Eli Lilly Canada Inc. v. Novopharm Limited

Eli Lilly Canada Inc. v. Novopharm Limited
Court (s) Database
Federal Court Decisions
Date
2009-10-05
Neutral citation
2009 FC 1018
File numbers
T-1048-07
Decision Content
Date: 20091005
Docket: T-1048-07
Citation: 2009 FC 1018
Ottawa, Ontario, October 5, 2009
PRESENT: The Honourable Mr. Justice O'Reilly
BETWEEN:
ELI LILLY CANADA INC., ELI LILLY AND COMPANY,
ELI LILLY AND COMPANY LIMITED
AND ELI LILLY SA
Plaintiffs
(Defendants by Counterclaim)
and
NOVOPHARM LIMITED
Defendant
(Plaintiff by Counterclaim)
REASONS FOR JUDGMENT AND JUDGMENT
I. Overview
[1] The plaintiffs are suing Novopharm for infringement of a patent for a medicine called olanzapine. Its brand name is Zyprexa. Olanzapine is regarded as a relatively safe, and often effective, medicine for treating schizophrenia. Olanzapine is widely prescribed and is a commercial success. It is the subject of a Canadian patent (No. 2,041,113) that I will refer to as the ‘113 patent. I will refer to the plaintiffs collectively as “Lilly”. Lilly applied for the ‘113 patent in 1991 and was granted it in 1998.
[2] Olanzapine was included within an earlier Lilly patent (No. 1,075,687) (the ‘687 patent). The main question before me is whether Lilly was entitled to a separate patent and a further monopoly on olanzapine.
[3] The ‘687 patent was a so-called “genus patent”. It covered 15 trillion compounds all with a similar chemical structure – three-ring molecules called “thienobenzodiazapines”. The ‘113 patent is a so-called “selection patent” which identifies an already-patented compound for separate patent protection based on the fact that it allegedly manifests unexpected, substantial and special properties in comparison with the other members of its chemical family.
[4] For ease of reference, I have set out relevant statutory provisions in Annex “A”, a summary of expert witnesses’ background and qualifications in Annex “B”, and a glossary of terms in Annex “C”.
II. Earlier Proceedings
[5] The ‘113 patent was litigated in two previous applications before the Federal Court under the Patented Medicines (Notice of Compliance) Regulations, SOR/98-166, as amended SOR/93-113. In the first, before Justice Johanne Gauthier, Lilly sought and obtained an order prohibiting the Minister of Health from issuing a Notice of Compliance (NOC) that would have allowed Apotex Inc. to make and sell olanzapine (Eli Lilly Canada Inc. v. Apotex Inc., 2007 FC 455). Apotex had alleged that the ‘113 patent was invalid on grounds of anticipation, obviousness, double patenting, and violation of s. 53 of the Patent Act, R.S.C. 1985, c. P-4. Justice Gauthier found that Apotex’s allegations were not justified. However, she also found that the question whether the ‘113 was a valid selection patent was not properly before her as it had not been specifically alleged by Apotex. The Federal Court of Appeal confirmed that finding: Apotex Inc. v. Eli Lilly Canada Inc., 2008 FCA 44.
[6] In a separate proceeding under the Patented Medicines (Notice of Compliance) Regulations, involving, essentially, the same parties as are before me, Justice Roger Hughes found that Novopharm’s allegation that the ‘113 patent’s disclosure was insufficient was justified. He refused to issue an order prohibiting the Minister from granting Novopharm an NOC to enter the olanzapine market (Eli Lilly Canada Inc. v. Novopharm Limited, 2007 FC 596). Soon thereafter, Novopharm obtained its NOC. Lilly launched an appeal, but the Federal Court of Appeal found the proceeding to be moot, given that Novopharm had already obtained its NOC (Eli Lilly Canada Inc. v. Novopharm Limited, 2007 FCA 359). By then, Lilly had commenced this action for infringement of the ‘113 patent.
[7] Justice Hughes addressed the issue that was not before Justice Gauthier – whether the ‘113 is a valid selection patent; in particular, whether the ‘113 patent’s disclosure is sufficient. He found that the ‘113 patent failed to describe what the “surprising and unexpected properties” of olanzapine were in comparison with the other compounds of the ‘687 patent. Before the Federal Court of Appeal, Lilly argued that Justice Hughes erred by requiring patents to set out comparative data. However, the Court did not agree that Justice Hughes had clearly stipulated that comparative data were required in selection patents.
[8] The essence of Justice Hughes’ decision is set out in his paragraph 162:
I find that the ‘113 patent fails to provide sufficient disclosure in its specification as to the invention, if any, in selecting olanzapine from a previously disclosed group of compounds. The prior art British Patent says that the whole class of compounds [is] to be useful in treating central nervous system disorders. The invention in selecting olanzapine is the so called “surprising and unexpected” properties of olanzapine in “comparison with flumezapine and other related compounds”. No such comparison is made anywhere in the ‘113 patent. No data was given. We are left only with rhetoric such as “high level of efficiency” and “mild and transient” and “lower” side effects. The puzzling and scant mention of a dog study refers only to ethyl olanzapine and tells nothing of flumezapine or other compounds.
