OPINION

JEFFREY V. BROWN, Justice.

In this case, appellant Susan Brockert raises two issues: (1) whether federal law preempts her failure-to-warn claims arising from her use of a prescription drug containing estrogen combined with progestin in hormone-replacement therapy; and (2) whether her design-defect claim that estrogen alone is a superior form of hormone-replacement therapy compared to estrogen with progestin is viable under Texas strict-liability and negligence law. For the reasons explained below, we hold that federal law does not preempt Brockert's failure-to-warn claims, but her design-defect claim is not viable under Texas law. We therefore reverse in part, affirm in part, and remand.

Background

Susan Brockert and several other plaintiffs sued Wyeth Pharmaceuticals, Inc., and its sales representatives including Kay Lamberth, Lana Pavlesic, James Ty, Ann Callan, and Patrick Tyrrell (collectively, "Wyeth"), along with other defendants, alleging that they were injured by various prescription drugs used in hormone-replacement therapy ("HRT"). The plaintiffs' claims included failure to warn, design defect, manufacturing defect, fraud, negligence, negligent misrepresentation, breach of warranty, and deceptive trade practices. Relevant here, Brockert contends that she experienced breast cancer as a result of taking Prempro, an HRT drug manufactured, marketed, and sold by Wyeth.

In the 1940s, well before the introduction of Prempro, Wyeth manufactured a hormone-replacement drug known as Premarin. Premarin was an estrogen-only compound used to treat menopausal symptoms. This treatment was commonly known as estrogen-replacement therapy, or ERT. However, after an increase in endometrial cancer became associated with the use of estrogen alone, scientists determined that the addition of another hormone, progesterone, to ERT could reduce the risk of endometrial cancer.

In 1992, Wyeth submitted a New Drug Application ("NDA") to the Food and Drug Administration ("FDA") for Prempro. Prempro was a combination of estrogen and progestin, a synthetic progesterone, to be used for treating menopausal symptoms and preventing osteoporosis, while protecting against endometrial hyperplasia in women with an intact uterus. As part of the NDA, Wyeth submitted a proposed warning labeling for Prempro.

During the FDA's consideration of the Prempro NDA, it evaluated scientific studies on the risk of breast cancer with estrogen alone and with estrogen and progestin in combination. As a result of that analysis, the FDA recognized that some studies had shown an increased risk of breast cancer with estrogen alone and that the combination of estrogen and progesterone may exacerbate that risk. The FDA viewed the effect of concomitant use of estrogen and progesterone on breast-cancer incidence as an important safety issue concerning this class of drugs, but recognized that "[b]ecause of the long latency between exposure to promotional agents and detection of clinical tumors . . . prospective studies take many years to conduct and require extremely large sample sizes to ensure statistically meaningful treatment group comparisons" and therefore "many more years are still needed before the relationship between HRT and breast cancer can be definitively determined."

The FDA also evaluated Wyeth's proposed labeling for Prempro and considered how it wanted the final label to read. The wording of the label evolved as Wyeth and the FDA communicated concerning the FDA's proposed revisions. Among other things, the FDA instructed Wyeth to revise portions of its warnings concerning the risk of breast cancer, and required language warning of the possible increased risk of breast cancer from the use of estrogen with progestin and warning that the risk may be higher than that associated with, the use of estrogen alone. The FDA's recommendation for approval of Prempro was specifically conditioned on Wyeth using the FDA's specified language for the Prempro labeling, including its modifications to the breast-cancer warnings. Wyeth complied with the FDA's specifications, and in 1995, the FDA approved Prempro.

Brockert used Prempro from 1997 until 2001, when she was diagnosed with breast cancer. At all times Brockert used Prempro, it was accompanied by the FDA-approved warning label.

