Full opinion text
RULINGS ON DEFENDANTS’ MOTIONS IN LIMINE AND ORDER GRANTING DEFENDANTS’ MOTION FOR SUMMARY ADJUDICATION MATZ, District Judge. TABLE OF CONTENTS I. Introduction.886 II. Background.888 III. Legal and Scientific Principles Applicable to the Expert Testimony in This Case.889 A. Qualifications .889 B. Reliability.889 1. Animal Studies .890 2. Differential Diagnosis.892 3. Epidemiological studies.892 C. Usefulness.893 IV. Dr. Richard Neugebauer (epidemiologist).894 A. Proposed Testimony.894 B. Qualifications .895 C. Reliability.895 1. Methodological flaws .895 2. “Suggestive” Evidence.898 D. Usefulness.898 E. Conclusion.899 V. Dr. Christopher Batich (polymer chemist).899 A. Proposed Testimony.899 B. Qualifications .901 1. Biodegradation of PUF-Coated Implants.901 2. Duty of Care.901 C. Reliability and Usefulness .902 D. Conclusion.903 VI.Dr. Marc Lappé (toxicologist). A. Proposed Testimony. B. Qualifications . C. Reliability. 1. Carcinogenic Properties of PUF vs. TDA 2. Animal Studies . D. Usefulness. E. Conclusion. VII. Dr. Douglas Shanklin (pathologist).913 A. Proposed Testimony.1.913 B. Qualifications .915 C. Reliability.916 1. General Causation. 916 2. Specific Causation.916 D. Usefulness.921 E. Conclusion.921 VIII. Defendants are Entitled to Summary Adjudication on Bradley Cagle’s Claims .921 A. Summary Judgment Standard.921 B. Analysis.922 IX. Conclusion.923 I. Introduction Toni Cagle was diagnosed with breast cancer approximately fourteen months after receiving breast implants. Her breast cancer later caused her death. Her husband Bradley Cagle (“Plaintiff’), as the administrator of Toni Cagle’s estate, alleges that her implants manufactured by The Cooper Companies, Inc., Surgitek/Medical Engineering Corp. (a subsidiary of Bristol Myers Squibb Co.), and Foamex, L.P. (“Defendants”) caused or accelerated her breast cancer. He alleges that Defendants are either the manufacturers or the successors in interest to the manufacturers of the devices implanted in Mrs. Cagle. His First Cause of Action is for “Strict Products Liability.” The others are as follows: (2) “Failure to Warn”; (3) Breach of Implied Warranty; (4) Breach of Express Warranty; (5) Fraud/Intentional Misrepresentation; (6) Negligence; (7) Loss of Consortium; (9) Wrongful Death; and (10) Pain and Suffering. This Order addresses four motions in limine filed by the Defendants and Defendants’ motion for summary judgment. The in limine motions seek to exclude the testimony of Plaintiffs four causation experts. Those experts are Dr. Neugebauer (an epidemiologist), Dr. Batich (a polymer chemist), Dr. Lappé (a toxicologist) and Dr. Shanklin (a pathologist). Defendants’ motion for summary judgment argues that because Plaintiffs experts are not qualified and the science they rely on is unsound, Plaintiff cannot demonstrate that Cagle’s implants caused or accelerated her breast cancer. In his Opposition to the summary judgment motion, Plaintiffs attempt to demonstrate a genuine issue about causation is based exclusively on the experts and scientific data at issue in Defendants’ motions in limine. Therefore, whether plaintiff can defeat the summary judgment motion turns on the extent to which Defendants’ motions in limine are meritorious. This case is different from most other breast implant cases, because Plaintiff is not alleging that Cagle’s cancer was caused by silicone. Cagle was one of a relatively small number of women whose implants were coated with a polyurethane foam (“PUF”). Plaintiff alleges that PUF breaks down in vivo into 2,4-toluene dia-mine (“TDA”) (also called 2,4-diaminoto-luene), which he claims is carcinogenic. Am. Compl. ¶43. Plaintiff alleges that TDA from Cagle’s implants caused or accelerated the progression of her disease. Id. ¶¶ 40, 48. More specifically, Plaintiffs causation theory is as follows (the relevant expert is indicated in parentheses): (a)One epidemiological study provides “suggestive evidence” of a causal link between PUF-coated implants and cancer (Neugebauer); (b) The polyurethane coating of PUF-coated implants biodegrades after implantation in humans (Batich); (c) The degradation products of the PUF-coating include TDA (Batich); (d) TDA is known to be carcinogenic in animals and is a “probable” human carcinogen (Lappé and Shanklin); (e) The amount of TDA likely released from Cagle’s implants, Cagle’s pregnancy (which began almost immediately after implantation) and the rare type of breast cancer Cagle suffered renders it more likely than not that her tumor was caused or its growth accelerated by TDA released from her implants (Lappé and Shanklin). Having analyzed Defendants’ challenges to the expert testimony of Lappé and Shanklin, I conclude that Plaintiff is unable to offer scientifically reliable evidence to support proposition (e). Therefore, even assuming that the evidence proffered to support propositions (a) through (d) is admissible, summary adjudication is appropriate, because Plaintiff cannot establish that breast implants caused Cagle’s cancer. This Order nevertheless addresses the content and admissibility of the evidence proffered to support propositions (a) through (d), in the event that on appeal the analysis and conclusion concerning proposition (e) is deemed incorrect. Section II provides the relevant background facts specific to this case. Section III provides an overview of the legal and scientific principles applicable to evaluating the relevant scientific evidence. In Sections IV through VII, I evaluate each expert’s qualifications and testimony based on the legal and scientific principles elaborated in Section III. Finally, Section VIII concludes that because Plaintiffs expert testimony that Cagle’s cancer was caused by her implants is not admissible, summary adjudication in favor of Defendants is appropriate. II. Background Defendants either manufactured or are successors in interest to companies that manufactured “Meme” polyurethane foam (“PUF”) coated breast implants. After the FDA requested data from the manufacturers about the chemical composition of PUF and safety testing data on that foam, the manufacturers suspended shipments of polyurethane-covered breast implants in May 1991. Lappé Opp. Exh. A at 8 (“Lappé Expert Report”). At that point, the FDA estimated that approximately ten percent of women with implants had the PUF-coated type. Id,.; Kern et al., Carcinogenic Potential of Silicone Breast Implants: A Connecticut Statewide Study, 100 Plastic & Reconstructive Surg. 737 (1997) (SJ Mot. Exh. 22). On May 2,1989, Toni Cagle, a registered nurse, received two PUF-coated breast implants. SJ Opp. Exh. 3 at 17:20-23, 22:1-3, 25:11-13, 26:1-6 (Cagle Depo.); Am. Compl. ¶38. Cagle was 30 years old at the time. Cagle testified that before receiving the implants she had been “overly cautious” in conducting self-breast examinations. She testified that she had examined her breasts every two to three weeks and that she had never detected any lumps or other problems with her breasts. SJ Opp. Exh. 3 at 28:16-29:12 (Cagle Depo.) Before receiving her implants, Cagle’s plastic surgeon examined her breasts and reported that he had found no masses. Id. at 28:1-5; Shanklin Opp. Exh. F at 67 (Patient History and Physical Examination Report). After Cagle received her implants, she continued to routinely examine her breasts. SJ Opp. Exh. 3 at 29:13-16 (Ca-gle Depo.). Within one week of receiving her implants, Cagle became pregnant. SJ Opp. Exh. 3 at 27:4-5 (Cagle Depo). On June 16, 1989, Cagle’s obstetrician conducted a prenatal physical examination and found no tumors in Cagle’s breasts. Shanklin Opp. Exh. G at 70-71 (Prenatal Record). At the end of June or the beginning of July 1990 — fourteen months after receiving her implants — Cagle discovered a lump in her right breast during one of her routine breast self-exams. Shanklin Opp. Exh. H at 113:21-114:15 (Maguire Depo.). She obtained a biopsy 451 days after receiving her implants on July 26, 1990. Shanklin Opp. Exh. F at 69. The resulting pathology report, dated July 30, 1990, reported “two portions of bright yellow firm tissue measuring 2.5 x 1.5 x 1.3 cm and 3.5 x 2.3 x 1.3 cm.” Id. Exh. J at 83. The diagnosis was infiltrating ductal carcinoma, the most common type of breast cancer, accounting for up to eighty percent of all breast cancers. Id.; Shanklin MIL Exh. 3 at 163:3-164:7 (Brusca Daubert Hearing) & Exh. 4 at 28:8-20 (Bylund Depo.). By the time the lump was diagnosed, the cancer had metastasized and spread to her lymph nodes. SJ Opp. Exh. 3 at 39:5-7 (Cagle Depo.). She underwent a mastectomy, radiation, and chemotherapy, id. at 35:11-12, but her cancer recurred and she died on July 26, 1993, at the age of 33—approximately 4 years after receiving the implants and 3 years after being diagnosed with breast cancer. Shanklin Opp. at 4. Cagle’s family has a history of cancer. Cagle testified that she had a maternal half-aunt who contracted breast cancer and had a mastectomy, SJ Opp. Exh. 3 at 9:13-16 (Cagle Depo.), that her mother died of kidney cancer, id. at 8:17-19, and a paternal aunt contracted stomach cancer. Id. at 11:6-17. One of Cagle’s experts, Dr. Lappé, has also reported that Cagle’s maternal grandfather had a gastric carcinoma (stomach cancer) and a paternal uncle had some unspecified type of cancer. Lappé Expert Report at 6. III. Legal and Scientific Principles Applicable to the Expert Testimony in This Case Once someone is qualified as a scientific expert, his testimony is generally admissible under Federal Rule of Evidence (“FRE”) 702 if it satisfies two criteria. First, the testimony must reflect “scientific knowledge ... derived by the scientific method.” Daubert v. Merrell Dow Pharms. Inc., 43 F.3d 1311, 1315 (9th Cir.), cert. denied 516 U.S. 869, 116 S.Ct. 189, 133 L.Ed.2d 126 (1995) (“Daubert II”) (quoting Daubert v. Merrell Dow Pharms. Inc., 509 U.S. 579, 590, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993) (“Daubert I ”)). This requirement “establishes a standard of evidentiary reliability.” Daubert I, at 590, 113 S.Ct. 2786. Second, the proposed expert testimony must be “relevant to the task at hand,” meaning that it “logically advances a material aspect of the proposing party’s ease.” Daubert II, at 1315 (quoting Daubert I, at 597, 113 S.Ct. 2786). The proponent of the expert testimony has the burden of establishing by a preponderance of the evidence that the admissibility requirements are met. Fed. R.Evid. 702 Advisory Committee’s Notes. A trial court’s decision to admit or exclude expert testimony is reviewed under an abuse of discretion standard. Gen. Elec. Co. v. Joiner, 522 U.S. 136, 138-139, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997). A. Qualifications A witness can qualify as an expert on the basis of “knowledge, skill, experience, training, or education,” Fed. R.Evid. 702; see also U.S. v. Cambindo Valencia, 609 F.2d 603, 640 (2d Cir.1979), and such qualifications are construed broadly. Thomas v. Newton Int’l Enterprises, 42 F.3d 1266, 1269 (9th Cir.1994); Pride v. BIC Corp., 218 F.3d 566, 577 (6th Cir.2000). A court abuses its discretion when it excludes expert testimony solely on the ground that the witness’s qualifications are not sufficiently specific if the witness is generally qualified. In re Paoli R.R. Yard PCB Litig., 35 F.3d 717 (3d Cir.1994) (abuse of discretion for trial court to preclude trained internist with broad experience in field of toxic substances, who had spent significant time reading literature on the effect of PCBs on human body, from testifying as to whether PCBs caused illness in plaintiffs, even though witness lacked expertise in other, more relevant, specialized fields). A lack of specialization affects the weight of the expert’s testimony, not its admissibility. Holbrook v. Dykes Bros. S.S. Co., 80 F.3d 777, 782 (3d Cir.1996). B. Reliability Although there is a presumption of admissibility, Daubert I, supra, at 588, 113 S.Ct. 2786, FRE 702 imposes a “gatek-eeping” duty on district courts to ensure that testimony based on scientific, technical, or other specialized knowledge rests on a reliable foundation. Id. at 597, 113 S.Ct. 2786; Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 141-42, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). “[T]he trial judge in all cases of proffered expert testimony must find that it is properly grounded, well-reasoned, and not speculative before it can be admitted.” Fed.R.Evid. 702 Advisory Committee’s Notes. “The trial court’s gatekeeping function requires more than simply ‘taking the expert’s word for it.’ ” Id. (citing Daubert II, supra, at 1319). In addition, “any step that renders [the expert’s] analysis unreliable ... renders the expert’s testimony inadmissible. This is true whether the step completely changes a reliable methodology or merely misapplies that methodology.” In re Paoli, supra, at 745. In Daubert I, the Supreme Court articulated the following factors that bear on the reliability inquiry: (1) whether the theory or technique used by the expert can be or has been tested, (2) whether the theory or technique has been subjected to peer review and publication, (3) the known or potential rate of error of the technique or theory when applied, and (4) the “general acceptance” of the theory or technique in the scientific community. Daubert I, supra, at 593-94, 113 S.Ct. 2786. These factors are not definitive or exclusive of others. Id. In addition, courts have also found the following factors relevant in assessing the reliability of expert testimony: (1) whether the expert is proposing to testify about matters growing directly out of independent research he or she has conducted or whether the opinion was developed expressly for purposes of testifying; (2) whether the expert has unjustifiably extrapolated from an accepted premise to an unfounded conclusion; (3) whether the expert has adequately accounted for obvious alternative explanations; (4) whether the expert is being as careful as he would be in his regular professional work; and (5) whether the field of expertise claimed by the expert is known to reach reliable results for the type of opinion offered. Fed. R.Evid. 702 Advisory Committee’s Notes. At issue in these motions in limine and the summary judgment motion is whether Plaintiffs experts can present reliable evidence of causation. To prevail on his claims, Plaintiff must show both general or generic causation (ie., that PUF implants have the capacity to cause breast cancer in humans) and specific causation (ie., that Cagle’s breast cancer was caused by her PUF implants). In re Hanford Nuclear Reservation Litig., 292 F.3d 1124, 1133-34 (9th Cir.2002). To demonstrate general and specific causation, Cagle’s experts rely on animal studies, differential diagnosis and epidemiological studies. The Court will discuss each such category of evidence. 