13
`
`Commercial Success:
`
`Economic Principles Applied
`to Patent Litigation
`
`jesse David and Marion B. Stewart
`
`A party accused of infringing a patent may contend that the asserted
`
`patent
`
`is invalid because of obviousness. That contention may be
`
`rebutted by a showing that the patented invention is a commercial suc—
`
`cessH—one of several secondary considerations that courts look to for iden-
`
`tifying the differences between the patented invention and the prior art.
`
`These secondary considerations—known as objective indicia of nonobvi-
`
`ousnesSv—also include such factors as copying, long-felt but unsolved
`need, failure of others, and licensing.1
`
`Determining whether an invention has, or has not, been a commercial
`
`success is primarily an economic exercise, and economists increasingly
`
`assist courts in evaluating this issue. Case law indicates that courts have
`
`traditionally looked for characteristics such as increasing revenues, gain
`
`in share in an appropriately defined market, and public acclaim in an
`
`attempt to determine whether a product has been a commercial success.
`
`Courts have also considered whether the patent holder has established a
`
`nexus between the claimed invention and the product’s commercial suc-
`cess—that is, whether the commercial success, if evident, is due to the
`
`patented feature as opposed to some other characteristic of the product
`
`or a mode of selling employed by the manufacturer.
`
`
`
`‘
`
`in re Denis Rouffet, Yannick Tanguy, and Frederic Berthault, 149 F.3d 1350, 47 USPde
`1453 (Fed. Cir. 1908). it is our understanding that courts may consider all of these
`indicia in an assessment ofa patent’s validity. For the purposes of our discussion, we
`consider only these factors that should weigh in a determination of commercial suc-
`cess, not whether or the extent to which those factors could support a finding of
`validity or invalidity.
`
`105
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`Page 1 of 13
`
`

`

`ECONOMIC APPROACHES TO INTELLECTUAL PROPERTY POLICY,
`LITIGATION, AND MANAGEMENT
`
`From an economic perspective, commercial success could in principle
`be defined by a single criterion: Does the patented invention earn a pos-
`
`itive net return (risk~adjusted) on invested capital after accounting for all
`relevant costs associated with developing and commercializing the patent
`
`as well as any alternatives available to the patent holder? Patents exist to
`
`protect the human and financial investment used to develop new prod-
`
`ucts, services, or processes. This investment, however, is only beneficial,
`
`from a social perspective, if consumers are willing to purchase an embod~
`
`iment of the invention at such a price as to fully compensate the inventor
`for all costs incurred in bringing the product to market.z Put simply,
`
`patents are not needed to protect inventors from making poor invest—
`ment decisions.
`
`The courts’use of the previously mentioned factors is not necessarily
`
`in conflict with this definition, and many—perhaps most—previous deci—
`
`sions made by courts are likely to have been consistent with it. Given the
`
`limitations on available data, it is entirely reasonable that an analysis of
`
`commercial success should consider and place significant weight on the
`
`traditional measures such as market share or revenue growth. However,
`
`under certain circumstances, rapid sales growth and gains in market share
`
`will not necessarily reflect a profitable underlying invention. Moreover,
`
`calculating the proper measure of profitability can be a complicated task
`
`and should be considered in an appropriate context—for example. relative
`
`to an appropriate benchmark or alternative. Consequently. it is our opin-
`
`ion that courts should look more deeply into the economic characteristics
`
`of the product before arriving at a determination of the commercial
`
`success of the patent.
`
`A Summary of the Case Law
`
`In Graham v. John Deere C0,, the seminal case identifying commercial suc—
`cess as a relevant secondary consideration in a determination of patent
`validity, the Supreme Court of the United States cited an article in the
`
`University ofPennsylvania Law Review that focused on the consumer per-
`spective for evaluating the commercial succeSs of a patent. The article
`stated that “{t]he operative facts...are the actions of buyers rather than
`those of prodUCers.”3 Case law since Graham has generally followed this
`
`3 One could imagine that, for reasons of public policy, a patented invention related to
`health care could be sold at an artificially low price, or even given away, but such a
`strategy would not reduce the true value of the invention.
`3 Graham v. John Deere Co., 383 US. 1 (1966): and Richard L. Robbins, "Subtests of
`‘Nonobviousness,'” University of Pennsylvania Law Review 112 (1963-1964}: 1175.
`
`196
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, |PR2017-01622
`Page 2 of 13
`
`