In 2002, the Women's Health Initiative ("WHI"), sponsored by the National Institutes of Health, released certain findings of a large clinical trial studying the use of Prempro to assess its risks and benefits in preventing heart disease and other conditions. The study was stopped early because the evidence of the risk of breast cancer and cardiovascular events outweighed the evidence of its benefits. Wyeth then changed the Prempro warning label to include the WHI findings and to, strengthen the warnings addressing the risk of breast cancer, cardiovascular effects, and other conditions. The FDA approved the revised warning labels. According to Wyeth, Prempro remains on the market, and the exact dosage Brockert was prescribed is still available and prescribed today.

Brockert alleges that, had Wyeth properly tested Prempro as the WHI did, it would have refrained from introducing the drug to the market or would have adopted the post-WHI label accepted by the FDA far earlier. Under either scenario, she contends that more likely than not she would have remained cancer-free.

Wyeth filed two motions for summary judgment. The first motion addressed Brockert's failure-to-warn claims. The second addressed all other claims. The trial court granted the first motion for summary judgment on federal preemption grounds, and did not specify the grounds for granting the second motion for summary judgment. On appeal, Brockert has abandoned all her claims except her failure-to-warn and design-defect claims.

Analysis

Brockert first challenges the trial court's grant of both Wyeth's first and second summary judgments. Brockert frames her first issue as "[w]hether a claim that a drug manufacturer should have studied a product more thoroughly and issued stronger warnings earlier is preempted by FDA regulations when the very warnings Plaintiff advocates are warnings the FDA approved when a third party finally performed the study the defendant should have undertaken, meaning there is no conflict between the tort claim and the FDA action." Brockert describes her second issue as "[w]hether a claim that estrogen combined with progestin is unreasonably dangerous and defective because it is more harmful than estrogen alone can serve as the basis for a design defect claim under Texas law." Both of Wyeth's summary-judgment motions were hybrid traditional and no-evidence motions. See Tex. R. Civ. P. 166a(c), (i). We therefore apply the established standards of review appropriate for each. See Joe v. Two Thirty Nine Joint Venture, 145 S.W.3d 150, 156-57 (Tex. 2004); Ford Motor Co. v. Ridgway, 135 S.W.3d 598, 600-601 (Tex. 2004).

I. Brockert's Failure-to-Warn Claims

Brockert contends that her claims are not preempted because the warnings she advocates are precisely the warnings the FDA approved after the WHI performed the study Wyeth failed to do, and therefore no conflict exists between her claims and the FDA regulations. Brockert also dismisses the FDA's own position, recently advanced in a preamble to a 2006 revision of the FDA's labeling rules, that federal law preempts state-law claims conflicting with or contrary to FDA-approved labeling. Courts in some jurisdictions have deferred to the FDA's position as stated in the preamble to hold that state-law failure-to-warn claims are preempted, while other courts have rejected the preamble as a basis for preempting state-law claims. Brockert contends the preamble either does not apply or, if it does apply, it is entitled to no deference from the court.

As discussed in greater detail below, while this appeal was pending, the United States Supreme Court issued Wyeth v. Levine, ___ U.S. ___, 129 S.Ct. 1187, 173 L.Ed.2d 51 (2009), in which the Court held that federal law did not preempt a plaintiffs claims that the FDA-approved warning label on the drug Phenergan, also made by Wyeth, inadequately warned of the drug's risks. Because of the Wyeth v. Levine decision, Wyeth has informed this court that it is "withdrawing" the preemption argument advanced in its brief. We therefore address Brockert's issue without reference to Wyeth's appellate argument.

A. Preemption

Federal preemption of state law is derived from the Supremacy Clause of the United States Constitution: "This Constitution, and the Laws of the United States . . . shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any State to the Contrary notwithstanding." U.S. Const., art. VI, cl. 2. Thus, when a state law conflicts with federal law, it is preempted and has no effect. Maryland v. Louisiana, 451 U.S. 725, 746, 101 S.Ct. 2114, 68 L.Ed.2d 576 (1981); BIC Pen Corp. v. Carter, 251 S.W.3d 500, 504 (Tex. 2008). In fields the states have traditionally occupied, such as health and safety regulation, there is a strong presumption against federal preemption. See Hillsborough County, Fla. v. Automated Med. Labs. Inc., 471 U.S. 707, 715, 105 S.Ct. 2371, 85 L.Ed.2d 714 (1985). To overcome the general presumption that Congress did not intend to preempt state law, Congress's intent to preempt must be "clear and manifest." See id. at 715-16, 105 S.Ct. 2371; Graber v. Fuqua, 279 S.W.3d 608, 611-13 (Tex. 2009).