1. Animal Studies Animal studies may be admissible to demonstrate general causation. According to the Reference Manual on Scientific Evidence (Federal Judicial Center 2d ed. 2000) (“Ref. Manual”), in extrapolating from animal data, “one can usually rely on the fact that a compound causing an effect in one mammalian species will cause it in another species. This is a basic principle of toxicology and pharmacology.” Id. at 410. Indeed, extrapolations to humans from animal experiments that involve significantly higher doses of the agent at issue are commonly used in the regulatory arena. Id. at 409. See also Cornell University, Program on Breast Cancer and Environmental Risk Factors in New York State, Fact Sheet No. 45 at p. 49, Environmental Chemicals and Breast Cancer Risk (2002) (“Animal studies... are important to help predict cancer risk when human studies are unavailable.”) When animal studies are offered to demonstrate causation in a tort case, experts also provide additional information to justify the extrapolation to humans. Ref. Manual at 409. Expert opinions based on animal data have been excluded where the expert did not review similarities and differences between humans and the animal species in which the compound was tested. E.g., Gen. Elec. Co. v. Joiner, supra, at 144, 118 S.Ct. 512 (district court did not abuse its discretion in ruling inadmissible expert testimony based on “seemingly far-removed” animal studies where party failed to explain why the extrapolation was scientifically proper); Domingo v. T.K., 289 F.3d 600, 606-607 (9th Cir.2002) (finding that district court’s exclusion of expert’s causation testimony was not an abuse of discretion, in part because expert did not provide “analytical support” for his extrapolation of animal study results to humans and also because there was no evidence that the expert had applied a valid scientific method in developing his theory and there were unexplained gaps between the expert’s premises and his conclusion); Turpin v. Merrell Dow Pharms., Inc., 959 F.2d 1349, 1360 (6th Cir.1992) (holding that expert’s testimony was inadmissible where the record failed to make clear why the effects of Bendectin in rats and rabbits could be extrapolated to humans); Hall v. Baxter Healthcare Corp., 947 F.Supp. 1387, 1410 (D.Or.1996) (“Extrapolations of animal studies to human beings are generally not considered reliable in the absence of a scientific explanation of why such extrapolation is warranted.”). Animal studies are not generally admissible where contrary epidemiological evidence in humans exists. See Richardson v. Richardson-Merrell, Inc., 857 F.2d 823, 830 (D.C.Cir.1988), cert. denied, 493 U.S. 882, 110 S.Ct. 218, 107 L.Ed.2d 171 (1989) (finding that animal studies of effects of Bendectin could not establish general causation of birth defects in humans where there was an “overwhelming” amount of contrary epidemiological evidence). There are two significant disadvantages in relying on animal studies. First, when extrapolating from animals to humans, differences in absorption, metabolism, and other factors may confound results. Second, toxicological expert opinions are “almost always” based on animal studies that involve doses of a suspected carcinogen that are significantly higher than animal doses comparable to expected human exposure. This is often necessary to obtain statistically significant predictions of the effects of realistic doses. Ref. Manual at 409. Extrapolation from high-dose animal studies, however, assumes a predictable relationship between dose and the probability that an exposed animal will be diagnosed with cancer. Ref. Manual at 345-46, 409, 414. 2. Differential Diagnosis Differential diagnosis, the process of elimination that physicians routinely use to identify the most likely cause of a particular individual’s illness, is an acceptable source of data on specific causation. Hall, supra, at 1413. By examining the patient’s symptoms, medical history, diagnostic test results, etc., a doctor can eliminate alternative causes and reach a conclusion about the most likely cause of a particular patient’s condition. “[T]o the extent that a doctor utilizes standard diagnostic techniques in gathering this information, the more likely [it is that a court will] find that the doctor’s methodology is reliable.” In re Paoli, supra, at 759. It is important to note, however, that differential diagnosis cannot demonstrate general causation, because it assumes, without proving, that all of the potential causes considered are capable of causing the condition at issue. “Indeed differential diagnosis assumes that general causation has been proven for the list of possible causes it eliminates[.]” Hall, supra, at 1413. 3. Epidemiological studies The field of epidemiology addresses the incidence, distribution and etiology (causation) of disease in human populations, Ref. Manual at 335, by comparing individuals exposed to a particular agent to unexposed individuals to determine whether exposure increases the risk of disease. Hall, supra, at 1403. A common approach to expressing the association between exposure to an agent and disease in a population is the agent’s “relative risk.” The relative risk is obtained by dividing the proportion of individuals in the exposed group who contract the disease by the proportion of individuals who contract the disease in the non-exposed group. For example, if a study found that 10 out of 1000 women with breast implants were diagnosed with breast cancer and 5 out of 1000 women without implants (the “control” group) were diagnosed with breast cancer, the relative risk of implants is 2.0, or twice as great as the risk of breast cancer without implants. This is so, because the proportion of women in the implant group with breast cancer is 0.1 (10/1000) and the proportion of women in the non-implant group with breast cancer is 0.05 (5/1000). And 0.1 divided by 0.05 is 2.0. “When [epidemiological] studies are available and relevant, and particularly when they are numerous and span a significant period of time, they assume a very important role in determinations of questions of causation.” Richardson v. Richardson-Merrell, Inc., supra. See also Ref. Manual at 335 n. 2 (“Epidemiologic studies have been well received by courts trying mass tort suits. Well-conducted studies are uniformly admitted.”) (citation omitted). Because epidemiology is concerned with the incidence of disease in populations, epidemiology is probative of general causation; “specific causation is beyond the domain of the science of epidemiology.” Ref. Manual at 381. However, and as described in the next sub-section, the Ninth Circuit has held that an epidemiological study is admissible to prove specific causation under California tort law if the study shows that the relative risk is greater than 2.0. C. Usefulness To be admissible, proffered expert testimony must assist the average trier of fact. Fed.R.Evid. 702. The Supreme Court characterizes this prong as “the ‘fit’ between the testimony and an issue in the case.” Daubert II, supra, at 1320 (citing Daubert I, at 591, 113 S.Ct. 2786). Testimony “fits” a case if it “logically advances a material aspect of the proposing party’s case.” Id. at 1315. Certain relative risk thresholds are required for the statistical results of an epidemiological study to “assist the trier of fact” in assessing causation. Daubert II, at 1320. Defendants argue at several points in their papers that to prove general causation, Plaintiff must establish that the relative risk