The central question before us is whether the "wherein" clauses here are limiting. Claim 1 of the '453 patent expressly recites two "wherein" clauses, which set out specific safety and efficacy benchmarks for the claimed formulation. However, Sandoz invites us to read these requirements out of the claim. Sandoz insists that these "wherein" clauses are mere statements of inherent results or general purpose that do not meaningfully limit the scope of the invention. For the reasons below, I agree with the majority's conclusion and most aspects of its thoughtful analysis. However, I would arrive at that conclusion by following a slightly different path.
I
As an initial matter, the posture in this case is unusual. The outcome of Sandoz's appeal turns purely on an issue of claim construction. If treated as non-limiting, Sandoz argues the asserted claims are automatically rendered invalid, based on prior rulings concluding that similar claims reciting the remaining elements are obvious. In turn, Sandoz contends that there can be no likelihood of success on the merits such that the district court's decision to grant Allergan a preliminary injunction must be reversed. Allergan agrees that Sandoz's challenge on appeal here rests entirely on the issue of claim construction.
Turning to the analysis, I agree with the majority's review of our relevant law. We have a well-established set of legal standards governing claim construction and the majority has already ably articulated those standards in detail. See Majority Op. 1373-74. Suffice it to say that "claim construction must begin with the words of the claims themselves," Amgen Inc. v. Hoechst Marion Roussel, Inc. , 457 F.3d 1293, 1301 (Fed. Cir. 2006), and then proceed to consider a term's meaning to one of ordinary skill in the art in light of the other intrinsic evidence, including the specification and prosecution history, Phillips v. AWH Corp. , 415 F.3d 1303, 1315-17 (Fed. Cir. 2005) (en banc).
In addition, our prior decisions have grappled with the more specific question of whether a particular clause in the body of a method claim is non-limiting. On one hand, we have given limiting effect to "wherein" clauses that "relate back to and clarify what is required by the count," giving "meaning and purpose to the manipulative steps." Griffin v. Bertina , 285 F.3d 1029, 1033 (Fed. Cir. 2002). Furthermore, if the clause "states a condition that is *1378material to patentability, it cannot be ignored." Hoffer v. Microsoft Corp. , 405 F.3d 1326, 1329 (Fed. Cir. 2005).
On the other hand, we have held that a clause that "merely states the result of the limitations" already in the claim "adds nothing to the patentability or substance of the claim." Texas Instruments Inc. v. U.S. Int'l Trade Comm'n , 988 F.2d 1165, 1172 (Fed. Cir. 1993). Similarly, a clause in a method claim "is not given weight when it simply expresses the intended result of a process step positively recited." Minton v. Nat'l Ass'n of Sec. Dealers, Inc. , 336 F.3d 1373, 1381 (Fed. Cir. 2003) ; see also Bristol-Myers Squibb Co. v. Ben Venue Labs., Inc. , 246 F.3d 1368, 1376 (Fed. Cir. 2001) ("[W]e agree with the defendants that this language is only a statement of purpose and intended result. The expression does not result in a manipulative difference in the steps of the claim.").
II
Turning to the application of these principles to the disputed clauses in claim 1 of the '453 patent, I agree with the majority in terms of the result. The "wherein" clauses here are limiting. I also agree with the majority's well-reasoned conclusion that the specification and prosecution history confirm the recited results were "material to patentability." But I would add one narrow but crucial point about the claim language at the outset of the majority opinion.
In my view, the claim language on its face confirms that these clauses give meaning and purpose to the other manipulative steps of claim 1. Griffin , 285 F.3d at 1033. The majority does not address the plain language of the claim. I take the opportunity to do so here. As explained below, Sandoz's arguments concerning the claim language fail.
Sandoz's principal argument about the claim language is that the "wherein" clauses merely describe results that occur whenever the claimed dosages are administered. To support this assumption, Sandoz provides no intrinsic evidence. Instead, it summarily asserts that "nothing in the intrinsic record identifies any combination of 0.2% brimonidine tartrate/0.68% timolol maleate administered twice daily that does not satisfy the wherein clauses." Reply Br. 25. But nothing in the claims-or the rest of the intrinsic record for that matter-states that any combination of these two compositions will satisfy the "wherein" clauses.
Thus, Sandoz's position falters from the start. Beginning with "the words of the claims themselves," Amgen , 457 F.3d at 1301, Sandoz has put forth no evidence how either clause "merely states the result of the limitations in the claim." Lockheed Martin Corp. v. Space Sys./Loral, Inc. , 324 F.3d 1308, 1319 (Fed. Cir. 2003). In my view, no part of the plain text of claim 1 shows these benchmarks are an inherent result of the dosage limitations.
