# Oral argument — In Re The Procter & Gamble Company > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A24386 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** August 4, 2016 ## Text I'm going to take this off. Okay, it's okay. Thank you. Oh. I'm close to joining you, like, early. Okay. Okay. Anything you say is going to make me feel bad. This is a shampoo case. Good morning. On that note, good morning, Your Honors, and may it please the court. Good morning. The core issue in this case is going to be claim construction. Procter & Gamble's 569 patent claims a novel dandruff shampoo, and it does so with great precision and great specificity. First, the claim specifies six components that must be present in the shampoo in specific percentage by weight ranges. But that's not all. The claims also set forth four specific properties that the composition must possess. These properties are characterized in the patent by precise index values that the composition must possess in order to meet the claim. These index values have no meaning, none, extrinsic to the 569 patent. They were coined by the inventors, and the test protocols were designed by them. And the specification contains excruciating detail, excruciating detail that the director has criticized us for, I might add, with regard to how these tests are to be conducted and how the index values are to be calculated. Now, what's the purpose of these index values? They define a sweet spot, a sweet spot that optimizes the properties of the claimed dandruff shampoo so that it kills dandruff but still leaves hair feeling clean, among several other things. The novelty and non-obviousness of this invention is the creation of this anti-dandruff shampoo that does not involve tradeoffs in conditioning performance on the one hand and clean hair feel on the other in anti-dandruff performance, yet again, and the other properties that are set forth in the index values. Now, what the board did in this case was to decide that these four specific indices were not actually required by the claims even though they are set forth right there. Instead, the board told us, they are merely proxies for a shampoo, and I quote the board here at A7, proxies for a shampoo that has been optimized for corresponding beneficial shampoo properties. Now, it was based on this claim construction that the board found that the Canabo reference rendered these claims obvious. This was, in short, a legal error. This court's cases time and again tell us that an inventor's definitions have to be honored. And there's no question here that these four index values and, moreover, the specific tests that are used to define them were defined in the 569 patent using the unmistakable language of definition. You start in the abstract at A542. The four index values are, quote, defined herein. In the summary of the invention at A543 column 2, the present invention relates to a shampoo composition having the four index values, quote, as defined herein. You go to the detailed description of the invention at A544 column 3. More particularly, these compositions exhibit certain characteristics which are measured by four indices defined herein. In the analytical methods that are set forth at A59 at column 33, starting there, the methods to be employed, the definite article, the methods to be employed for determining the values of these indices are described in detail below. And this is followed by 11 columns of detailed descriptions of the methods. So let me tell you something I'm concerned about. Sure. You've got 11 columns that you're reading into these claims when you construe those claims. And I just, it's a lot to read into the claim constructions. And I'm just wondering, are you aware of any cases where something like this has happened before? Well, Judge Stoll, I might take issue with the characterization of reading into the claims. Because that's what the board said, and that's what the director is saying in this case. What we do have in the claims, though, are specific index values. Now, in a lot of cases, for example, you might take a look at the, I think it's the Callaway golf case, for example, recently decided by this court. An ASTM standard that was extrinsic to the patent was incorporated, in essence, by reference. But it was decided by the court that that was the standard that was going to be used. In the Invitrogen case, which is also cited in our briefs, the specification referred to a specific gel assay for measuring what was called RNase H activity. The court held in that case that it was that test that had to be used to define the particular value that was claimed in the claim. In those two cases, you've got a situation where you're going to an outside test. You're saying, OK, this index value is something that's determined by this standard that's used in the industry. This is a little different. I don't know if it's a meaningful difference. But this is a little different in that you're telling us to go to specific techniques that are described in the 11 columns in the specification. So I think, Judge Stoll, the Invitrogen case doesn't meet your description there. I think Invitrogen, the test was actually set forth in the specification itself. Callaway Golf has an external ASTM test. I was just beginning to mention the Cheney against PPG case, which is cited in our brief. There was an industrial standard called DIN 53583. That was extrinsic. Whether it's extrinsic to the patent or intrinsic, the point is that it's defined. I don't think there's a meaningful difference. I don't think there's a legal difference. In fact, quite the contrary, I think that it's