State laws may conflict with federal laws and be preempted in three ways: (1) `the federal law may expressly preempt state law; (2) federal laws or regulations may impliedly preempt state law or regulations if the statute's scope indicates that Congress intended federal law or regulations to occupy the field exclusively; or (3) state law is impliedly preempted if it actually conflicts with federal law or regulations because it is either impossible for a private party to comply with both state and federal requirements or the state law obstructs accomplishing and executing Congress's full purposes and objectives. See Graber, 279 S.W.3d at 610-11 (citing Crosby v. Nat'l Foreign Trade Council, 530 U.S. 363, 372, 120 S.Ct. 2288, 147 L.Ed.2d 352 (2000)); BIC Pen Corp., 251 S.W.3d at 504. Here, we are concerned with only the third basis for preemption — an actual conflict with federal law or regulations.

Congress delegated to the FDA the authority to promulgate regulations to enforce the Food, Drug Cosmetic Act ("FDCA"). See 21 U.S.C. § 371(a); see also Medtronic, Inc. v. Lohr, 518 U.S. 470, 496, 116 S.Ct. 2240, 135 L.Ed.2d 700 (1996) ("FDA is the federal agency to which Congress has delegated its authority to implement the provisions of the [FDCA]."). The FDA has exercised its delegated authority to promulgate comprehensive prescription-drug-labeling regulations. See 21 C.F.R. 201.56-.57 (2006).

A drug manufacturer like Wyeth must submit a New Drug Application to the FDA and that NDA must be approved before a new drug can be marketed. 21 U.S.C. § 355(a). If the FDA finds that, among other requirements, there is "substantial evidence" that the drug is "safe for use under the conditions prescribed, recommended or suggested in the proposed labeling," and that "the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the . . . proposed labeling," then the FDA will grant the NDA. 21 U.S.C. § 355(d).

In the NDA, a manufacturer like Wyeth is also required to provide proposed labeling for the FDA's consideration. 21 U.S.C. § 355(b)(1)(F). The FDA must determine that the label is not "false or misleading in any particular." 21 U.S.C. § 355(d). After a drug is approved, a manufacturer generally cannot change the labeling unless it first submits a supplemental application to the FDA and obtains approval for the revision. 21 C.F.R. 314.70(b)(2)(v)(A)(2008). In some instances, the FDA permits changes to be made by submitting a supplement to the labeling, which the manufacturer may implement before the FDA approves it. Id. at 314.70(c). Such changes include changes in the labeling to reflect newly acquired information to add or strengthen a contraindication, warning, precaution, or adverse reaction. Id. at 314.70(c)(6)(iii)(A). If a manufacturer makes a change before receiving FDA approval, the agency may later reject the change and order the manufacturer to cease distribution of the product. Id. at 314.70(c)(7).

In its 2006 preamble to its revised regulations, the FDA explicitly stated its intent to preempt certain state-law failure-to-warn claims as conflicting with federal regulatory objectives. See Requirements on Content and Format of Labeling for Human Prescription Drug and Biological Products, 71 Fed. Reg. 3922, 3934 (Jan. 24, 2006). In the preamble, the FDA states that it "believes that State laws conflict with and stand as an obstacle to achievement of the full objectives and purposes of Federal law when they purport to compel a firm to include in labeling or advertising a statement that FDA has considered and found scientifically unsubstantiated." Id. at 3935. The FDA also expresses its disagreement with those cases viewing FDA-labeling requirements as a minimum standard that may be supplemented by state law: "FDA interprets the act to establish both a `floor' and a `ceiling,' such that additional disclosures of risk information can expose a manufacturer to liability under the act if the additional statement is unsubstantiated or otherwise false or misleading." Id. at 3934-35.

