Opinion of the Court filed by Circuit Judge PROST. Dissenting opinion filed by Circuit Judge BRYSON.
PROST, Circuit Judge.
Appellants Novartis Pharmaceuticals Corporation, Novartis AG, Novartis Phar-[1330]*1330ma AG, and Novartis International Pharmaceutical Ltd. (collectively “Novartis”) appeal the claim construction and entry by the United States District Court for the District of Delaware of a judgment as a matter of law (“JMOL”) for noninfringement of United States Patent No. 6,007,-840 (“the '840 patent”) in favor of Abbott Laboratories (“Abbott”). Novartis Pharms. Corp. v. Abbott Labs., Inc., 294 F.Supp.2d 557 (D.Del.2003). While we reverse the claim construction of “lipophilic component,” we perceive no error in the district court’s finding that no reasonable juror could conclude that an equivalent to a pharmaceutically acceptable non-surfactant lipophilic excipient, which is capable of dissolving cyclosporin, is present in Abbott’s accused product. Accordingly, we affirm the grant of JMOL for noninfringement. .
BACKGROUND
Novartis commenced a lawsuit against Abbott for allegedly infringing United States Patent No. 5,342,625 (“the '625 patent”) and the '840 patent. The '625 and '840 patents relate to pharmaceutical compositions of the drug cyclosporin, which help prevent organ rejection in transplant patients. Cyclosporin is highly hydrophobic, and as a consequence, has been difficult to administer in a convenient form that provides the desired bioavailability to a patient. '840 patent, col. 3, 1. 32, col. 4, 11. 32-40. The two patents asserted by Novartis are directed to compositions that facilitate human absorption of cyclosporin by use of a microemulsion. '625 patent, col. 5, 11. 1-5; '840 patent, col. 4, 1. 64-col. 5,1.1.
After the district court conducted a Markman hearing and construed the disputed terms in the '625 and the '840 patents, Novartis narrowed its allegation of infringement in its complaint to claim 1 of the '625 patent1 and claim 81 of the '840 patent2. Each of the remaining asserted claims calls for a microemulsion pre-con-centrate composition that, upon addition of water," forms an oil-in-water (“o/w”) mi-croemulsion. The parties disputed, among other things, whether Abbott’s product includes a “lipophilic phase component” in the '625 patent or a “lipophilic component” in the '840 patent. In addition, the parties disputed whether “surfactant,” as it is used in each of the asserted patents, encom[1331]*1331passes both hydrophilic and lipophilic surfactants.
In its claim construction of the phrase “lipophilic phase component,” the district court held that the term requires:
at least one excipient meeting the following criteria: (1) a pharmaceutically acceptable lipophilic solvent in which cyclosporin is soluble, which is (2) immiscible with both water and the hydro-philic phase component(s) (in the absence of a surfactant), and which (3) lacks the amphiphilic function characteristic of a surfactant (i.e. it must not be a surfactant).
Novartis, 294 F.Supp.2d at 561 (emphasis added). The district court construed “surfactant” to encompass both hydrophilic and lipophilic surfactants. While Novartis disputes whether the claim construction of “lipophilic phase component” is correct, the parties agree that this term bears the same meaning as the term “lipophilic component,” as used in the '840 patent.3
According to Abbott, its accused product, Gengraf, is a cyclosporin formulation that contains only hydrophilic excipients and surfactants. The accused product contains an ingredient called Span 80, which Abbott contends is a surfactant and not a “lipophilic component.” After the parties put on their evidence, the district court permitted Novartis to submit to the jury the question of whether Abbott infringes either of the asserted patents under the doctrine of equivalents. The jury returned a verdict finding that Abbott did not infringe claim 1 of the '625 patent, but did infringe claim 81 of the '840 patent. Abbott moved for a judgment as a matter of law (“JMOL”) in its favor on claim 81, or in the alternative, a new trial because the district court had erroneously excluded relevant evidence and because the '840 verdict is facially inconsistent with the '625 verdict and against the clear weight of evidence.