of contracting breast cancer from PUF-coated implants is at least 2.0. E.g., Lappé Reply at 8. That argument is based on a misunderstanding of relative risk, a mis-reading of Ninth Circuit precedent and a lapse in basic logical reasoning. As explained below, this “doubling of the risk” requirement applies to statistical evidence proffered to prove specific, not general causation. As explained in Section III.B.3, supra, the relative risk is a statistical term derived from a study of hundreds or thousands of subjects. It is obtained by dividing the proportion of individuals in an exposed group who contract the disease by the proportion of individuals who contract the disease in a non-exposed group. Thus, any properly-performed epidemiological study that finds a relative risk greater than 1.0 signifies that exposure to an agent increases the probability of contracting the disease. Where the study properly accounts for potential confounding factors and concludes that exposure to the agent is what increases the probability of contracting the disease, the study has demonstrated general causation — that exposure to the agent “is capable of causing [the illness at issue] in the general population.” In re Hanford, supra, at 1134. A relative risk of 2.0, sometimes referred in certain contexts as a “doubling dose,” is not necessary to establish generic causation. Id. at 1137. When statistical analyses or probabilistic results of epidemiological studies are offered to prove specific causation, however, under California law those analyses must show a relative risk greater than 2.0 to be “useful” to the jury. Daubert II, supra, at 1320. This is so, because a relative risk greater than 2.0 is needed to extrapolate from generic population-based studies to conclusions about what caused a specific person’s disease. When the relative risk is 2.0, the alleged cause is responsible for an equal number of cases of the disease as all other background causes present in the control group. Thus, a relative risk of 2.0 implies a 50% probability that the agent at issue was responsible for a particular individual’s disease. This means that a relative risk that is greater than 2.0 permits the conclusion that the agent was more likely than not responsible for a particular individual’s disease. Ref. Manual at 384, n. 140 (citing Daubert II). Thus, suppose a study finds that 100 out of a group of 100,000 women with breast implants developed breast cancer. If, as a result of a different study comparing rates of cancer in women with implants to women without implants, a risk factor of 2.0 was known to be associated with breast implants, this would mean that the cancers in 50 of those 100 cancerous women were caused by implants. Since the study alone provides no way of knowing which of the 100 women with breast cancer are in the group of 50 whose cancer was caused by implants, the probability that any one of those women’s cancer was caused by her implants would be only 50% — just short of “more likely than not.” However, a risk factor that is greater than 2.0 would mean that the cancers of more than 50 of the 100 women were caused by breast implants, making it more likely than not that breast implants were the cause of any one of those women’s cancer. This is the reasoning underlying the “doubling of the risk” requirement of Daubert II. Proof of general causation is a prerequisite to applying this statistical “doubling risk” approach to specific causation. Ref. Manual at 383-84. Additionally, this approach to proving specific causation assumes that the plaintiff is comparable to the subjects of the epidemiological study and that there were no other causal agents present in the plaintiffs case not accounted for by the study. Id. at 385. Depending on the differences between the plaintiff and the subjects of the study, this can weigh in favor or against specific causation. For example, here Plaintiff argues that because Cagle became pregnant shortly after receiving her implants, the release of hormones associated with her pregnancy made her breast cells more susceptible to potential mutation by TDA, thus making her different from the groups evaluated in the epidemiological studies. The foregoing discussion demonstrates the inherent difficulties associated with use of statistical data to prove specific causation. As the Ninth Circuit has observed, No doubt, there will be unjust results under this substantive standard. If a drug increases the likelihood of [harm], but doesn’t more than double it, some plaintiffs whose injuries are attributable to the drug will be unable to recover. There is a converse unfairness under a regime that allows recovery to everyone that may have been affected by the drug.... One can conclude from this that unfairness is inevitable when our tools for detecting causation are imperfect and we must rely on probabilities rather than more direct proof. Daubert II, supra, at 1320, n. 13. IV. Dr. Richard Neugebauer (epidemiologist) The crux of Dr. Neugebauer’s testimony seeks to explain why in his view the existing epidemiological studies of PUF breast implants and cancer rates are either unreliable or not relevant to this case. At his deposition, Dr. Neugebauer agreed that a “fair summary of his bottom line opinion” is that “epidemiologic evidence from which to rule out a relationship between (PUF) implants and breast cancer is lacking.” Dunleavy Decl. Exh. C at 36-37. He also states that “the one study to examine this question in a rigorous manner reported at least a doubling of breast cancer risk ....” See Neugebauer Opp. Exh. B (“Neu-gebauer Expert Report”). A. Proposed Testimony Neugebauer intends to testify to the following: The epidemiological studies concluding that there is no association between breast implants and cancer are either unreliable or inapplicable to this case Support • Defendants rely on studies which failed to eliminate confounding factors. Those studies also failed to address that women who select breast augmentation are not a random sample of the female population. Neugebauer Expert Report at 2-3. • Defendants rely on studies which did not have a sufficient number of subjects. Given the relative infrequency of individual cancers, “truly enormous” study sizes are needed. Id. at 3. • Defendants rely on studies for which response rates (i.e. the proportion of persons solicited who responded to study questionnaires) fell below 70%, a rate considered “borderline acceptable.” High non-response rates may suggest participants and non-participants differ on basic so-ciodemographic characteristics. Id. at 3. • Defendants rely on studies which employed self-report questionnaires rather than objective validation or clarification by independent researchers. Id. at 4. • Many of the epidemiological studies of breast implants are totally unhelpful, because they fail to identify which type of implants were used in each subject. This is significant, because PUF-covered implants comprise a relatively small proportion of all implants. Id. at 4-5. One epidemiological study provides “suggestive evidence” that there is an association between breast implants and breast cancer. Support • A study that monitored cancer rates by implant type found that women with PUF implants experienced a doubling of their cancer risk. Id. (citing Brinton et at., Breast Cancer Following Augmentation Mammo-plasty, 11 Cancer Causes and Control 819 (2000) (Neugebauer MIL Exh. 7)); Neugebauer MIL Exh. 1 (Neugebauer Depo.) at 36:3-37:3. B. Qualifications Defendants argue that Neugebauer is not qualified to testify about epidemiological studies, because he “is not a medical doctor” and “has never conducted an epidemiologic study examining the relationship between breast implants and cancer.” Neugebauer MIL at 2. They also point out that Neugebauer has merely “review[ed] certain articles” and that “Neugebauer never took any interest in the alleged health effects of breast implants until after he was contacted to serve as a paid expert witness.” Id. at 3. Defendants’ argument is meritless. Neugebauer is a general epidemiologist, with extensive experience in designing, conducting and analyzing epidemiological studies as well as teaching others how to design, conduct, and analyze such studies. Neugebauer’s credentials are specific to an inquiry about the methodological soundness of Defendants’ epidemiological data. Furthermore, Neugebauer has testified that “the methods of epidemiology are fundamentally the same whether the possible risk factor is [PUF], asbestos, or serum cholesterol levels. The available study designs are also the same whether the outcome is breast cancer, lung cancer, or coronary heart disease.” Neugebauer Opp. Exh. E (Neugebauer Deck) at 2. That Neugebauer specializes in perinatal and psychiatric epidemiology should not preclude him from offering an opinion regarding cancer epidemiology. Given these facts and the liberal construction of expert qualifications FRE 702 requires, Neugebauer is certainly qualified to evaluate and explain the available epidemiological evidence concerning breast implants. C. Reliability As mentioned previously, Neugebauer intends to provide the following testimony: (1) Defendants’ epidemiological studies are methodologically flawed and do not specifically address whether PUF-covered breast implants are carcinogenic and (2) there is “suggestive [epidemiological] evidence” that PUF-covered breast implants are carcinogenic. Neugebauer MIL Exh. 1 (Neu-gebauer’s Depo.) at 36:3-37:3. 1. Methodological flaws Neugebauer’s analysis of the existing epidemiological evidence is admissible. Meta-analyses that probe the methodological validity of medical studies are not unprincipled or unscientific. Neuge-bauer’s testimony is based upon well known statistical principles — that the size of the study makes a difference, that the ability to control for certain variables is essential and that response rates and data collection methods affect results. Neuge-bauer has evaluated the available studies and has concluded that, given what he perceives as design flaws, those studies do not support Defendants’ contention that implants do not cause breast cancer. The various Daubert factors also support a finding that Neugebauer’s criticisms of the available studies are admissible. The statistical underpinnings of epidemiology are well-tested. They have been subjected to peer review and publication. They are generally accepted in the scientific community. Furthermore, Neugebauer did not develop the techniques he employs to evaluate Defendants’ studies for the purposes of litigation, nor has he extrapolated from an accepted premise to an unfounded conclusion. Moreover, Neugebauer’s (correct) assessment that the available epidemiological studies do not reach even tentative conclusions about whether PUF-coated implants (as opposed to implants generally) are carcinogenic, is also reliable and hence admissible. In their summary judgment Motion, Defendants cite many studies that they contend support their argument that there is no association between PUF-coated implants and breast cancer. SJ Mot. Exhs. 15-33. However, those studies determined that the silicone in breast implants are not associated with cancer. Not one of those studies purported to reach any conclusion about the rate of breast cancer in women receiving PUF-coated implants. Many of the studies did not report the number of subjects that had received PUF-coated implants and in those that did, it is clear that the number of PUF subjects was a tiny fraction of the total number of subjects studied. For example, in the 1997 study by Deapen et al. only 69 out of the 3182 subjects had received PUF-coated implants. Id. Exh. 15 at 1348; Neugebauer MIL Exh. 6 at 114. The study reported that one patient with polyurethane sponge implants developed breast cancer, but did not purport to reach any conclusion about PUF-coated implants generally. SJ Mot. Exh. 15 at 1351; Neu-gebauer MIL at 117. Similarly, the 1996 Brinton study that Defendants proffered concluded that “[f|urther studies are needed on risks associated with ... [PUF]coated implants.” SJ Mot. Exh. 16 at 274. Defendants appear to argue that a study published by Brinton in 2000 reached a conclusion about the risk of cancer from PUF-coated implants. Neugebauer MIL at 12-13. That study did no such thing. The study’s focus was the risk of breast cancer from breast implants in general and only 1.3% of the implants in the 13,488 patients studied were PUF-coated. Id. Exh. 7 at 121, 125. Based on the very limited sample of individuals with PUF-coated implants, Brinton found that the relative risk of PUF-covered implants, was 1.99 with a margin of error between 0.5 and 8.0 at the 95% confidence level. Id. at 125. That means that 95 times out of 100 a study of that type would yield a relative risk value somewhere between 0.5 and 8.0. This huge margin of error associated with the PUF-specific data (ranging from a potential finding that implants make a woman 50% less likely to develop breast cancer to a potential finding that they make her 800% more likely to develop breast cancer) render those findings meaningless for purposes of proving or disproving general causation in a court of law. Plaintiff does not dispute that the National Academy of Sciences Institute of Medicine Committee on the Safety of Silicone Breast Implants commented as follows on PUF implants: Given the relatively small number of women with polyurethane implants still in place, the natural breast cancer incidence in women and the lack of evidence for polyurethane carcinogenesis, which implies at most a small effect, if any, of polyurethane in causing human cancer, it is unlikely that any study of patients with existing implants will be able to provide sufficient evidence of an association between these implants and cancer. At present, evidence is lacking to conclude that there is an association between polyurethane-coated implants and cancer, and the way that [sic] existing evidence suggests that there is no such association. Since the implantation of polyurethane-coated breast implants within the United States is unlikely, these conflicting studies may never be reconciled. Item 7 on Revised Statement of Uncontro-verted Facts and Conclusions of Law. The absence of meaningful epidemiological data on PUF-covered implants is not surprising, given that manufacturers of PUF-covered implants suspended shipments in May 1991 and only ten percent of women with implants had the PUF-coated type at that time. Lappé Expert Report at 8; Kern et al., Carcinogenic Potential of Silicone Breast Implants: A Connecticut Statewide Study, 100 Plastic & Reconstructive Surg. 737 (1997) (SJ Mot. Exh. 22). 