Since the FDA published the preamble, federal courts that have considered its effect have reached differing conclusions. For example, a number of courts have deferred to the FDA's interpretation of its own labeling regulations as having some preemptive effect and dismissed state-law failure-to-warn claims. See, e.g., Sykes v. Glaxo-SmithKline, 484 F.Supp.2d 289, 317 (E.D.Pa. 2007); Colacicco v. Apotex, Inc., 432 F.Supp.2d 514, 530, 537 (E.D.Pa. 2006), aff'd, 521 F.3d 253 (3d Cir. 2008), vacated, ___ U.S. ___, 129 S.Ct. 1578, 173 L.Ed.2d 672 (2009); Dobbs v. Wyeth Pharmaceuticals, 530 F.Supp.2d 1275, 1288 (W.D.Okla. 2008); In re Bextra and Celebrex Mktg. Sales Practices and Prod. Liab. Litig., No. M: 05-1699 CRB, 2006 WL 2374742, at *8-9 (N.D.Cal. Aug. 16, 2006). Others have refused to defer to the FDA and concluded that its labeling regulations have no preemptive effect. See, e.g., In re Vioxx Prods. Liab. Litig., 501 F.Supp.2d 776, 788 (E.D.La. 2007); In re Zyprexa Prods. Liab. Litig., 489 F.Supp.2d 230, 273-74 (E.D.N.Y. 2007); Perry v. Novartis Pharm. Corp., 456 F.Supp.2d 678, 683-84 (E.D.Pa. 2006); Jackson v. Pfizer, Inc., 432 F.Supp.2d 964, 968 (D.Neb. 2006).

But the United States Supreme Court has recently resolved the differing opinions about the deference to be accorded the preamble. See Wyeth v. Levine, ___ U.S. ___, 129 S.Ct. 1187, 173 L.Ed.2d 51 (2009). This case also informs our resolution of Brockert's issue. In Wyeth v. Levine, Diana Levine lost her arm to gangrene after Phenergan, an anti-nausea drug, was injected intravenously using a procedure known as an "IV push." Id. at 1191. Levine alleged that Phenergan's labeling was defective because, even though it included warnings about the risks of intravenous administration, it should have instructed clinicians not to use the IV-push method, as other, safer options were available. See id. at 1191-92. A jury found that Phenergan was a defective product because its warnings and instructions were inadequate. Wyeth appealed, arguing that FDA regulations directly preempted Levine's claims. Id. at 1193. The Vermont Supreme Court affirmed the judgment below, holding that the jury's verdict did not conflict with the FDA's labeling requirements because Wyeth could have warned against IV-push administration without prior FDA approval, and because federal-labeling requirements "create a floor, not a ceiling" for state regulation. Id. The United States Supreme Court granted Wyeth's petition for certiorari to consider whether the FDA's labeling requirements provide Wyeth with a complete defense to Levine's tort claims. Id. at 1191.

Before the Supreme Court, Wyeth argued that Levine's claims are preempted because it would have been impossible for it to comply with a state-law duty to modify its labeling without violating federal law. Id. at 1193-94. Wyeth further argued that recognition of Levine's state-law tort action creates a unacceptable obstacle to the accomplishment and execution of the full purposes and objectives of Congress because it substitutes a lay jury's decision about drug labeling for the expert judgment of the FDA. Id. at 1194. In advancing the latter argument, Wyeth relied on the FDA's declaration in the 2006 preamble that state-law failure-to-warn claims are preempted. Id. at 1200.