The district court granted Abbott’s motion for a JMOL. Id. at 567. The court noted that under its claim construction, “the ‘lipophilic phase component’ cannot be a surfactant.” Id. at 565; see also id. at 566 (“Under the Court’s claim construction, Novartis is precluded from claiming that surfactants are the equivalent of the ‘lipophilic phase component.’ ”). In construing the claims and concluding that surfactants cannot be a component of the lipophilic phase, the district court relied on a. passage from the '625 specification, which states:
Suitable components for use as lipophilic phase include any pharmaceutically acceptable solvent which is non-miscible with the selected hydrophilic phase, e.g. as defined under (1.1.) or (1.2.). Such solvents will appropriately be devoid or substantially devoid of surfactant function. ■
'625 patent, col. 8, 11. 58-63. The court effectively read this language as a disavowal of any lipophilic component having surfactant function. In addition, the district court noted that the enumeration of three excipient components in the claim language gave support to viewing each component as separate and distinct. Lastly, the district court observed that the patentee had acknowledged during the prosecution of the application that led to the '840 patent that distinctions exist between surfactant and lipophilic compounds that differentiate the chemical entity of each class of compound. According to the district court, “[bjecause surfactants are expressly excluded from coverage, a surfactant acting as the ‘lipophilic phase com[1332]*1332ponent’.is inconsistent with the language of the claim, as construed by the Court.” Novartis, 294 F.Supp.2d at 567. Because Span 80 is a surfactant, the district court reasoned that Novartis could not, as a matter of law under the specific exclusion principle, establish that the substance is an equivalent to the “lipophilic component.” Id. Accordingly, the district court concluded that Novartis could not demonstrate that Abbott infringes under the doctrine of equivalents. Id.
STANDARD OF REVIEW
In a determination of infringement, the claims must be first construed by the court, and then the properly construed claims must be compared' to the accused product by the fact finder. Cybor Corp. v. FAS Techs. Inc., 138 F.3d 1448, 1454 (Fed.Cir.1998) (en banc). If challenged, we review claim construction de novo. Golight, Inc. v. Wal-Mart Stores, Inc., 355 F.3d 1327, 1330 (Fed.Cir.2004); Markman v. Westview Instruments, Inc., 52 F.3d 967, 970 (Fed.Cir.1995) (en banc), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996). We likewise review the grant of a JMOL for noninfringement de novo, reapplying the JMOL standard used by the district court. Ericsson, Inc. v. Harris Corp., 352 F.3d 1369, 1373 (Fed.Cir.2003); Markman, 52 F.3d at 975. “To prevail on a renewed motion for JMOL following a jury trial,, a party ‘must show that the jury’s findings, presumed or express, are not supported by substantial evidence or, if they were, that the legal conclusion(s) implied [by] the jury’s verdict cannot in law be supported by thpse findings.’ ” Pannu v. Iolab Corp., 155 F.3d 1344, 1348 (Fed.Cir.1998) (citing Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 893 (Fed.Cir.1984)); Fed. R.Civ.P. 50(b). If the former situation applies, “we must view the evidence in the light most favorable to the non-moving party, and determine whether the record contains the ‘minimum quantum of evidence from which a jury might reasonably afford relief.’ ” Glenn Distribs. Corp. v. Carlisle Plastics, Inc., 297 F.3d 294, 299 (3d Cir.2002) (quoting Parkway Garage, Inc. v. City of Philadelphia, 5 F.3d 685, 691 (3d Cir.1993)).
DISCUSSION
I. Collateral Estoppel
As a threshold issue, Abbott asserts that Novartis is barred in this appeal from challenging the jury verdict on the '625 patent or the JMOL on the '840 patent. Because the parties agree that the terms “lipophilic component” and “surfactant” are common to the '625 and '840 patents, Abbott argues that Novartis’ failure to challenge the construction of these terms in connection with the '625 patent estops Novartis from challenging their meaning in the related '840 patent. According to Abbott, “any legal determinations or factual findings necessary to the jury’s verdict as to claim 1 of the '625 patent, the parent of the '840 patent, must be accorded pre-clusive effect.” (Citing Mycogen Plant Science, Inc. v. Monsanto Co., 252 F.3d 1306, 1311 (Fed.Cir.2001)).