`

`COMMERCIAL SUCCESS: ECONOMIC PRINCIPLES
`APPLIED TO PATEN'I‘ LITIGATION
`
`position. For example, in Demaco Corp. 1:. Fl. Von Langsdorff Licensing Ltd,
`
`the court stated the following:
`
`The rationale for giving weight to the so-called “secondary consid-
`
`erations” is that
`
`they provide objective evidence of how the
`
`patented device is viewed in the marketplace, by those directly
`interested in the product.4
`
`Based on this approach, courts appear to have turned to a few standard
`
`measures of consumers’ demand for the patented product, such as total
`
`unit sales or revenues. Although not universally, the courts have generally
`
`recognized that this information must be placed in a “meaningful context”
`
`and consequently have noted that the sales must represent a significant
`
`and/or growing share of that product in some “market.” This also follows
`
`the University of Pennsylvania Law Review article, which stated that “[t]he
`
`basic measure of commercial success should be the proportion of the total
`market for the product that the patentee has obtained.”5 Subsequent deci-
`sions have reinforced the standard that sales figures must at least be con-
`
`sidered in light of the size of the overall market, although the method for
`identifying the appropriate market has not generally been specified}[1
`However, achieving a significant volume of sales or even a large mar-
`
`ket share does not necessarily indicate that the inventor should view a
`
`patent as a success. For example, sales may be driven by characteristics
`
`other than the patented invention, such as other patented features, non-
`
`patented characteristics, and brand name. For some products, market
`
`share may also be affected by advertising. (The basic formulas for Coke
`
`and Pepsi haven’t changed in decades, yet market shares appear to be
`
`affected by changing marketing strategies on the part of the two compa-
`
`nies.) As an extreme example, increasing sales and market share of a
`
`product could also be generated by simply lowering price, a tactic some-
`
`times employed by companies seeking to create customer awareness early
`
`4 Demnco Corp. 1-. Fl. Von Langsdorff Licensing Ltd, 851 F.2d 138?, v US PQ2d 1222 (Fed.
`Cir. 1988).
`5 University of Pennsylvania Law Review, 1175.
`6
`For example. see Ecolocliem Inc. v. Southern California Edison (30., 22.? F.3d 1361 (Fed. Cir.
`zooo): Cable Electric Products Inc. v. Cenmorlz Inc, 770 F.2d 1015, 22.6 USPQ 881 (Fed.
`Cir. 1985); and Iberitech Inc. v. Monoclonal Antibodies Inc, 802 F.2d 136?. 231 USPQ 81
`(Fed. Cir. 1936). An exception where a decision considered sales explicitly outside the
`context of the size of the overall market is Neupalz Inc. 1;. Ideal Manufacturing and Sales
`Corp, 41 Fed. Appx. 43s; 2002 US. App. LEXIS 13843 (Fed. Cir. 2002}. In}?! Eaton and
`Co. v. Atlantic Paste and Glue Ca, 106 F.3d 1563. 41 USPQ2d 1641 (Fed. Cir. 1997). the
`court similarly found that a large number of units sold did represent evidence of com-
`mercial success, without any showing of a share in a well-defined market.
`
`197
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, |PR2017~01622
`Page 3 of 13
`
`