probably better to set forth in detail in the patent the precise way to do the tests. And that's what these inventors did. And by the way, to the extent that it's helpful to understand what these tests are about, Judge Stoll, these tests are about comparing the composition to a control. Each of them is compared to a control. And the control is basically what was as good as existed, at least in these inventors' minds beforehand. So the idea was to come up with something that was actually better in these various categories than what existed before. Just turning back very briefly to the language of definition that's in this patent, there are no fewer than 11 specific places where it says that these index values are defined in the patent, defined herein. In cases that we've cited like AstraZeneca and Brain Tree Labs, both of those cases involved one reference to definition. Defined below in AstraZeneca, Brain Tree Labs, as used herein. That was enough for a clear lexicography in that case. Now, Judge Stoll, let me go back to your concern about reading in to the claim, which I think is really not what's happening here. The specification is used exactly as it's supposed to be used to define and help the person of ordinary skill understand what the four index values that are in the claims means. But I want to make a comment about what the consequences of the director's arguments and the board's holding in this case might be. Now, the director, in her brief, acknowledges the defined herein language in the patent, but actually denies that this constitutes lexicography. Well, cases like AstraZeneca and Brain Tree say, and actually even Phillips would be a very good authority for this as well, with regard to the meaningfulness of a clear definition that's set forth in the specification. This is lexicography. The director doesn't cite any cases for the proposition that this isn't. And instead, what Judge Stoll, the director, says is, well, no claim should ever be this long. This is, quite frankly, an astonishing argument. And honestly, if this compared to some of the biotechnology cases that come before this court, even if you were to incorporate all of this and say, well, this is actually part of the claim, this would be a relatively reasonably linked claim compared to some of those biotech cases. But the more important point is that this argument, which we've called an astonishing one with a very purposeful understanding of what that word means in our brief, is that it would encourage inventors to be less specific in their patents, to not specify as much in the specification. And I would have thought that Section 112 encouraged exactly the opposite. Construction, by the way, though, is not the same as incorporation. Would it instead encourage applicants to put the details into the claim as opposed to not having them at all in the specification? I'm not sure it would, because the criticism is that the tests are incorporated into the claim. And remember, the line from the director's brief is no claim should ever be this long. So the director's position is actually, no, we don't want it in the spec, and we don't want it in the claim either. We're giving the public what 112 says we should be giving them here. And the director's argument to the contrary is really just an invitation to create what is effectively a reverse Honeywell situation. We've talked about the Honeywell case. This is not an indefiniteness case, but Honeywell is. And the reason that Honeywell was indefinite was that they set forth a particular numerical limitation that had to be determined by a test, but they didn't tell what the test was. And there were at least four different tests in the prior art which gave different results. And that's why you need to specify the tests in the patents like this. Do you think there's any variability in these tests that you have? For example, the one where you have three people who are trained to feel the ponytail in a particular way. Is there variability there in the subjective review by these people who are hired to feel the ponytail? I think it's only subjective insofar that it's dependent on human senses, which might have some variability. But again, the control really should minimize or eliminate any subjectivity or variability there, because it's just a matter of determining whether it's better than the control with regard to the particular field. If I could just wind up very quickly here, and I'll give some time back to the court. We believe that the correction of this claim construction error resolves this case. The Canabo reference contains none of these index values, and there's no proof in the record, and the board did not find that the claims as correctly defined were obvious over the Canabo reference. And the director doesn't offer any argument for affirmance or remand under our construction, and particularly so in light of the fact that this is administrative review where the actual petitioner from the IPR is no longer in the case. I think the Magnum oil tools case, recent decision from this court, may say something about the fact that there's no point in remanding at this point. The record is the record, and there is a total and absolute failure of proof of any unpatentability under the correct claim construction. I'll give the rest of my time back to the court and come back on rebuttal, unless the court has further questions. Thank you, counsel. Thank you. Thank you, Your Honors, and may it please the court. I have a few points. The board's construction was reasonable, and I can walk through the board's opinion how it addressed it in two ways. One, at page 7, it said, well, the