Novartis claims that it preserved its right to challenge the jury verdict on the '625 patent, and that collateral es-toppel does not apply to preclude its challenge to the JMOL on the '840 patent. As evidence, Novartis notes that when discussing the construction of the term “surfactant,” it pointed out that the correct construction of this term is relevant only to the '625 patent. Novartis further observes that it requested that “if this Court reverses the district court’s claim construction” and “if this Court does not reinstate thé jury [infringement] verdict,” then “it is requested that ... the case [be] remanded for further proceedings consistent with the [1333]*1333proper construction of the claim terms.” According to Novartis, the “further proceedings” necessarily include proceedings on claim 1 of the '625 patent as well as claim 81 of the '840 patent. Novartis moreover notes that Abbott admitted that the Notice of Appeal may be broad enough to encompass the appeal of the jury verdict on the '625 patent. In sum, Novartis asserts that (1) its conditional request of this court to construe a term for the '625 patent, (2) its request to remand “for further proceedings,” and (3) Abbott’s concession that the Notice of Appeal may be sufficiently broad to encompass an appeal of the '625 patent verdict, suffice to show that it preserved its right to challenge the claim construction, the noninfringement verdict on the '625 patent, and the JMOL that followed.
We apply the law of the regional circuit to the issue of collateral estoppel in circumstances such as those presented here. RF Del., Inc. v. Pac. Keystone Techs., Inc., 326 F.3d 1255, 1261 (Fed.Cir.2003). The party seeking to invoke collateral estoppel bears the burden, Dici v. Pennsylvania, 91 F.3d 542, 548 (3d Cir.1996), of showing the following four elements are present: “(1) the previous determination was necessary to the decision; (2) the identical issue was previously litigated; (3) the issue was actually decided in a decision that was final, valid, and on the merits; and (4) the party being precluded from relitigating the issue was adequately represented in the previous action.” Hawksbill Sea Turtle v. FEMA, 126 F.3d 461, 475 (3d Cir.1997) (citing Raytech Corp. v. White, 54 F.3d 187, 190 (3d Cir.1995)).
Starting with Novartis’ contention that its appeal of the jury verdict on the '625 patent was preserved, we note that Novartis, from the beginning, has maintained that “this is an appeal from entry of JMOL overturning a jury verdict finding Abbott guilty of infringement under the doctrine of equivalents.” (Emphasis added). According to Novartis, “[t]he element at the core of this appeal (in that it formed the basis for the district court’s grant of JMOL) is the ‘lipophilic component’ element.” (Emphasis added). As a consequence of the district court’s allegedly erroneous claim construction, Novartis urged this court to “overturn the JMOL ruling ” and reinstate the jury’s verdict. (Emphasis added). Novartis urged that in the alternative, if this court decides against reinstating the verdict, but rather “orders a retrial, [this court] should construe the term ‘surfactant’ to provide guidance ” to the district court. (Emphasis added). The purpose of examining the term “surfactant,” therefore, is not to attack the '625 patent jury verdict, as Novartis belatedly now contends in response to Abbott’s collateral estoppel argument, but rather to provide guidance to the district court if we order a retrial of the '840 patent case. In support of its proposed alternative, Novartis cited Altiris, Inc. v. Symantec Corp., 318 F.3d 1363, 1366 (Fed.Cir.2003), as a parallel situation. There, this court consented to construe terms that did not form a part of the appealed summary judgment because they “may be relevant to the remand determination of infringement” and because they were found to be construed erroneously. Id. We find no language in Novartis’ opening brief stating that this is also an appeal from the jury verdict on the '625 patent or arguments explaining why the allegedly erroneous claim construction compels us to conclude that the jury’s non-infringement verdict on the '625 patent mandates a reversal. It is our understanding from Novartis’ statement of its position that “this is an appeal from entry of JMOL” and that this court need not broach the meaning of “surfactant,” which itself does not form a part of the appealed [1334]*1334JMOL, unless this court remands the case for a retrial. For reasons discussed below, we affirm the JMOL, and thus we find remand unwarranted. Based on our affir-mance, our review of the term “surfactant,” as it pertains to the '625 patent, is uncalled for.