`

`ECONOMIC APPROACHES TO INTELLECTUAL PROPERTY POLICY,
`LITIGATION, AND MANAGEMENT
`
`in the product life cycle. The Manual of Patent Examining Procedure, pub-
`
`lished by the United States Patent and Trademark Office, identifies this
`
`nexus between the success of the product and the patent itself as a key
`
`component of a nonobviousness claim:
`
`An applicant who is asserting commercial success to support its
`contentiOn of nonobviousness bears the burden of proof of estab-
`
`lishing a nexus between the claimed invention and evidence of
`commercial success?
`
`Courts have recognized some of these possibilities and have generally
`
`required a showing that any commercial success be directly linked to
`
`demand for the patented feature rather than any other factors.
`
`Cansequently, for any data on sales or market share to be relevant, one
`
`must be able to demonstrate that whatever demand for the product
`
`exists, it is due. at least in part, to the patent, not some other features or
`actions by the seller.8 A simple thought experiment can shed light on the
`concept of a nexus. Suppose the patented inventiou were made unavail-
`able and removed from the product. Could the seller attain the same level
`
`of commercial suCcess? Or, from an economic perspective, what is the dif-
`
`ference in net profits that would accrue to the patent holder if the
`
`patented inventiOn were removed from the product?
`
`Despite the courts’ tendency to view commercial success from only
`the consumers’ perspective, a few decisions have recognized profitability
`
`as a factor that might be considered along with other objective economic
`
`evidence. For example, in Cable Electric Products Inc. v. Genmorie Inc, the
`court stated:
`
`Without further ecouomic evidence, for example, it would be
`
`improper to infer that the reported sales represent a substantial
`
`share of any definable market or whether the profitability per unit
`is anything out of the ordinary in the industry involved.9
`
`Discussions of profitability or other “supply-side” considerations have
`
`been included in assessments of commercial success in only a few other
`
`3" United States Patent and Trademark Office, Manual of Patent Examining Procedure,
`February 2003 revision. 5 1:16.03. See also Dt-maco, 851 F..zd 1337.
`3 Although the courts have consistently recognized that the issue of a nexus is critical
`in a determination of commercial success. in many cases they have found that the
`existence of a significant advertising budget does not in itself rebut the presumption
`that the commercial success of the product at issue must be clue to the patented
`invention. For example, see Merck and Co. v. Dunbury Pharmacai Inc, 694 F.Supp. 1. at
`(D. Del. 1988): and Hybritech, 802 P.2d 1367.
`9 Cable, v70 F.3d 1015.
`
`198
`
`___—‘
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`Page 4 of 13
`
`

`

`COMMERCIAL SUCCESS: ECONOMIC PRINCIPLES
`APPLIED TO PATENT LITIGATION
`
`cases.10 As these cases properly point out, ultimately an inventor’s suc-
`
`cess should be judged by the returns to his investment relative to that
`inventor’s next-best alternatives.
`
`Economic Criteria
`
`In the first edition of his ground—breaking book, Economic Analysis ofLaw,
`the distinguished jurist Richard Posner discussed the normative (i.e.,
`
`prescriptive) and positive (i.e., descriptive) roles of economics in the law:
`
`Economics turns out to be a powerful tool of normative analysis of
`law and legal institutions—a source of criticism and reform...The
`
`normative role of economic analysis in the law is fairly obvious.
`
`The positive role—that of explaining the rules and outcomes in
`
`the legal system as they arewis less obvious, but not less impor-
`
`tant. As we shall see, many areas of the law, especially the great
`
`common law fields of property, torts, and contracts, bear the stamp
`
`of economic reasoning. Few legal opinions, to be sure, contain
`
`explicit references to economic concepts and few judges have a
`
`substantial background in economics. But the true grounds of
`
`decision are often concealed rather than illuminated by the charac-
`teristic rhetoric of judicial opinions.11
`
`As described above, we suggest that there is a straightforward norma-
`
`tive role for economies in determining commercial success: A patented
`invention should be considered a commercial success if it can be shown
`
`to have earned, or can reasonably be expected to earn, a positive net
`
`return on invested capital after accounting for all relevant costs associ-
`
`ated with development and commercialization as well as any alternatives
`
`available to the patent holder and the amount of risk borne by the patent
`
`holder. Although courts would do well, in our view, to adopt more explicit
`
`economic reasoning along these lines in their analysis of commercial suc—
`
`cess issues, our reading of the relevant cases suggests that a substantial
`
`amount of economic analysis has already found its way into judicial opin-
`
`ions regarding commercial success.
`
`Under certain circumstances, it appears that economic analysis could
`
`provide a definitive answer to the question “Has a patented invention
`been a commercial success?” For example, suppose that:
`
`1° For example. see Miles Laboratories Inc. v. Shandon inc, 1992 WL 503432 (W. D. Pa};
`and In re Ben Huang. 100 F.3d 135, 40 USPde 1635 (Fed. Cir. moo).
`‘1 Richard A. Posner, Economic Analysis ofLaw (Boston. MA: Little, Brown and Co.,
`1971). 6-
`
`199
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`Page 5 of 13
`
`