2. “Suggestive” Evidence Neugebauer’s assertion that the 2000 Brinton study contains “suggestive” evidence of an association between PUF-coated implants and cancer is not reliable. As discussed above, only 1.3% of the implants in that study were PUF-coated, a sample size so small that the risk factor could be anywhere between 0.5 and 8.0 ninety-five percent of the time. The Brinton study never reached any conclusion about PUF-coated implants specifically and, given the very large margin of error, any causation opinion based on the PUF-specific results of that study would be unreliable. Indeed, when asked to clarify what he meant by “suggestive” evidence, Neugebauer testified that he meant that “more work has to be done to rule it in or out.” Neugebauer MIL Exh. 1 (Neugebauer Depo.) at 38:13-16. This does not satisfy the preponderance of the evidence causation standard. Additionally, Neugebauer’s proposed testimony about the Brinton study is unreliable, because it ignores the rigorous standards and methodology Neugebauer himself applies to the other studies. “[A]ny step that renders [the expert’s] analysis unreliable ... renders the expert’s testimony inadmissible. This is true whether the step completely changes a reliable methodology or merely misapplies that methodology.” Fed.R.Evid. 702 Advisory Committee’s Notes 2000 (quoting In re Paoli, supra, at 745 (emphasis omitted)). Neugebauer ignores that the PUF-specific findings of the Brinton study suffer from the same flaws he points to in the other studies: (a) the sample size of subjects with PUF implants was vanishingly small; (b) the overall response rate (approximately 71%, Neugebauer MIL at 122) was only “borderline acceptable,” Neugebauer Expert Report at 3; and (c) the study relied on questionnaires rather than objective validation or clarification. Neugebauer MIL Exh. 7 at 122; Neugebauer Expert Report at 4. D. Usefulness Defendants argue that Neugebauer’s testimony criticizing the available epidemiological studies about breast implants and opining that “more work has to be done” to ascertain whether PUF-coated implants cause cancer is not useful to the jury. Defendants appear to be arguing that because the testimony does not show that breast implants cause breast cancer, Neu-gebauer’s opinion does not advance Plaintiffs case and is thus irrelevant. This argument lacks merit. It is clear from Defendants’ papers that they intend to proffer at least fifteen epidemiological studies concluding that there is no evidence that silicone breast implants cause cancer. Plaintiff is entitled to use Neugebauer, an expert in epidemiology, to point out the main problem with those studies' — that they reach no conclusion about the propensity of PUF-coated implants to cause cancer. This is classic rebuttal expert testimony. Aside from his testimony about “suggestive evidence” (which will be excluded), Neugebauer’s purpose is to explain why the studies Defendants intend to proffer are inapposite. Plaintiff is entitled to inform the jury that Defendants’ studies about silicone implants did not reach conclusions about the carcinogenicity of PUF-coated implants and about those studies’ purported methodological limitations. Defendants’ reliance on Kelley v. Am. Heyer-Schulte Corp., et al., 957 F.Supp. 873 (W.D.Tex.1997) and Hall, supra, is unfounded. In Kelley, the plaintiffs epidemiology expert had testified that the available studies provided limited evidence of an association between breast implants and Sjogren’s Syndrome (the injury at issue in that case) and that “more studies are needed.” Kelley, 957 F.Supp. at 881. At the same time, however, the expert also intended to rely on those inconclusive studies to assert that a causal relationship did exist between breast implants and Sjo-gren’s Syndrome. Unlike in Kelley, Plaintiff does not intend to rely on the same studies Neugebauer distinguishes to advance a positive theory of causation. In Hall, the court excluded an expert’s “reanalysis” of existing epidemiological studies because it was unreliable and the witness’s criticisms of the studies were not accepted within the scientific community. The Court also noted that it had already determined that the studies in question were inadmissible, thus rendering the expert’s testimony about them irrelevant. Id. at 1406-07. E. Conclusion For the foregoing reasons, Neuge-bauer’s testimony regarding the reliability and applicability of Defendants’ studies is admissible under Fed.R.Evid. 702. Neugebauer’s claim that the 2000 Brin-ton study offers “suggestive evidence” of the carcinogenicity of PUF-coated implants is not admissible. That study did not make any finding about whether PUF-coated implants are carcinogenic and the error rate associated with the PUF-specific data makes it wholly unreliable. Additionally, Neugebauer’s conclusions regarding the Brinton study are vulnerable to the same criticisms he levies against Defendants’ studies, namely, that the sample sizes were too small and the response rates were too low. V. Dr. Christopher Batich (polymer chemist) A. Proposed Testimony Batich intends to testify to the following. PTJF-eoated implants degrade after implantation in humans, releasing TDA Support • 1991 and 1995 research findings that PUF-covered breast implant patients who had previously tested negative for TDA in their blood and urine tested positive post-implantation. Chan et al., Detection of toluenediamines in the urine of a patient with polyurethane-covered breast implants, 37 Clinical Chemistry, 756-58 (1991) (Batich Opp. Exh. W); Sepai, et al., Exposure to toluenediamines from polyurethane-covered breast implants, 77 Toxicology Letters 371-78 (1995) (Batich Opp. Exh. T). • A 1993 in vitro study demonstrating that Meme PUF-covered implants contained “significant amounts” of residual TDA while degrading at 0.8% per year. Benoit et al., Degradation of polyurethane foams used in Meme breast implant, 27 J. of Biomedical Material Research, 1341-48 (1993) (Batich Opp. Exh. P). See also Luu, et al, Characterization of Polyesteru-rethane Degradation Products, 5 J. of Applied Biomaterials, 1-7 (1994) (Ba-tich Opp. Exh. R) • Summarizing the available research, the National Toxicology Program (“NTP”) stated in 2001 and in December 2002 that TDA is a degradation product of the PUF used in Meme silicone breast implants. The NTP reported that elevated levels of TDA (between 0.4 to 6 ng/ml) were detected in the urine and plasma of patients up to two years after implantation. SJ Opp. Exh. 54 at 850 (Ninth Report); http://ehp.niehs.nih.gov/roc/ tocio, html (Tenth Report). • Defendants’ expert Dr. Gale has testified that she has “no reason ... to controvert the basic scientific premise that TDI-based foam [PUF] will degrade and release TDA.” Batich Opp. Exh. BB at 362:5-10 (Cross-examination of Dr. Gale in Brusca litigation). • Batich conducted his own 1989 study intended to characterize the degradation products of the polyurethane used in PUF-coated implants. He identified the chemical reaction that causes PUF to degrade and release TDA. Although the conditions used in the experiment (150 C and concentrated sodium hydroxide) were quite different from in vivo conditions, Batich maintains that the harsh conditions merely accelerated the same chemical reaction as would occur in vivo. The purpose of the experiment, Batich says, was not to assess how much TDA will be released under physiologic conditions, but to figure out what the hydrolysis products of PUF are. Batich et al., Toxic Hydrolysis Product From a Biodegradable Foam Implant, 23 J. Biomed. Mater. Res.: Applied Biomaterials, 311-19 (1989) (Batich Opp. Exh. F); Batich, et al., Letter to the