As to collateral estoppel, we find it inapplicable to the JMOL on the '840 patent. Here, the jury determined that Novartis failed to prove by a preponderance of evidence that Abbott infringed the '625 patent under the doctrine of equivalents. The verdict form on which the jury indicated its decision did not require the jury to specify the one or more limitations in claim 1 that it found Abbott’s product did not contain. Further, there is no record evidence explaining the jury’s rationale for its verdict. Assuming that the district court’s claim construction of “lipophilic component” was the dispositive element that proved necessary to the jury’s noninfringement verdict, as Abbott urges, it would appear that the jury’s infringement verdict on the '840 patent is quite possibly contradictory. One explanation may be as Novartis suggests. According to Novartis, the jury may have “include[d] all of the SPAN® 80 and CREMOPHOR® EL ingredients within the ‘surfactant’ component of the ['625] claim, thereby leaving no ingredient to serve as the lipophilic component.” Because claim 81 called more specifically for a hydrophilic surfactant, the jury could have concluded that Cremophor EL is the only surfactant that meets the hydrophilic surfactant element, leaving the jury free to conclude that the unaccounted for Span 80 was, based on testimony from Novartis’ expert, not acting as a surfactant but rather as a lipophilic component in the claimed composition. In any event, we cannot conclude from the limited information available that the district court’s construction of “lipophilic phase component,” proved necessary to the jury’s noninfringement decision on the '625 patent. Accordingly, Abbott has not met its burden to show that the claim construction of that term by the district court was necessary to the jury’s noninfringement decision. Collateral estoppel therefore should not apply to preclude Novartis from challenging the claim construction and the JMOL in regards to the '840 patent.
II. Claim Construction
In construing claims, our analysis must begin and remain centered on the claim language itself. Interactive Gift Express, Inc. v. Compuserve, Inc. 256 F.3d 1323, 1331 (Fed.Cir.2001). We generally presume that the words of a patent claim have the meaning that a person of ordinary skill in the relevant art would ordinarily attribute to them. Tex. Digital Sys., Inc. v. Telegenix, Inc., 308 F.3d 1193, 1202 (Fed.Cir.2002); Johnson Worldwide Assocs., Inc. v. Zebco Corp., 175 F.3d 985, 989 (Fed.Cir.1999). The presumption may be rebutted when: (1) the patentee has chosen to be his own lexicographer, or (2) a claim term lacks such clarity that there is “no means by which the scope of the claim may be ascertained from the language used.” Johnson Worldwide, 175 F.3d at 990.
If the disputed claim term “is a term with no previous meaning to those of ordinary skill in the prior art[,][i]ts meaning, then, must be found [elsewhere] in the patent.” J.T. Eaton & Co. v. Atl. Paste & Glue Co., 106 F.3d 1563, 1568 (Fed.Cir.1997). Most often, the specification “is the single best guide to the meaning of a disputed term.” Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed.Cir.1996). Even when guidance is not provided in explicit definitional format, “the specification may define claim terms ‘by implication’ such that the meaning may be ‘found in or ascertained by a reading of the patent documents.’ ” Bell Atl. Network [1335]*1335Servs., Inc. v. Covad Communications Group, Inc., 262 F.3d 1258, 1268 (Fed.Cir.2001) (quoting Vitronics, 90 F.3d at 1582, 1584 n. 6). In addition to the specification, we must examine the prosecution history to determine whether the patentee- relinquished claim coverage by amendment or through argument to overcome or distinguish a reference. Id. Although extrinsic evidence may be used by the court to help understand the disputed limitation, it may not be used to vary, contradict, expand, or limit the claim language from how it is defined, even by implication, in the specification or file history. Vitronics, 90 F.3d at 1584-85.