Lest there be any doubt, the other language of claim 1 further exposes the lack of support for Sandoz's position. Claim 1 is written in open format. See '453 patent col. 9 ll. 16-17 ("A method of treating a patient with glaucoma or ocular hypertension comprising ...."). Therefore, while the claimed treatment must include the two recited compositions (0.2% w/v brimonidine tartrate and 0.68% w/v timolol maleate), other compositions can be present in the formulation, such as solvents, buffers, and preservatives typical for ophthalmic solutions. See id. col. 2 ll. 46-65, col. 3 ll. 23-30. Sandoz provides no explanation as to how *1379the recited dosages of brimonidine tartrate and timolol maleate will necessarily achieve these results if these other formulation parameters vary. In addition, the claims are not directed to a composition alone. Instead, they are directed to a method of "administering." Id. col. 9 l. 17. Without the "wherein" clauses, the only other limitation guiding the physician is that the drug is administered "twice" per day. Id . Sandoz provides no basis for us to conclude with any certainty that the safety and efficacy requirements of the "wherein" clauses would always result from two doses of (1) any formulation of the combination at (2) any interval in a 24-hour period.
Nor does the text of claim 1 suggest that these specific benchmarks are "only a statement of purpose and intended result." Bristol-Myers Squibb , 246 F.3d at 1376. On their face, these clauses state specific requirements rather than a general purpose or aspirational result for the claimed method. The efficacy clause does not simply require some general level of therapeutic effectiveness. Instead, it specifically requires that the claimed formulation be "as effective as the administration of 0.2% w/v brimonidine tartrate monotherapy three times per day." '453 patent col. 9 l. 20-col. 10 l. 1.
Likewise, the safety clause does not recite that the claimed method of treatment is generally intended to improve safety. Rather, with equal particularity, the safety clause of claim 1 here requires that the claimed formulation "reduces the incidence of one of more adverse events"-specifically "selected from the group consisting of conjunctival hyperemia, oral dryness, eye pruritus, allergic conjunctivitis, foreign body sensation, conjunctival folliculosis, and somnolence"-in comparison once again "to the administration of 0.2% w/v brimonidine tartrate monotherapy three times daily." Id. col. 10 ll. 1-7.
In light of this claim language, Sandoz's attempt to liken the clauses here to the non-limiting term in Minton founders. In Minton , we reasoned that the term "efficiently" "on its face does not inform the mechanics of how the trade is executed, and nothing in the specification or the prosecution history suggests otherwise." Minton , 336 F.3d at 1381 (emphasis added) (concluding term was a non-limiting statement of intended result of a claim directed to a system for executing securities trades). Here, Sandoz has provided no affirmative basis for us to conclude from the claim language alone that these detailed clauses-which specify clear and measurable metrics that the formulation must meet-are a mere intended result.
Therefore, I would reject Sandoz's position that the claim language on its face recites either an inherent or intended result.
III
The specification serves to reinforce the clear import of the claims. Indeed, this consistency between the claims and the specification is punctuated by the majority's own thorough review of the specification.
As explained by the majority, the specification confirms that "the claimed invention is ultimately a formulation" that meets the specific safety and efficacy outcomes in claim 1. See Majority Op. 1375. To support this conclusion, the majority points to various parts of the specification that essentially mirror the structure of the "wherein" clauses' language in claim 1, which uses specific prior art as a clear benchmark for measuring safety and efficacy. In the majority's *1380own words, these safety and efficacy "benefits are described throughout the specification with reference to prior art topical ophthalmic treatments, viz., Timolol BID and '0.2% w/v brimonidine tartrate' TID, as recited in the claims ." Id. at 1375 (emphasis added). Thus, the majority recognizes the parallel nature of the claims and specification.
Sandoz's attempts to undermine this consistency between the claims and the specification fail. The specification does not support Sandoz's position that any formulation involving the two compositions administered twice daily will satisfy the clauses. Indeed, Sandoz cannot point to any place in the specification that suggests these traits are inherently found in all formulations that combine 0.2% w/v brimonidine tartrate and 0.68% w/v timolol maleate.
Sandoz's reliance on Bristol-Myers Squibb is therefore misplaced. We have no reason to conclude that the safety and efficacy standards required by the "wherein" clauses "essentially duplicate[ ] the dosage amounts recited in the claims." Bristol-Myers Squibb , 246 F.3d at 1375 (finding claimed dosages are described in the specification as "antineoplastically effective"). Thus, these claims stand in stark contrast to the non-limiting clauses in both Bristol-Myers Squibb and other cases decided on similar facts. Id. ("The express dosage amounts are material claim limitations; the statement of the intended result of administering those amounts does not change those amounts or otherwise limit the claim."); see also In re Copaxone Consol. Cases , 906 F.3d 1013, 1023 (Fed. Cir. 2018) (finding clause "does not change the express dosing amount or method already disclosed in the claims").