`

`ECONOMIC APPROACHES TO INTELLECTUAL PROPERTY POLICY,
`LITIGATION, AND MANAGEMENT
`
`1.
`
`a start-up company, faunded solely to exploit a single patented
`
`invention, incurred costs (in present value terms} of $1 million to
`
`develop a single saleable product;
`
`2. over its entire life cycle—mow completedu-sales of that product
`
`generated net profits of $2. million (again, in present value terms);
`and
`
`3.
`
`there is no doubt that the product characteristics and/or other
`
`factors that led consumers to purchase the product were all due to
`the invention.
`
`The first assumption allows us to say with certainty that it cost pre-
`
`cisely $1 million to develop a product embodying the patented invention,
`since we assume away any difficulties that would be caused by the need to
`
`associate “common costs” in, say, a central research-and—development
`
`(R&D) facility with the development of a particular invention. The second
`
`assumption eliminates the difficulty of evaluating the potential profits still
`
`to be earned by a product Currently on the market. The third assumption
`
`assures that the nexus between patented invention and sales success has
`
`been established. Assuming that an appropriate interest rate has been used
`
`to “discount” (or appreciate) the investment and the resultant profits, 3 $2
`
`million return on a $1 million investment would surely count as a com-
`
`mercially successful venture from the perspective of the producer. Since
`
`(by assumption) the patented invention is what made that return possible,
`
`then the patented invention should be deemed a commercial success.
`
`In our experience, however, the issues that need to be addressed are
`
`always more complicated than the stylized example above, so it is hardly a
`surprise that—as far as we know—no reported case has reached a decision
`
`regarding the commercial success of a patented invention simply by compar-
`
`ing the Cost of developing and selling the patented product with profits
`
`earned on that product. Our own research has made clear that even large,
`
`technology-oriented companies have difficulty associating early-stage R8cD
`
`costs with what ultimately became a commercially viable product, inevitably
`
`leading to some uncertainty regarding the total cost of bringing a patented
`invention to market.12 In addition, determining profitability for a single
`product sold by a multiproduct company can be further complicated if the
`growth in that product’s sales comes at least partly at the expense of profits
`
`
`‘3 Even iI detailed prtxluctvspecil'ic R&D cost data were not available. however. one
`might be able to make a reasonable evaluation of commercial success by comparing a
`product’s profits to the average cost of developing and commercializing broadly simi-
`lar products.
`
`200
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`Page 6 of 13
`
`

`

`COMMERCIAL SUCCESS: ECONOMIC PRINCIPLES
`APPLIED T0 PATENT LITIGATION
`
`elsewhere in the company or, alternatively. if sales of the patented product
`
`generate additional profits for the company by drawing consumers to other
`products. Another complication arises from the fact that most patent dis-
`
`putes involve products currently or not yet on the market. not products
`
`whose life cycles have ended, adding further uncertainty regarding the prof-
`its that will ultimately be generated.13 And finally. while there are certainly
`instances in which there is no doubt that the patented invention has created
`the performance characteristics that were responsible for the product’s suc-
`
`cess. our studies have also revealed contrary examples in which it was clear
`
`that a patent played little. if any. role in generating product sales. Given the
`
`data imperfections that frequently make a “direct” measure of commercial
`success impractical,“1 it is therefore not surprising that courts have tended
`to focus on “indirect" evidence. such as growth in market share. As Judge
`
`Posner suggested, however. many of the courts' decisions on commercial
`
`success nevertheless “bear the stamp of economic reasoning."
`
`For example, economic reasoning makes clear that pharmaceutical
`
`companies would not invest in research on a particular class of drugs.
`
`such as antibiotics. unless they believed that on an expected—value basis
`
`that research would be profitable. If companies' expectations are rational.
`
`then a bundle of “average" marketed antibiotics will generate enough prof-
`
`its over their life cycles to yield an acceptable return on the companies’
`R&D investments. A drug that clearly does much better than average is
`very likely. therefore, to be a commercial success.Is
`
`How would we know that a drug is much better than average? A large
`
`(i.e.. much above-average) market share would be a likely indicator. and
`
`rapid growth in market share—particularly if the product is not too far
`
`into its life cycle—would also likely be relevant, since the expected pres-
`
`ent value of a product's profit stream will be greater the sooner those
`
`13 Sometimes. however. as with a blockbuster prescription drug that has performance
`features clearly due to the patent at issue. the sales and profits generated during the
`first few years of the product's life will be sufficiently large to leave little doubt about
`the patent ’5 commercial success even if uncertainties remain regarding the precise
`magnitude of early-stage R&D costs.
`‘4 A further complication relates to the possibility that infringement by a competitor
`may affect the profitability of a patent holder's product. and therefore its apparent
`commercial success. In order for an analysis of profitability to be of use in assessing
`commercial success, one should account for the actions of the infringer. Moreover.
`aseessing only the infringer’s profitability. rather than the patent holder's. may not
`provide an appropriate measure of commercial success. Such a problem: could arise if.
`for example. the infringe: had a different cost structure or sold to a different group of
`customers than the patent holder.
`15 Note, however. that the inverse is not necessarily true. For example. in a highly prof~
`itahle industry. a "below-average" product may still be a commercial success.
`
`20].
`
`UNITED THERAPEUTICS. EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, |PR2017-01622
`Page 7 of 13
`
`