Novartis claims that the district court erred by not giving the term “lipophilic component” the broadest meaning possible. More specifically, Novartis asserts the district court unnecessarily added limitations concerning whether the “lipophilic component” may include surfactants and whether the constituents of the lipophilic component must be immiscible.
A.
Novartis begins its analysis by observing that “lipophilic” has a well-defined meaning to those skilled in the art. As defined by Novartis’ expert, lipophilic means “fat- or oil-loving, or water hating.” Novartis then notes that the specification teaches that:
The relative proportions will also vary, depending on the particular function of ingredients in the composition, for example, in the case of a surfactant component of a “microemulsion pre-concen-trate”, on whether this is employed as a surfactant only or both a surfactant and a co-solvent.
’840 patent, col. 16,11. 36-41. Further, the specification teaches that:
When the surfactant comprises an effective solvent for the cyclosporin active ingredient, as in the case e.g. of surfactants or mixtures of surfactants under (3.1.1.) to (3.2.7.) above, it may be incorporated into compositions as defined under (A), not only as surfactant, but in excess as an additional carrier or co-solvent phase, i.e. as part of the hydro-philic or lipophilic phase.
Id., col. 12,11. 44-49. Novartis claims that these teachings are inconsistent with the court’s construction that a surfactant can form no part of the “lipophilic component.” We agree with Novartis.
Claim 81 of the '840 patent recites a pharmaceutical composition that forms a microemulsion upon dilution by water. The composition comprises, in addition to the active ingredient, three other components. Namely, it contains (1) a hydro-philic component, (2) a “lipophilic component,” and (3) a hydrophilic surfactant. Neither party cites any persuasive record evidence that the term “lipophilic component” (or “lipophilic phase component”) had a well-defined meaning to those of ordinary skill in the art. Accordingly, the claim term requires that we turn to the intrinsic evidence, and if necessary, the extrinsic evidence, to elucidate the meaning that one skilled in the art would give the term. Johnson Worldwide, 175 F.3d at 990.
The passages of the specification quoted above indicate that a surfactant or a mixture of surfactants, provided it is an “effective solvent for the cyclosporin,” may form, contrary to the district court’s construction, a part of the lipophilic phase that dissolves the cyclosporin. Furthermore, the '840 patent discloses that if a surfactant forms a part of the lipophilic phase, it does so as a co-solvent. Col. 12, 11. 43-49, col. 16,11. 36-41. A co-solvent is a “solvent that in conjunction with another solvent can dissolve a solute.” Webster’s Third New International Dictionary (‘Webster’s ”) 514 (1993) (emphasis added). No[1336]*1336vartis’ expert opines that the specification’s disclosure indicates that a surfactant “could function alone ” as the solvent in the lipophilic phase. (Emphasis added). By definition, however, a co-solvent operates with “another solvent” to dissolve a solute, such as cyclosporin. Novartis’ expert testimony therefore directly contradicts the express teaching of the specification, and may not be used to give us appropriate guidance to reach the correct claim construction.
At oral argument, Novartis attempted to make a similar argument, asserting that the description of surfactants as making up only a “part” of the lipophilic phase, see '840 patent, col. 12,11. 48-49, indicates that surfactants could constitute the whole lipo-philic phase. “Part,” however, means “something less than a whole.” Webster’s at 1645. Because Novartis supplies insufficient record evidence to validate its unconventional interpretations of “co-solvent” and “part,” as those words are used in the specification, Novartis’ reading of the specification must be rejected.