properties are definitely part of the claim, and those properties are in Canabo and other prior art references as well-known properties, anti-dandruff, conditioning, combability, and anti-dandruff again. And the board also, if that wasn't enough, the board compared the earlier tests, and Procter & Gamble only challenged two of the wearing clauses as to not having a prior art companion test. And the board compared the Shin and P&G's anti-dandruff test, and those tests have great commonality. They have the same dandruff organism put on an agar plate to incubate for a period of time, the same anti-dandruff agent, ZPT, skin is washed with the shampoo, the skin is then placed on the agar plate, and then the distances of death for the organism are analyzed. And the board also discussed the only other challenged wearing clause by P&G to the board, and that was the feeling clean property test, and both Canabo and P&G do this similarly. The board said the tests were comparable. Both Canabo and P&G use about 10 people to run their fingers through shampooed hair. These are feeling clean sensory tests done by people. Another point I'd like to make is that in Vitrogen, it's very different from this situation. On page 1077 of that case, the court says that it's only looking to about 13 lines of the patent owner's specification there. And so the test there was really specific, really concrete, and a lot shorter than what we have here. Here we have matters that are subjective. For example, respectfully inviting the court's attention to page 560 of the record in column 36. Under line 30, we have something called a force-time curve. A force-time curve is generated. The sum of the integration of these force-time curves is then generated. And more subjectivity in the next column, combing forces within some range of the mean combing force, typically two standard deviations from the mean. I just have two more examples. An analog in the next column in the middle around line 25. Are you in column 38? I'm sorry. Now I'm in, yes. Got it. Thank you, Your Honor. In column 38, about line 23, we learn about an analog digital interface to accumulate and integrate the force-time data. And about 20 lines down at line 46, we now learn that we have to use the Fisher's least significant difference procedure. So we probably have to look to more textbooks or things like that to get to what that is, the Fisher's least significant difference procedure. So in short, what we have here are comparable property tests in the prior art. And we have very broad ranges here in the claim with no showing of criticality for any of the claimed ranges. Accordingly, the board's decision should be affirmed. Unless the court has any further questions. Thank you. Thank you, Your Honor. I think I have three points to make with regard to rebuttal here. First, my friend started with the reasonableness of this claim construction. I don't have no case from this court that says it's reasonable to say that specific claim limitations can be read out of a claim and instead substituted with something that's comparable. Secondly, I heard just at the end of my friend's argument here talking about how there was no proof, no indication. P&G only showed this, that, or the other thing. We didn't have any burdens. We were the respondent in this IPR. The failures were on the part of Unilever, which did not test the Knievo example 10 using our rubric to prove that anything was comparable. They said it was comparable because there were some things in the prior art that are generally aimed at dandruff prevention and generally aimed at conditioning. But as we pointed out in our briefs, it's not just these four index values were available to the inventors. There were at least 46 different qualities that could have been subjected to this claim in just two references. The Hachowski and Sorkin references. And with 46 different choices, that's a possible 70 trillion options with regard to what could have gone into the claim. It's over 70 trillion. Even if you told the inventors, well, you have to pick four, it would still be over 160,000 different options. It was the picking of these precise four, then the designing and implementation of the particular limitations that are set forth in the index values that constitute the invention. What Unilever was able to do before the PTAB was, without any testing, convince the PTAB that comparable tests, comparable finding made on the same sort of argument that is being made here, can substitute for applying the plain claim language. That's not permitted under any case that I can possibly think of. And the director and the board did not cite any for that proposition either. Finally, what I heard was that there was no showing of criticality of any of the claimed ranges. Again, if there was no showing of criticality of the ranges, that was a showing that Unilever needed to make. Or they needed to show that it was not critical. The claim tells us what's critical here because the claim is the invention. And in this case, I do not see how it is possible for the board's decision to be affirmed under broadest reasonable construction because it's not reasonable. It reads half of the claim out. It's not possible, certainly wouldn't have been possible under the Phillips standard. And the director cites no case that suggests that comparable tests instead of the ones actually set forth in a patent can be used to prove what is in the claim. Unless the court has further questions, we'll give that 15 seconds. Thank you. Thank you, counsel. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A24386. Public record. Not legal advice.