`

`ECONOMIC APPROACHES TO INTELLECTUAL PROPERTY POLICY,
`LITIGATION, AND MANAGEMENT
`
`profits are earned.16 The courts’ reliance on market—share data and
`growth in market share, as described in the previous section, appears to
`
`be sensible in light of the likely imperfections in the data that would have
`
`shed a more direct light on the issue of a product’s commercial success.
`
`Despite the fact that, for the most part, Courts' general approach to
`
`determining commercial success has been Consistent with these eco-
`
`nomic concepts, it appears that some decisions would have benefited
`
`from more. or at least more explicit, economic analysis. For example, in
`
`the Neupak Inc. v. Ideal Manufacturing and Sales Corp. case, the court of
`
`appeals found the following:
`
`Because the record shows that between 1995 and 2000 Neupak’s
`
`patented mobile filling carts enjoyed a significant increase in sales
`
`and constituted an increasing share of Neupak’s business, the dis-
`
`trict court did not clearly err in concluding that Neupak demon-
`strated a nexus between commercial success and the '233 patent.”
`
`In this case, not only did the courts (both the district and appeals)
`apparently fail to put Neupak’s sales into any “meaningful context,” there
`appears to be a possibility that the product embodying the patented
`
`invention became succeSSful at the expense of other Neupak products. It
`
`is likely that a relatively simple analysis of the company’s financial
`
`records could provide a definitive answer to that question.
`
`In another case, In re Ben Huang, the United States Court of Appeals
`upheld a finding by the Board of Patent Appeals and Interferences that the
`pending claims made by Huang for a patent covering a particular kind of
`
`tennis racket grip were obvious, in part through a finding that Huang had
`not presented sufficient evidence of commercial success.13 In this case, the
`patent holder had cited several factors that he claimed were indicators of
`
`commercial success, including 1) sales of over 1 million units for use on
`
`both new and resold rackets and a) the fact that since Huang began selling
`
`the claimed grip, sales of his company’s prior grips had decreased by about
`50 percent. In this case, a relatively basic review of the economics of the
`
`claimed product by the patent holder would likely have provided a sounder
`
`basis for his claim. For example, from a review of the product and patent
`
`descriptions, it appears that development costs were likely quite low—the
`
`‘6 The discounted net present value (NPV) of a product that generates $10 per year in
`profits over the next three years will be greater than the NW of a product that gener-
`ates as next year. $10 the second year. and $15 the third year. even though both prod-
`ucts Will generate $30 in (undiscountedl profits.
`'7 Neupak, 41 Fed. Appx. 435.
`13 Huang, 100 F.3d 135.
`
`202
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017—01622
`Page 8 of 13
`
`