In an attempt to further support its view that surfactants may constitute the entire lipophilic phase, Novartis notes that Example 4.1 in the '840 patent describes a “regular emulsion pre-concentrate” in which only a surfactant is used to dissolve the cyclosporin. The composition defined by claim 81 is not a regular emulsion pre-concentrate, however, but rather a microe-mulsion pre-concentrate, and therefore the example does nothing to advance our understanding of what composes the “lipo-philic component” in the claimed composition.4
Because the specification teaches that “mixtures of surfactants,” if used to dissolve the cyclosporin, form only a “part” of the lipophilic phase, '840 patent, col. 12, 11. 43-49 (emphasis added), the other solvent forming the rest of the lipophilic phase must not be another surfactant. Rather, according to the specification, the remainder of the solvent constituting the lipophilic phase is a pharmaceutically acceptable non-surfactant lipophilic excipient that is capable of dissolving cyclosporin, such as those described under section (2) of the specification. Id., col. 8, 11. 66-67; col. 9, 11. 5-46; col. 12, 11. 43-49. Accordingly, while we agree with Novartis that surfactants may form a part of the lipophilic component, the intrinsic record shows that this component cannot be composed entirely of surfactants, as urged by Novartis. Although Abbott initially argued in its opposition brief that the '840 patent “requires the lipophilic (phase) component to be comprised of an excipient other than a surfactant,” Abbott later admitted in its cross-appeal brief and at oral argument that “a surfactant [may] contribute[ ] to the lipophilic phase so long as the lipophilic phase also contained at least one non-surfactant excipient.” In view of the intrinsic evidence, the claimed “lipophilic component” must contain, at minimum, a pharmaceutically acceptable lipophilic substance capable of dissolving cyclosporin, and that is a non-surfactant excipient.
B.
Novartis also challenges the district court’s inclusion of the requirement that the lipophilic component must be im[1337]*1337miscible with the hydrophilic phase in the absence of surfactants. The specification states that “suitable components ” for use in the lipophilic phase may include any pharmaceutically acceptable solvent that is non-miscible with the selected hydrophilic phase. '840 patent, col. 9, 11. 1-3 (emphasis added). The specification, contrary to the district court’s reading, does not teach that the entire lipophilic phase must be immiscible with the hydrophilic phase, but rather only components, e.g., parts, of the phase, in one embodiment,. should be immiscible. We find nothing within the intrinsic record that compels reading the immiscibility limitation into the “lipophilic component” term.
III. Jury Verdict
Novartis asserts that even if we uphold the district court’s construction of “lipo-philic component,” the jury’s verdict on the '840 patent should be allowed to stand because there was substantial evidence for the jury to conclude that Abbott’s cyclos-porin composition infringed under the doctrine of equivalents. In granting JMOL, Novartis argues that the district court compounded the error of its claim construction by erroneously applying the specific exclusion principle and the all-elements rule. Moreover, Novartis suggests that even if we allow the claim construction to stand, the latter errors marídate a reversal of the JMOL.
In light of its exclusion of surfactants from the lipophilic component, the district court found that the specific exclusion principle precluded it, as a matter of law, from treating the surfactant as an element forming the lipophilic component.’ Novartis, ' 294 F.Supp.2d at 567. Left with no identifiable compound in the accused product corresponding to the lipophilic component, the district court held, under the all-elements rule, that no reasonable fact finder could conclude that Abbott infringes under the doctrine of equivalents. Id.
As reviewed above in connection with claim construction, the specification informs us that the lipophilic component encompasses compositions that may include one or more surfactants, so long as the lipophilic component also contains at least one non-surfactant lipophilic excipient, which is capable of dissolving cyclosporin. Novartis argued that Span 80, which is a surfactant and an ingredient of the accused product, meets the lipophilic component limitation under the doctrine of equivalents. Novartis cites the testimony of its expert who opined at trial that Span 80 “does not exhibit the amphiphilic function of a surfactant” nor does it “function as a surfactant ... but, rather, that it functions as the lipophilic component.” (Emphases added). The fact that Span 80 may “function” as a co-solvent to cyclosporin does not answer, however, whether a suitable non-surfactant excipient is also present in the accused composition, thereby bringing the alleged co-solvent mixture within the scope of the term “lipophilic component,” either literally- or under the doctrine of equivalents.