Editor and Response, 1 J. Applied Biomaterials 193-95 (1990) (Batich Opp. Exhs. G & H). As to Cagle, in her Meme implant a polyester based polyurethane foam was used rather than the more stable olip-hatic or polyether based one. This led to more rapid degradation. Defendants never conducted studies before placing PUF-coated implants on the market to determine the biodegradation products of those implants. Because such studies were “simple to carry out and inexpensive!,][tjhere was no good reason not to have done them, and the failure to test ... was unreasonable and unsafe. ” Batich Opp. Exh. B (“Ba-tich Expert Report”) at 12. Support • Batich says that before marketing a biomaterial, studies ought to be conducted to determine (a) what degradation products are formed; (b) what quantity of them are released; (c) the biological properties of the released components; (d) where the products travel within the body and (e) what biological changes they cause. Id. B. Qualifications 1. Biodegradation of PUF-Coated Implants Defendants argue that Dr. Batich is not qualified to opine about whether TDA released from PUF-coated implants can cause breast cancer. Plaintiff responds that Batich is not a “causation” witness, but “is offered as an expert ... on the issue of the defective design of defendant’s Meme implant.” Batich Opp. at 1. According to Plaintiff, the purpose of Ba-tich’s testimony is to support Plaintiffs negligence claim (see Am. Compl. ¶¶ 95-104) by testifying that Defendants breached a duty of care when they failed to conduct degradation studies on the polyurethane foam and when they failed to use a more stable oliphatic or polyether-based foam. Id.; Batich Expert Report. Defendants respond that Batich really is a causation witness, because his duty-of-care opinion is “wholly based on his opinion that polyurethane breast implants or their breakdown products cause cancer.” Ba-tich Reply at 2. Both parties are partly correct. Ba-tich’s testimony is an essential part of Plaintiffs causation theory, but in a much more limited way than Defendants represent. Defendants are correct that Dr. Ba-tich is not qualified to testify that TDA is a probable human carcinogen or that TDA that may have been released from Toni Cagle’s implants caused her breast cancer, but nothing in Batich’s expert report indicates that he intends to so testify anyway. Batich Expert Report. However, Batich, a polymer chemist who directs a biomedical engineering program at the University of Florida, is qualified to testify about the mechanisms of hydrolysis of PUF and the in vitro and human studies finding that PUF degrades to TDA. Although Batich is not a physician or biochemist, he may opine about these studies because they primarily involve chemical analyses of patients’ blood and urine using classic tools of the analytical chemistry trade, such as gas chromatography and mass spectrometry. E.g. Sepai et al. (Batich Opp. Exh. T) at 871-72. A physician isn’t needed to interpret a study merely because the analysis involves a urine sample. 2. Duty of Care Defendants argue that Batich’s duty-of-care testimony is inadmissible because it relies on the premise that PUF implants cause cancer. Not so. First, Batich’s testimony is not based on that premise; he intends to testify that Defendants breached their duty of care by marketing PUF implants without carrying out what he characterizes as simple and inexpensive tests to determine whether the polyurethane would break down in vivo. He also intends to testify that Defendants should have used a more stable polyether-based foam than the polyester-based one they actually used. While it is true that Plaintiff cannot prevail on his negligence claim unless he shows causation, Batich’s testimony — essentially that Defendants breached a duty of care by not taking reasonable precautions — does not require a prior finding of causation. Second, to the extent that Batich may refer to the TDA breakdown product of PUF as a known animal carcinogen, he may permissibly rely on the testimony of Dr. Lappé (see infra Section VI) as support. Defendants also argue that Ba-tich’s duty-of-care testimony is inadmissible, because Batich is not qualified to testify about medical device testing standards. Batich MIL at 16. More specifically, they argue that Batich’s five-pronged pre-mar-ket testing requirement (see Section V.A., supra) ■ is inadmissible, because it merely constitutes “what [Batich] think[s] would be good prudent practice for someone who is producing a medical device.” Batich MIL Exh. 1 (“Batich Depo.”) at 62:6-23. Defendants are correct that Batich is not qualified to opine that Defendants breached a duty of care by failing to conduct certain tests. As a basis for Batich’s expertise, Plaintiff notes that Batich worked in the quality control department of a pharmaceutical company between 1965 and 1967, that he published several papers about medical devices and bioma-terials, and that he holds several patents on medical devices. Batich Opp. at 6. This is insufficient to qualify Batich as an expert on when and whether it is “unreasonable” not to conduct biodegradation tests on a biomaterial before using it in- an implantable medical device. Plaintiff proffers no evidence that Batich has any experience developing an implantable medical device for general use or that he has any foundational knowledge about what standard practices exist in the industry in this regard. Batich has testified that other than his studies of the hydrolysis products of PUF and “gel bleed” from implants, he has not tested any other medical device for biodegradation products. Batich Depo. at '65:18-67:2. Batich is qualified, however, to describe how one could test the PUF-coating to determine whether it releases toxic products. He is also qualified to testify about the amount of time it would take to conduct those tests and how much it would cost. Defendants nonetheless argue that even this limited testimony would be inadmissible, because Batich is not familiar with the applicable FDA regulatory standards. That argument lacks merit. What the FDA requires a medical device manufacturer to do is not the per se standard for determining what that manufacturer’s duty is in a state law negligence case. See Goodlin v. Medtronic, Inc., 167 F.3d 1867, 1382 (11th Cir.1999) (FDA’s pre-market approval of medical device did not preempt state law tort claim); Hill v. Searle Labs., 884 F.2d 1064, 1068 (8th Cir.1989) (“FDA approval is not a shield to liability.... FDA regulations are generally minimal standards of conduct unless Congress intended to preempt common law, which Congress has not done in this area.”). Although Batich cannot say whether the FDA requires manufacturers to test for biodegradation products of new biomaterials, this is not fatal to his expert testimony. C. Reliability and Usefulness Dr. Batich’s testimony is both reliable and relevant. His testimony regarding the in vivo breakdown products of PUF is based on his own published, peer-reviewed study and other published, peer-reviewed studies. See supra Section V.A. That Defendants can point to contrary studies is not sufficient to render Batich’s testimony inadmissible. In addition, Ba-tich’s testimony is relevant, because the breakdown of PUF is an essential aspect of Cagle’s causation theory. Batich may not testify, however, about his 1989 study findings on the amount of TDA released from PU foam. That study employed extreme experimental conditions. He