In support of its first argument, Wyeth contended that it would be impossible for it to comply with both state-law and federal-labeling duties because it could only have changed Phenergan's label in response to new information that the FDA had not previously considered. See id. at 1196. However, the Court rejected this argument, noting that the FDA itself had explained that newly acquired information is not limited to new data, but also encompasses "`new analyses of previously submitted data.'" Id. at 1197 (citing 73 Fed. Reg. 49609, 49604). The Court explained that "[t]he rule accounts for the fact that risk information accumulates over time and that the same data may take on a different meaning in light of subsequent developments." Id. Thus, the Court concluded, as Wyeth became aware of information about amputations resulting from Phenergan injections, it "could have analyzed the accumulating data and added a stronger warning about IV-push administration of the drug." Id. Further, the Court rejected Wyeth's claim that unilaterally changing the warning could have subjected it to liability for misbranding and unauthorized distribution under the FDA regulations, noting that "it has remained a central premise of federal drug regulation that the manufacturer bears responsibility for the content of its label at all times." Id. at 1197-98. Absent clear evidence that the FDA would not have approved a change to Phenergan's label, the Court refused to conclude that it was impossible for Wyeth to comply with both federal and state requirements. Id. at 1198-99. Cautioning that "[i]mpossibility pre-emption is a demanding defense," the Court held that "the mere fact that the FDA approved Phenergan's label does not establish that it would have prohibited such a change." Id. at 1199.

The Supreme Court also rejected Wyeth's second argument that requiring it to comply with a state-law duty to provide a stronger warning would obstruct the purposes and objects of federal drug-labeling regulation, concluding instead that "all evidence of Congress's purposes" was contrary to this argument. Id. at 1199. The Court noted that, although Congress provided an express-preemption provision for medical devices, it did not enact a similar provision for prescription drugs, and it considered this "powerful evidence" that "Congress did not intend FDA oversight to be the exclusive means of ensuring drug safety and effectiveness." Id. at 1200.

Further, the Court dismissed Wyeth's argument that the 2006 preamble requires that courts defer to the FDA's judgment in "perform[ing] a precise balancing of risks and benefits" and "establish[ing] a specific labeling standard that leave no room for different state-law judgments." See id. at 1200. The Court explained that the preamble was not an agency regulation with the force of law such that it could preempt conflicting state law, but was only a "mere assertion" of the FDA's position; therefore, "[t]he weight we accord the agency's explanation of state law's impact on the federal scheme depends on its thoroughness, consistency, and persuasiveness." Id. at 1201 (citing United States v. Mead Corp., 533 U.S. 218, 234-35, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001); Skidmore v. Swift Co., 323 U.S. 134, 140, 65 S.Ct. 161, 89 L.Ed. 124 (1944)). Applying this standard, the Court concluded that the 2006 preamble "does not merit deference" because when the FDA first proposed the new rule in December 2000, it represented that the rule would not contain policies implicating federalism or preempting state law; yet, in 2006 when the rule was finalized, without notice or opportunity for comment, the preamble "articulated a sweeping position on the FDCA's pre-emptive effect." Id. The Court determined that this procedural failure rendered the agency's views "inherently suspect." Id. Further, the Court determined that the preamble "is at odds with what evidence we have of Congress' purposes, and it reversed the FDA's own longstanding position without providing a reasoned explanation." Id. The Court ultimately concluded that "Congress has repeatedly declined to pre-empt state law, and the FDA's recently adopted position that state tort suits interfere with its statutory mandate is entitled to no weight." Id. at 1204. Accordingly, the Court affirmed the Vermont Supreme Court's judgment. Id.

B. Brockert's Failure-To-Warn Claims Are Not Preempted

Against this backdrop, we turn to our analysis of Brockert's issue, recognizing that the parties did not have the benefit of the Supreme Court's decision in Wyeth v. Levine when they made their arguments. Brockert argues that no conflict exists with FDA regulations, because the warnings she advocates are the very warnings the FDA actually approved. Further, Brockert distinguishes this case from others in which preemption was found when the FDA considered and rejected stronger warnings, because the warnings she advocates were based on newly discovered evidence.