In light of the specification’s implicit teaching that surfactants do not compose the entire portion of the lipophilic component, Novartis is foreclosed from arguing that Span 80, which the specification expressly acknowledges is a surfactant, '840 patent, col. 9,11. 49-50; col. 11, 1L58-62, is an equivalent to a pharmaceutically acceptable non-surfactant lipophilic exeipiént, as required by the lipophilic phase under our claim construction. See SciMed Life Sys., Inc. v. Advanced Cardiovascular, 242 F.3d 1337, 1347 (Fed.Cir.2001) (“[I]f a patent states that the claimed device must be ‘non-metallic,’ the patentee cannot assert the patent against a metallic device on the ground- that a metallic device is equivalent to a non-metallic device.”); Durel Corp. v. Osram Sylvania Inc., 256 F.3d 1298, 1305 (Fed.Cir.2001) (“A finding of equivalence [1338]*1338[with Sylvania’s AIO(OH) and Al(OH)3 coatings] would vitiate the limitation ‘oxide coating,’ which we have concluded is defined to primarily consist of a binary compound.”); see also Athletic Alternatives, Inc. v. Prince Mfg. Inc., 73 F.3d 1573, 1582 (Fed.Cir.1996) (“As a corollary to the ‘all limitations’ rule ..., we have held that ‘the concept of equivalency cannot embrace a structure that is specifically excluded from the scope of the claims.’ “ (citation omitted)).
Novartis claims that the district court’s holding that Span 80 may not, as a matter of law, be regarded as an equivalent to a non-surfactant excipient under the asserted patents is contrary to the Supreme Court’s decision in Graver Tank & Manufacturing Co. v. Linde Air Products Co., 339 U.S. 605, 70 S.Ct. 854, 94 L.Ed. 1097 (1950), and this court’s decision in Wright Medical Technology, Inc. v. Osteonics Corp., 122 F.3d 1440 (Fed.Cir.1997). We disagree. In Graver Tank, the patentees asserted four claims directed to welding flux made of alkaline earth metal silicates. 339 U.S. at 607, 70 S.Ct. 854. Except for the type of metal silicate present, the accused product was alike in all respects to the claimed welding flux. Id. at 610, 70 S.Ct. 854. The parties disputed whether silicates made of manganese metal in the accused product could be treated as an equivalent to silicates made of alkaline earth metal. Id. The Court found that the. difference between the metal silicates was insubstantial, and they therefore could be considered equivalents. Id. Notably, there was no dispute that the metal silicate element was present in the accused product. Rather, the dispute focused on a limitation of the metal silicate element.
Similarly, in Wright Medical, there was no dispute that the accused product employed intramedullary rod elements, just as in the claimed invention. 122 F.3d at 1445. The asserted independent claim defined a femoral surface shaping guide comprising “an intramedullary rod portion adapted to closely fit in and extend through the narrowest portion of the human femur.” Id. at 1442 (emphases added). The district court construed “closely fit” to mean “a tight or snug fit between the rod and the isthmus of the femur” and “extend through” to require that the “rod pass through the entire length of isthmus.” Id. at 1444. Under this claim construction, the parties agreed that the intramedullary rods of the accused product did not fit tightly against or extend through the isthmus of the femur, and thus the product did not literally infringe the asserted patent. Id. The parties disputed, however, whether the rods in the accused product were un-substantially different from that claimed, and thus, whether the accused product infringed under the doctrine of equivalents. Id. at 1445. The district court granted summary judgment to the accused infringer, rejecting the patentee’s theory for infringement by equivalence because the theory “attempted] to ignore claim limitations on which the public is entitled to rely.” Id. We found that summary judgment for the accused infringement was improper, and held that the fact finder must be given the opportunity to determine whether the intramedullary rod in the accused product is equivalent to that used in the claimed invention. Id. at 1446.