placed a 500 milligram sample of PUF in concentrated sodium hydroxide overnight at a temperature of 150 C to determine what the hydrolysis products of PUF were. He determined that in those conditions about 85 milligrams of TDA (17 percent of the sample) was released. Batich Opp. Exh. F at 52. Since there is no evidence that PUF degrades at such a fast rate in vivo, Batich lacks foundation to testify about how much TDA was produced in his study and it would be prejudicial for him to do so. D. Conclusion Dr. Batich’s testimony that the polyurethane coating of PUF-coated implants hydrolyzes in vivo to produce TDA is admissible. He may discuss the available literature on the subject, including the amounts of TDA found by researchers in the blood and urine of patients shortly after those patients received PUF-coated implants. However, because he carried out his own 1989 in vitro study under extreme conditions, his own findings regarding the amount of TDA released would not be reliable or relevant and may be prejudicial. Batich also may testify about how one would test for biodegradation products of a biomaterial, how long such a test may take and how much it would cost. He may not testify that “wide use” of all biomaterials must be preceded by such tests or that any failure to test the product(s) used in this case was “unreasonable.” (Ultimately, of course, the question of whether Defendants breached a duty of care by not testing the PUF coating or when they chose to use a polyester based polyurethane instead of a more stable plastic would be an issue for the jury.) VI. Dr. Marc Lappé (toxicologist) A. Proposed Testimony Lappé intends to testify to the following: Studies have shown that polyurethanes, when implanted in rats, are carcinogenic. Support • A 1964 study by Dr. Hueper in the Journal of the National Cancer Institute in which polyurethane foams were implanted in rats. After the foams were implanted either in the animals’ neck or abdomen, carcinomas and sarcomas were observed. Degradation of the foams in vivo was also observed. SJ Opp. Exh. 60 at 964-986. • A 1975 study by Dr. John Autian in Cancer Research in which 17 chemical varieties of polyurethane were implanted in the abdomens of rats. Autian found that “the data clearly indicate an increased incidence of fi-brosarcomas” in rats that received implants. Again, biodegradation of the polyurethanes was observed. Id. Exh. 63 at 1063-68. • A 1976 study by Dr. Autian in Cancer Research in which one of the polyurethanes from the 1975 study was implanted in the lungs of rats at varying doses. Autian reported that the frequency of fibrosarcomas increased with increased polyurethane doses and that biodegradation of the polyurethanes had occurred. He concluded that “our data are consistent with a mechanism of biological degradation of the polymer to yield an active carcinogenic compound.” Id. Exh. 64 at 1069-72. PUF-coated implants including the Meme implanted in Cagle, degrade after implantation in humans, releasing TDA. Support • Lappé Expert Report at 9-10 (citing many of the same studies apparently relied on by Batich). Studies show that TDA is an animal carcinogen, Support • Cancer bioassays (animal studies designed to measure the propensity of an agent to cause cancer in animals) are used by regulatory agencies to estimate the cancer risk to humans. Shanklin Opp. Exh. U at 48-49. In the case of TDA, it appears that cancer bioassays were conducted by feeding TDA to rats or by injecting TDA under a rat’s skin. Lappé Report at 7. The bioassays conducted by the National Cancer Institute revealed that TDA produced neoplastic nodules in both male and female rats and mammary carcinomas (breast cancer) in female rats. The bioassays also found TDA to be carcinogenic in female mice. Lappé Report at 7 (citing SJ Opp. Exh. 52 at 841 (NCI Bioassay)). • Other studies have shown that TDA produces cancer in animals after being painted on the skin, injected under the skin or provided in the animals’ diet. Lappé Report at Ills. • A Joint Report of the United Nations Environment Programme, the International Labour Organization and the WHO stated that TDA had been found to increase the incidence of mammary tumors in rodents. Lappé Report Exh. P at 119. TDA is a “probable” human carcinogen. Support • Based on the rodent bioassays (not the polyurethane implant studies) several groups have concluded that TDA is probably or possibly carcinogenic in humans: • The National Toxicology Program (“NTP”) has stated that TDA is “reasonably anticipated to be a human carcinogen” and observed that TDA increases the incidence of breast cancer in female rats when administered in the diet. SJ Opp. Exh. 54. • In 1986 the EPA found that TDA would likely meet the criteria for classification as a “probable human carcinogen” based on the NTP determination. SJ Opp. Exh. 53. • The WHO International Agency for Research on Cancer (“IARC”) classified TDA as “possibly carcinogenic to humans” in 1987. SJ Opp., Exh. 74. • Studies have revised the risk of breast cancer from PUF-coated implants upwards based on biodegra-dation data. • The FDA initially estimated the risk ■ as between 1 and 5 in 1,000,000 (1995). Lappé Expert Report at 15. • The FDA revised this estimate to 1 in 400,000, which the FDA team concluded presented “an unreasonable risk to health for the patient.” (1998). Id. at 16-17. • A study by Sepai et al., which Lappé contends is more accurate, estimated the risk as 1 in 8,431 (1995). Id. at 16. It is “highly likely, beyond a reasonable scientific doubt, that the rapid development of Ms. Cagle’s tumor was at a minimum accelerated, and possibly caused, by her exposure to a known animal carcinogen [TDA].” Id. at 20. Support • Using data specific to Cagle and concepts derived from in vitro PUF biodegradation research, Lappé estimates that 413.6 nanograms of TDA was released from Cagle’s implants per day during the first year after implantation. Using that data, the assumed mass of Cagle’s breast tissue, and TDA’s “potency factor,” Lappé concludes that there was a 1 in 3,484 chance that Cagle’s breast cancer was due to her implants. Id. at 17-19. Since the baseline risk of breast cancer for a person of Cagle’s age was apparently 1 in 3,333, Lappé concludes that the additional 1 in 3,484 risk created by Cagle’s implants “effectively doubled” Cagle’s risk of breast cancer. Id. at 29. However, he admits that “by itself, such a numerical exercise does not prove or disprove causation.” Id. at 19. He later revised that risk assessment, concluding that the risk of developing breast cancer as a result of TDA is approximately 1 in 2 million. Lappe MIL Exh. 1 (Lappe depo.), at 7:9-9:25, 96:22-97:14. The principles Lappe applies from the in vitro research can be extrapolated to humans because he cites an in vivo study that found an even higher cancer risk than the in vitro study upon which he relies. Id. at 18 (citing Sepai et al, Exposure to Toluenediamines from polyurethane-covered breast implants, Toxicology Letters 77 (1995) 371-378, SJ Opp. Exh. 50). • Lappé cites several factors that would purportedly increase Cagle’s risk factor or suggest that his conclusion is sound: • Cagle chose some of the biggest implants available, thus, increasing TDA levels. • Cancer risks increase relative to the placement of the TDA; here, TDA was administered locally via implantation. • Cagle’s cancer was extremely aggressive and TDA can accelerate carcinogenesis. In this