`

`COMMERCIAL SUCCESS: ECONOMIC PRINCIPLES
`APPLIED T0 PATENT LITIGATION
`
`patent claimed a change in the ratio of the thickness of the various materi-
`
`als used in the grip. If this were the case, then net profitability could have
`been reliably estimated for both the patented version of the product and
`
`the older version that it replaced. Assuming that Huang’s sales of the new
`
`grip Were not a result of discounting relative to the preexisting product
`
`and that manufacturing costs for the two products were similar, then a
`determination of commercial success could be made based on an evalua-
`
`tion of the increased revenues generated by the patented product relative
`to an appropriate benchmark (such as Huang’s revenues prior to introduc-
`
`tion of the new product or to revenues of competitors in the industry).
`
`Two Case Studies
`
`We were asked to evaluate and testify on commercial success issues in
`
`two recent cases. These cases provide an illustration of how traditional
`
`measures may be insufficient to prove commercial success, as well as an
`
`illustration of how, if properly applied, economic analysis can previde the
`
`complete picture.
`
`In the first case, we were asked to carry out research and testify on
`
`behalf of an accused infringer who was challenging the validity of a patent
`
`allegedly covering a particular type of packaged snack product. Despite
`
`rapid growth in sales of the product embodying the patented invention
`
`(approximately $30 million in revenues during the first year rising to about
`
`$110 million by the fourth year) and attainment of a substantial share of any
`
`reasonably defined market, we identified several key facts that nonetheless
`
`indicated that the patent may not have been a commercial success.
`
`Our first concern was that, although revenues were increasing rapidly,
`
`the trend in profits was not so promising. As shown in Figure 1, the com-
`
`pany experienced a cumulative net operating loss of approximately $10
`
`million to $15 million during the first five years of the product’s life cycle.
`
`Moreover, the trend through the last two years was downward—offering
`
`no indication that profits would be forthcoming in the near future.
`
`Furthermore, our analysis found that sales of the product were coming, in
`
`part, from customers that were switching from other snack products man—
`
`ufactured by the same company. We estimated that an additional $13 mil-
`
`lion in profits had been lost due to "cannibalization” of other product
`
`lines. These data indicated to us that although the product apparently had
`
`been deemed a success in the marketplace by consumers, it did not appear
`
`to be a commercial success from the perspective of the patent holder.
`
`203
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`Page 9 of 13
`
`

`

`ECONOMIC APPROACHES TO INTELLECTUAL PROPERTY POLICY,
`LITIGATION, AND MANAGEMENT
`
`
`
`Figure 1. Snack Product Patent Holder Profitability,
`Annual and Cumulative
`
`10.000.000
`
`5.000.000
`
`0
`
`(5,000,000)
`
`(10.000.000)
`
`(15,000.000)
`
`
`
`.
`
`Development
`Development
`:1.)
`Expenses
`Expenses
`:
`
`
`1998
`
`1999
`
`2000
`
`2001
`
`2002
`
`2003 Cumulative
`Total
`
`n.
`
`a2 .
`
`3..2
`“6
`
`E 3
`
`‘
`33
`3.O
`
`A second major concern related to the issue of the putative nexus
`
`between the revenues earned by the company and the patented invention.
`
`In this case, a competitor had entered the market one year after the
`
`patent holder with a product—eacknowledged to be noninfringing and
`
`apparently not protected by any other patent or critical trade secret—that
`provided virtually the same benefits to the consumer as the disputed
`
`product, including such characteristics as ease of preparation, portion
`
`control. and shape of the package (important for product placement on
`
`the store shelf). Moreover, as shown in Figure 2. this product experienced
`
`a path of revenue growth almost identical to the product at issue. The
`
`patent holder claimed that the product embodying the patented invention
`
`was one way to achieve the benefits cited by customers. However, despite
`
`the dramatic growth in revenues, in our opinion the performance of this
`
`alternative product demonstrated that causation had not been established.
`
`Based on information we reviewed, it appeared that rapid growth in rev-
`
`enues and market share for products ofthis type were not dependent on
`
`the patented invention. Finally, we pointed out that the patent holder had
`
`a very well-known brand name and had used innovative techniques to
`introduce and market the product at issue. These factors further weak—
`
`ened the link between the patented invention and any success (at least in
`
`terms of gross revenues) that the product had in the marketplace.
`
`204
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017—01622
`Page 10 of 13
`
`

`

`COMMERCIAL SUCCESS: ECONOMIC PRINCIPLES
`APPLIED TO PATENT LITIGATION
`
`
`
`Figure 2. Snack Product Annual Revenues Based on Number
`of Years Since Product Introduction
`
`I Competitor
`
`Patent Holder
`
`120.00 0,000
`
`
`
`Revenues(Dollars)
`
`60,000,000
`
`40,000,000
`
`100,00 0,000
`
`00,000,000
`
`20.000.000
`
`1st Year
`
`2nd Year
`
`3rd Year
`
`4th Year
`
`The facts were very different in the second patent dispute in which we
`
`testified for the patent holder regarding the Commercial success of an anti-
`
`infective drug. First, there did not appear to be much dispute about the
`
`nexus between the patented invention and the product’s efficacy, since—as
`
`one might expect with pharmaceutical products—the patent disclosed the
`
`drug itself and its methods of use. Second, although the accused infringer
`
`contended that the product’s success was due to advertising and promo-
`
`tion, that argument was weakened by (1) physicians’ testimony and other
`
`evidence that while promotional activities may well lead physicians to try
`
`a new product, repeated prescribing for patients is likely only if the prod-
`
`uct performs well; and (2) our analysis, which shOWed that the product at
`
`issue had the second-lowest ratio of promotional spending to sales of all
`
`major anti-infective products introduced in the past decade.
`
`Traditional metrics, such as growth in market share, also pointed to
`
`the product’s commercial success, as did a direct comparison of profits
`and R&D expenditures. Figure 3, for example, shows that after just four
`
`years on the market, the product ranked fourth among all oral tablet
`
`antibiotics, a market that included well over zoo products.
`
`205
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`
`Page 11 of 13
`
`