In both Graver Tank and Wright Medical, the fact that certain claimed limitations in the element at issue were missing in the accused product did not change the fact that the element, albeit different from that expressly claimed, was indeed present in the accused product. In Graver Tank, for instance, although the accused product used manganese metal instead of an alkaline earth metal, that substitution did not transform the disputed element, i.e ., metal silicate, into something that was not a metal silicate. Similarly, in Wright Medical, [1339]*1339although the intramedullary rod in the' accused product did not tightly fit and extend through the isthmus of the femur, as required by the literal claim language, the absence of those claimed limitations did not vitiate the fact that the accused product possessed an “intramedullary .rod.”
When the substitution of one feature, however, for another into an element of the accused product places it outside the scope of the recited claim element, the doctrine of equivalents may not be applied. Sage Prods., Inc. v. Devon Indus., Inc., 126 F.3d 1420, 1423 (Fed.Cir.1997); see, e.g., Durel Corp., 256 F.3d at 1305 (finding that the addition of hydroxyl groups to the alleged oxide coating in the accused product made it no longer an “oxide coating” as properly construed under the asserted patent, and thus placing the accused product outside the reach of the doctrine of equivalents). Permitting such an element in .the accused product to come within the bounds of the claimed element would impermissibly extend the scope of the claim language beyond what the patented actually claimed.
Here, the “lipophilic component” element requires at least a pharmaceutically acceptable non-surfactant lipophilic excipi-ent capable of dissolving cyclosporin. Under Novartis’ theory of infringement by equivalence, Span 80, admittedly a surfactant according to the '840 patent, see col. 9, 11. 47-50, col. 11, 11. 58-59, is an equivalent to the non-surfactant excipient required by the claimed “lipophilic component.” Including a chemical composition made entirely of surfactants within - the scope of “lipophilic component” would bé inconsistent with the claim construction of that term. We must reject Novartis’ theory of infringement by equivalence. Because Novartis failed to adduce evidence from which a reasonable juror could conclude, consistent with this court’s claim construction, that there is a “lipophilic component” in Abbott’s accused product, we find no basis to reverse the JMOL for noninfringement of the '840 patent.
IV. Cross-Appeal.
In Abbott’s conditional cross-appeal, it claims that in the event we reverse the JMOL, we should not reinstate the jury verdict. Rather, Abbott urges tliis court to grant it a new trial on the '840 patent, because the jury verdict of infringement of the '840 patent - is irreconcilably inconsistent with' the jury verdict of noninfringement as to the '625 patent. In addition to the inconsistency, Abbott contends that it is entitled to a new trial on the '840 patent because the jury’s verdict is contrary to the clear weight of the -evidence and because the district court erroneously excluded evidence that Abbott holds its own patents for its accused product.
The district court granted Abbott’s motion for JMOL for noninfringement and denied Abbott’s alternative motion for a. new trial as moot. Novartis, 294 F.Supp.2d at 560. “[A] party who prevails on noninfringement has no right to file a ‘conditional’ cross-appeal to introduce new argument or challenge a claim construction, but may simply assert alternative grounds in the record for affirming the judgment.” Phillips v. AWH Corp., 363 F.3d 1207, 1216 (Fed.Cir.2004). “It is only necessary and appropriate to file a cross-appeal when a party seeks to enlarge its own rights under the judgment or to lessen the rights of its adversary under the judgment.” Bailey v. Dart Container Corp. of Mick, 292 F.3d 1360, 1362 (Fed.Cir.2002). The conditional cross-appeal is dismissed as improper because the district court entered a judgment of noninfringement in Abbott’s favor. See Phillips, 363 F.3d at 1216.
CONCLUSION
Although the district court’s claim construction was erroneous, we find, under [1340]*1340our amended claim construction, no error in the district court’s grant of the JMOL. We reverse the claim construction but affirm the grant of JMOL for noninfringement on claim 81 of the '840 patent.
AFFIRMED.