In light of Wyeth v. Levine, we must agree with Brockert. Below, Wyeth had argued that a direct conflict existed because it would have been impossible for it to comply with both the FDA's labeling requirements and the stronger warnings Brockert advocates. In the absence of evidence that the FDA would not have approved a change to the labeling at issue, the Wyeth Court rejected this "impossibility" argument. See id. at 1198-99. The Court instructed that the mere fact that the FDA approved a prescription drug's label does not establish that it would have prohibited a change in response to new evidence of a risk. Id. at 1199. The Wyeth Court also resoundingly rejected the argument that the FDA's 2006 preamble was entitled to deference, instead concluding that the preamble was entitled to no deference whatsoever, as it was "inherently suspect." Id. at 1201, 1204.

Additionally, in reaching its conclusions, the Supreme Court dismissed Wyeth's argument that labeling changes were permitted only to reflect "newly acquired information" the FDA had not previously considered. As the Court explained, the FDA's own regulations interpret "newly acquired information" to include new analyses of previously submitted data. See id. at 1196-97. Here, Brockert argues that Wyeth should have engaged in the type of testing eventually done by the WHI that revealed the greater risk, and that Wyeth should have adequately labeled Prempro to warn of this risk. Further, Brockert maintains that, if adequately warned, she would not have used Prempro and more likely than not would not have acquired breast cancer. Given the Wyeth Court's statement that a "central premise" of federal-drug regulation is that "the manufacturer bears responsibility for the content of its label at all times," and the lack of evidence that the FDA would have rejected a stronger warning concerning the risk of breast cancer, we cannot affirm the trial court's dismissal of Brockert's failure-to-warn claim at the summary-judgment stage based on federal conflict preemption.

We therefore sustain Brockert's first issue.

II. Brockert's Design-Defect Claims

Wyeth's second motion for summary judgment addressed all claims other than the failure-to-warn claims. On appeal, Brockert challenges the trial court's grant of this summary-judgment motion only as to her design-defect claims.

Relevant here, Wyeth alleged that it was entitled to summary judgment on Brockert's design-defect claims because (1) in Texas, the Restatement (Second) of Torts, section 402A, comment k, applies to exempt FDA-approved prescription drugs from strict-liability claims for design defect; and (2) Brockert has no evidence of a safer alternative design. Because we hold that the trial court correctly granted summary judgment on the second basis, we do not address the first.

A design defect renders a product unreasonably dangerous as designed, taking into consideration the utility of the product and the risk involved in its use. Gen. Motors Corp. v. Sanchez, 997 S.W.2d 584, 588 (Tex. 1999). A plaintiff must prove that there is a safer alternative design to recover under a design-defect theory. Id.; Caterpillar, Inc. v. Shears, 911 S.W.2d 379, 384 (Tex. 1995). In the absence of a safer alternative, a product is not unreasonably dangerous as a matter of law. Caterpillar, 911 S.W.2d at 384.

Brockert argues that she has presented substantial evidence of a safer alternative design. Specifically, she contends that the safer alternative design to estrogen in combination with progestin is estrogen alone. As Brockert acknowledges, this alleged alternative design already exists in prescription drugs like Premarin, also made by Wyeth. According to Brockert, estrogen with progestin is defective because the overall risk of breast cancer to users outweighs the overall risk of estrogen alone to the endometrium. Wyeth responds that Premarin (estrogen only) is not an alternative design for Prempro (estrogen with progestin), but instead is a completely different prescription drug intended for a different population of women.

As evidence of a safer alternative design, Brockert presented the testimony of Dr. Daniel Lehane, a medical oncologist whose practice, emphasizes the treatment of breast cancer. Dr. Lehane testified that estrogen alone is a safer product than estrogen with progestin, and that the addition of progestin to estrogen makes hormone therapy unreasonably dangerous by substantially increasing the risk of breast cancer. He also opined that it is more likely than not that Brockert would not have acquired breast cancer had she used estrogen alone and ceased all hormone therapy use shortly after the WHI results were announced. Further, Dr. Lehane opined that the benefits attributed to hormone therapy are achieved by estrogen alone, and not the progestin, explaining that progestin was added to reduce the incidence of endometrial hyperplasia. Dr. Lehane also testified that the increased risk of breast cancer combination-hormone therapy generates outweighs the reduction in the risk of endometrial cancer associated with estrogen alone.