Given the lack of support for its position, Sandoz tries to show inherent results by redrawing the claim's scope. Sandoz implies that the scope of the claims is limited to the preferred embodiment, Combigan®. The clinical trials of Combigan® discussed in the specification (Example II) achieved the benchmarks recited in the efficacy and safety clauses of claim 1. See Appellant's Br. 41-42; Reply Br. 23, 25. Sandoz jumps to the conclusion that the "wherein" clauses therefore only recite inherent or intended results.
Sandoz's position fails for at least two reasons. First, Sandoz did not argue that the claims should be construed as being limited to Allergan's commercial embodiment Combigan®. Second, there is no basis in the record before us for assuming that all formulations of the combination necessarily behave like Combigan®. Combigan® is "a sterile, preserved, aqueous solution" that uses the specific solvent, preservative, and buffers listed in Table 1 of the specification. '453 patent col. 3 ll. 65-67. Thus, Sandoz's arguments based on the clinical trial data for Combigan® carry little force.
In sum, no part of the specification justifies reading out the express language of the "wherein" clauses defining the claimed invention.
IV
Finally, the prosecution history cements the view that the claim language at issue is clearly limiting. Again, one need look no further than the majority's opinion for proof. See Majority Op. 1372-76. The majority's cogent analysis of the prosecution history underscores the consistency between the claim language and the rest of the intrinsic record.
The majority opinion relies on arguments Allergan made during prosecution *1381that were directly based on the claim language itself. As the majority observes, "Allergan relied on the efficacy and safety of the claimed methods during prosecution of the Patents-in-Suit in responding to the examiner's rejections." Id. at 1375. In particular, the majority points to Allergan's statements to the Examiner that parrot the express language of claim 1. "Allergan explained that the prior art 'does nothing to teach or suggest that the claimed fixed combination of brimonidine tartrate and timolol maleate administered twice daily would be as effective as the administration of 0.2% w/v brimonidine tartrate monotherapy three times per day ....' " Id. (quoting J.A. 944) (emphasis added); see also '453 patent col. 9 l. 20-col. 10 l. 1 (wherein method is "as effective as the administration of 0.2% w/v brimonidine tartrate monotherapy three times per day").
In turn, the majority emphasizes that the Examiner credited Allergan's arguments based on the claim language, which showed that "the prior art 'fail[ed] to teach the reduction in adverse events as compared to the administration of 0.2% w/v brimonidine tartrate monotherapy three times a day as claimed .' " Id. at 1376 (quoting J.A. 976-77, 998-99, 1009-10) (emphasis added).
Accordingly, I would hold that the claim language, specification, and prosecution history all consistently show that the "wherein" clauses place meaningful limitations on the scope of the invention. On its face, claim 1 is limited to only those formulations of "0.2% w/v brimonidine tartrate" and "0.68% w/v timolol maleate" administered twice daily that meet the specific safety and efficacy benchmarks. The specification and prosecution history expunge any lingering doubts that these clauses define the scope of the invention. The "wherein" clauses are important for "giving meaning ... to the manipulative steps," rather than simply being an "inherent result of performing the manipulative steps." Griffin , 285 F.3d at 1033-34.
V
I offer one final point about the importance of the plain language of the claim here. Sandoz's arguments raise general concerns about the role of claim language in claim construction.
Sandoz attempts to label over half the claim language in claim 1 as a mere "intended result." By doing so, Sandoz invites us to start from a place of uncertainty about whether most of the text in the body of the claim is limiting. Accepting that invitation threatens the broader notice function of the patent claim. "It is a 'bedrock principle' of patent law that 'the claims of a patent define the invention to which the patentee is entitled the right to exclude.' " Phillips , 415 F.3d at 1312 (quoting Innova/Pure Water, Inc. v. Safari Water Filtration Sys., Inc. , 381 F.3d 1111, 1115 (Fed. Cir. 2004) ). Indeed, "[b]oth the Supreme Court and this court have adhered to the fundamental principle that claims define the scope of patent protection .... The claims thus give notice of the scope of patent protection." Johnson & Johnston Assocs. v. R.E. Serv. Co. , 285 F.3d 1046, 1052 (Fed. Cir. 2002) (en banc).
As it is, claim construction can be difficult. For instance, litigants often encounter uncertainty over whether a claim's preamble is limiting or not. I see no reason to inject further uncertainty into the notice provided by the body of a claim. Given the specificity, clarity, and material limits the "wherein" clauses add to the scope of claim *13821 on their face, Sandoz's position deserves rigorous scrutiny from the start. We should not begin with the presumption that text in the body of the claim may be meaningless and can only be saved by clear statements in the specification or prosecution history.
Therefore, I would affirm the district court's decision for the reasons stated in the majority's opinion, except that I would start by explaining why the plain language of the claim compels us to reject Sandoz's position that the "wherein" clauses here are mere statements of inherent or intended results.