`

`ECONOMIC APPROACHES TO INTELLECTUAL PROPERTY POLICY,
`LITIGATION, AND MANAGEMENT
`
`
`
`Figure 3. Revenues for Best-Selling Competing Brands in
`the Oral Tablet Antibiotic Market
`
`- - Patented Product — Products In the
`Same Class
`
`—- Products in the
`Other Classes
`
`Dollars)
`
`Revenues(Constant
`
`1991
`
`1992
`
`1993
`
`1994
`
`1995
`
`1996
`
`199?
`
`1998
`
`1999 2000 2001
`
`2002 2003
`
`Although competition among antibiotics spans several classes of
`
`drugs, each class has a unique mechanism of action and therefore repre-
`
`sents a distinct market segment that should be examined as part of an
`
`evaluation of the commercial success of a patented invention. As Figure 3
`
`also shows, sales of the patented product grew faster than any other com-
`
`peting antibiotic. Figure 4 shows that within drugs of the same class, the
`
`market leader began losing share as soon as the patented product was
`
`introduced. The product’s rapid acceptance as the treatment of choice for
`
`dangerous infections such as hospital-acquired pneumonia demonstrated
`
`both the product’s commercial success and the importance of its per-
`
`formance characteristics (since no amount of advertising or promotion
`
`would be likely to influence the use of a product in life-threatening situ-
`
`ations). The huge sales of the product were even more impressive in light
`of the long odds against success in the pharmaceutical industry” and a
`history of failed attempts to develop safe and effective anti—infective
`
`drugs, leaving no doubt in our minds that the product and the patented
`invention were commercial successes.
`
`206
`
`__‘
`
`UNITED THERAPEUTICS, EX. 2079
`WATSON LABORATORIES v. UNITED THERAPEUTICS, IPR2017-01622
`
`Page 12 of 13
`
`

`

`COMMERCIAL SUCCESS: ECONOMIC PRINCIPLES
`APPLIED TO PATENT LITIGATION
`
`
`
`Figure 4. Shares of Revenue for a Class of Antibiotic Drugs
`
`90
`
`
`
`
`
`MarketShare(Percentage)
`
`80
`
`'30
`
`40
`
`Patented Product
`
`3'3
`
`20
`
`10
`
`60 50
`
`
`1991
`
`1992
`
`1993
`
`1994 1995 1996 1997
`
`1998 1999 2000 2001 2002 2003
`
`Conclusions
`
`Based on our understanding of the purpose of patent protection and our
`
`interpretation of precedents, it is our opinion that commercial success
`should be evaluated on the basis of the economic contribution of a
`
`patented invention to an inventor’s financial well-being. Thus, from the
`perspective of economics. a key indicator of commercial succeSS ought to
`
`be the profits gene rated by the patented invention, relative to an appro-
`
`priate benchmark or alternative. When available, financial data on these
`factors should be considered in an evaluation of commercial success.
`
`Courts’ historic use of factOrs such as revenue growth and large market
`
`share is likely to be consistent with this standard in most cases, if
`
`applied correctly.
`
`1° Recent studies estimate that only one out of every 5.000 potential drug compounds
`synthesized during applied research ends up being marketed. See Alfonso Gambardella.
`Science and Innovation: The US. Pharmaceutical industry During the 19805 {Cambridge
`University Press. 1905}, 20; and Pharmaceutical Research and Manufacturers of
`America. Pharmaceutical Industry Profile 2003, 1-6. Only 3 of every 10 approved drugs
`have sales revenue that exceed the average after-tax development costs of a new drug
`product. See Pharmaceutical Research and Manufacturers of America,“The Lengthy
`and Costly Challenge of Drug Development