The Texas Supreme Court has held that a plaintiff cannot prove design defect by claiming that defendant should have sold an entirely different product. See Caterpillar, 911 S.W.2d at 384-85. In Caterpillar, the allegedly defective product was a front-end loader with a rollover-protective structure ("ROPS") that was removable, which allowed the loader to be used in low-clearance areas. The plaintiffs design-defect expert testified that the front-end loader should have been configured so that the ROPS was not removable or so that the removal of the ROPS would render the loader inoperable. The supreme court reversed a jury verdict of design defect, holding that this was "no evidence . . . of a safer alternative design for a front-end loader that could fulfill the multi-purpose role of Caterpillar's model 920 with a removable ROPS." Id. at 384. The court rejected the plaintiffs argument that the defendant should have changed the product in such a way that it was essentially transformed into a different product:

A motorcycle could be made safer by adding two additional wheels and a cab, but then it is no longer a motorcycle. A convertible can be made safer by fully enclosing the cab, but then it is just an ordinary car. The law of products liability demands that manufacturers and distributors take feasible steps to make their products reasonably safe. It is not rational, however, to impose liability in such a way as to eliminate whole categories of useful products from the market.

Id. at 385; see also Am. Tobacco Co., Inc. v. Grinnell, 951 S.W.2d 420, 433 n. 10 (Tex. 1997) (granting summary judgment to defendant because there was no safer alternative design for cigarettes; "[c]ategorical liability is not only an unworkable solution, but also a position repeatedly rejected by courts.").

The Fifth Circuit has also found that a plaintiff cannot prove that a safer alternative design exists by pointing to a substantially different product, even when the other product has the same general purpose as the allegedly defective product. See Theriot v. Danek Med., Inc., 168 F.3d 253 (5th Cir. 1999) (applying Louisiana law). In Theriot, the court held that a plaintiff who contended that pedicle screws were defectively designed was required to demonstrate a safer alternative design that involved pedicle screws, and that the plaintiff was not permitted to point to other products intended to provide biomechanical stability, such as internal systems using hooks or wires, or external neck braces. Id. at 255.

Thus, a safer alternative design must be one for the product at issue — here, Prempro. As Dr. Lehane points out in his affidavit, in prescription drugs like Prempro, "synthetic progestin was added to estrogen to reduce the incidence of endometrial hyperplasia." But Brockert does not explain how Prempro could have been modified or improved; she instead argues that progestin should not have been added to estrogen. In essence, Brockert argues that the product Prempro should have been a different product: its predecessor Premarin. But, as the supreme court has explained, Texas law does not recognize this sort of categorical attack on a product. See Caterpillar, 911 S.W.2d at 384-85.

Finally, Wyeth contends that Brockert's failure to raise a fact issue concerning a safer alternative design disposes of all of her design-defect claims, including her negligent-design-defect claim. We agree. See Toshiba Intern. Corp. v. Henry, 152 S.W.3d 774, 784-85 n. 3 (Tex.App.-Texarkana 2004, no pet.) ("When a plaintiff seeks recovery because of negligence or a theory of strict liability in tort, the burden is on the plaintiff to prove that the injury resulted from a defect in the product.").

Therefore, we overrule Brockert's second issue.

Conclusion

We hold that Brockert's failure-to-warn claims, as alleged, are not preempted by federal-labeling regulations, and we therefore sustain her first issue. But we overrule Brockert's second issue because we hold that she failed to present any evidence of a safer alternative design to support her design-defect claims. We remand the case for further proceedings consistent with this opinion.