims. Id. at *29.

The Board found claim 12 not anticipated by NCT 653 because Petitioner failed to establish why a treatment that extends for only 15 weeks, even if described as examining suitability for chronic use, “necessarily discloses the step of finding structural joint damage inhibition at week 24 or week 48.” Id. at *32-33. Importantly, the Board found that “petitioner’s argument amounts to what POSA could do, not what it necessarily would do based on the disclosure of NCT ’653,” which is insufficient to show inherent anticipation. Id. at *33.


The Board also found that claims 1-3 and 6-11 are unpatentable as having been obvious over NCT 653 in view of Morichika.

Patent Owner argued that a person of ordinary skill in the art (“POSA”) would not have reasonably expected success in arriving at the claimed invention based on the teachings of the cited references “given the many issues with highly concentrated antibody formulations, like increased viscosity, aggregation, and what kind of device would be used to administer the drug,” and because Morichika’s formulations were never actually administered to humans. Id. at *38-39.

The Board disagreed, finding that Patent Owner applied “an improperly heightened standard to its reasonable expectation of success argument.” Id. at *39. Noting that Morichika “teaches a highly concentrated antibody-containing formulation ‘suitable for subcutaneous administration[,]’” the Board credited Petitioner’s experts in determining that “Morichika and the state of the art provide ample information and direction to provide a POSA with a reasonable expectation of success in achieving the claimed invention.” Id. at *40-41, 45. Importantly, the Board found Morichika addresses the same concerns raised by Patent Owner relating to high-concentration antibody formulations, such as deamidation, aggregation, and viscosity, “and includes examples with recipes of formulations of high-concentration tocilizumab that address those concerns.” Id. at *40-42.

The Board also found claim 12 obvious over the combination of NCT 653 and Morichika and further in view of Kremer 2009, which taught that patients receiving tocilizumab had no progression of structural joint damage after 104 weeks. The Board credited Petitioner’s expert, who opined that a POSA would have understood that those patients would have had the same scores at 24 and 48 weeks. Id. at *49-50.


Expert testimony can be an important factor in navigating complex technologies. This was especially evident here, where the Board expressly credited Petitioner’s expert in reaching its findings of anticipation and obviousness.

For claims related to treatment of disease, a claim element directed to safety and efficacy of the treatment may require claim construction. When claim construction is disputed, it may be advantageous, depending on the circumstances, to develop patentability arguments based on both sides’ proposed constructions.

The ’677 patent


The ’677 patent is directed to a subcutaneous administration device for anti-IL-6R antibody. IPR2002-00579, Paper 72 at *7. Claims 1 and 5 are independent—each claim recites in relevant part:

An article of manufacture comprising a subcutaneous administration device, which contains and delivers to a patient a 162 mg fixed dose.

Claim Construction

The parties disputed whether the term “delivers to a patient” imposed any meaningful restriction on the claimed device’s structure or the intended recipient’s condition. The parties also disputed whether the claimed “subcutaneous administration device” must administer all of the fixed dose in one injection.

The Board declined to adopt Patent Owner’s construction that the patient be a person “with an IL-6-mediated disorder” and that the device “safely and effectively deliver the drug subcutaneously” to “accommodate the risk of serious immunogenic reaction.” Id. at *15. To do otherwise, the Board determined, “would impermissibly extend the scope of the claims” and “place an additional limitation on the device that is not functional, but rather relates to the nature of the individual.” Id. at *15. Accordingly, the Board interpreted the “functional language as requiring the claimed device to be capable of delivering the contained dosage to any ‘patient’ whether or not they have a specific medical condition or without respect to ancillary conditions or concerns that may apply” and construed the term to require “that the claimed device be capable of subcutaneously delivering the dose contained within it to a person.” Id. at *16.

The Board also declined to adopt Patent Owner’s construction that “subcutaneous administration device” requires a single injection. Id. at *18. Finding no requirement in the specification or claims that the dose be administered “via a single injection,” the Board construed the language “to mean that the claimed apparatus must be ‘capable of containing and delivering the 162 mg fixed dose subcutaneously,’ without specifying the method by which the dose is to be delivered (e.g., single or multiple injections) or reading any further method limitations into the device claim.” Id. at *19.


The Board determined that NCT 653 does not anticipate claims 1 and 5 of the ’677 patent because NCT 653 does not expressly or inherently disclose “that the subcutaneous administration device necessarily ‘contains and delivers’ the 162 mg fixed dose.” Id. at *28 (emphasis in original). Regarding inherency, the Board noted that although it would be “logical” for a POSA to assume that a device used to administer the dose in NCT 653 was capable of delivering the whole dose in a single administered injection, the reference does not rule out the possibility that another method could be used, e.g., two injections, each from a device capable of containing and delivering a dose only of 81 mg. Id. at *28. The Board also reasoned that even though this hypothetical route of administration is “perhaps unlikely” it is not “categorically excluded” by NCT 653.


The Board nevertheless found claims 1-8 of the ’677 patents obvious.

As with the ’264 patent, the Board first noted that claims 1-8 are not limited by “any requirement for a showing of the efficacy or safety of administering tocilizumab.” Id. at *55. Regarding the proposed combination, the Board found that “subcutaneous injection of a 162 mg fixed dose of tocilizumab had been performed in the NCT ’653 clinical trial prior to the priority date of the ’677 patent” and that such injection “self-evidently requires a device designed to contain and deliver the 162 mg fixed dose.” Id. at *60. The Board determined that it would have been obvious to a POSA to use an autoinjector pen for the treatment of RA and to modify it to contain and subcutaneously deliver the fixed dose of 162 mg tocilizumab, as taught by NCT ’653. Id. at 58. Furthermore, because the claims do not require injection to be comfortable for the patient or administered in a single injection (per its claim construction), the Board determined that a POSA would understand that a subcutaneous injection device “could deliver a high concentration of human anti-IL-6 human antibodies in a volume suitable for subcutaneous injection (approximately 1 milliliter)” and would have been motivated to use one because it is popular with patients, easy to, use and consequently increases compliance with the treatment regimen. Id. at *58-61.


Inherency is challenging to prove. Interestingly, the Board determined that NCT 653 does not inherently disclose “that the subcutaneous administration device[s] necessarily ‘contains and delivers’ the 162 mg fixed dose” but that the disclosed “[s]ubcutaneous injection self-evidently requires a device designed to contain and deliver the 162 mg fixed dose.” Id. at *60 (emphasis added). This, and the Board’s finding that certain possibilities could not be “categorically excluded” by the prior art, highlight the importance for a challenger to present alternative grounds such as obviousness when relying on inherency to prove anticipation. Using obviousness as a backup, as demonstrated here, could be the difference between winning and losing.

Case: Celltrion, Inc., v. Chugai Seiyaku Kabushiki Kaisha, No. IPR2022-00578, Paper 78 (P.T.A.B. August 29, 2023); Celltrion, Inc., v. Chugai Seiyaku Kabushiki Kaisha, No. IPR2022-00579, Paper 72 (P.T.A.B. August 29, 2023)