[9] The compounds Justice Hughes was referring to in this passage (flumezapine and ethyl olanzapine) are other compounds from the ‘687 patent. His conclusion was that the ‘113 did not adequately distinguish between olanzapine’s qualities and the characteristics of the previously patented family of compounds, of which olanzapine was a member. As will be seen below, I disagree with Justice Hughes’ conclusion that the ‘113 does not contain any comparisons between olanzapine and the other ‘687 compounds. But I agree with him that the ‘113 patent does not adequately describe the alleged advantages of olanzapine. In fact, I conclude that in 1991 Lilly did not have evidence supporting the alleged advantages of olanzapine over the ‘687 compounds.
III. Issues
[10] The main issue before me is the validity of the ‘113 patent. Novopharm mounted its attack on numerous grounds, but I am persuaded that the main one – that the ‘113 is not a valid selection patent – is supported by the preponderance of evidence and, therefore, I deal with the others briefly at the end of my reasons. Novopharm’s other grounds for challenging the ‘113 patent included anticipation, double-patenting, wrong Inventorship, obviousness, s. 53 of the Patent Act (misrepresentation), and s. 73 of the Patent Act (deemed abandonment).
[11] There is no serious issue about infringement. Novopharm admits that, if the ‘113 patent is valid, it is infringing it by marketing a generic version of olanzapine.
[12] Lilly bears the burden of proof on a balance of probabilities on the infringement issue. Novopharm bears the burden of proof in respect of invalidity. In assessing whether it has met that burden, I must bear in mind that the Commissioner of Patents already concluded, by virtue of issuing the ‘113 patent to Lilly, that the patent is valid. The Commissioner’s conclusion is entitled to some deference; I can overturn it only if the evidence in Novopharm’s favour shows that it was unreasonable (Apotex Inc. v. Wellcome Foundation Ltd., 2002 SCC 77, at para. 41-44).
[13] I find, on a balance of probabilities, that the ‘113 patent is invalid and, therefore, not infringed.
IV. Treating Schizophrenia
[14] Schizophrenia is a form of psychosis affecting about 1 percent of the population. Generally speaking, symptoms fall into two categories. The first, called “positive” symptoms, includes hallucinations and delusions. The second, called “negative” symptoms, includes withdrawal, lack of motivation and impaired mental functioning.
[15] There is no known cure for schizophrenia. However, over the course of the past fifty or sixty years, scientists have found some drugs that mitigate some of the worst symptoms. The drug chlorpromazine was a breakthrough in the early 1950s. But the problem with chlorpromazine was that it came with a serious side-effect liability. In particular, it induced an array of uncomfortable motor effects called “extra-pyramidal symptoms”, or EPS. EPS include restlessness, stiffness, twitching and facial contortions. Chlorpromazine and analogous drugs that share this liability to cause EPS are referred to as “typical” or “first-generation” antipsychotics.
[16] In time, scientists discovered a better drug, called clozapine, which came onto the market in the late 1960s. Clozapine’s main advantage was that it did not induce EPS. However, after years on the market, it was found to cause a rare but serious blood disorder, called agranulocytosis, in which the body abruptly stops making white blood cells. Clozapine was taken off the market in the 1970s, but returned in the late 1980s. Patients taking clozapine must take frequent blood tests to ensure their white blood cell count is normal. Clozapine and other drugs with a low EPS liability are referred to as “atypical” or “second-generation” antipsychotics.
[17] Once clozapine was off the market, many scientists, including those at Lilly, were looking for a safe, clozapine-like compound - one that would treat both the positive and negative symptoms of schizophrenia, have little liability to produce EPS, and not affect production of white blood cells.
[18] Various tests can be used to determine a compound’s potential as an antipsychotic. The same tests have been used for decades. Compounds are tested in mice to see if they reduce locomotor activity and induce hypothermia (good signs for an antipsychotic). A compound’s capacity to block a conditioned avoidance response (CAR) in rats is of interest because it, too, indicates antipsychotic activity. Essentially, a CAR test measures a compound’s ability to interfere with rats’ learned behaviour (e.g., avoiding electric shock). On the other hand, a compound’s liability to induce catalepsy (CAT) in rodents is an important indicator of its liability to produce EPS in humans. A compound will be a promising atypical or second-generation antipsychotic if it shows good CAR-CAT separation (i.e., its CAR score is high and its CAT score is low).
V. The Story of Olanzapine
[19] The story of the development of olanzapine begins with a discussion of the family of compounds of which it is a member and the corresponding patent for that family – the ‘687 patent. As described above, in the 1970s, the search was on for a safe clozapine. Lilly was exploring compounds that were chemically similar to clozapine as part of that quest.
[20] Having heard about clozapine and its potential as an antipsychotic, Dr. Jiban Chakrabarti, a Lilly chemist, attended a conference in Prague in the early 1970s. He met the scientists who had made and developed clozapine. Dr. David Tupper, another Lilly chemist, recalls that, when Dr. Chakrabarti returned, he was very excited about what he had heard there. He believed that he could make compounds that would be clozapine-like in terms of their antipsychotic effect but would avoid the issues encountered with clozapine. Dr. Chakrabarti suggested replacing one of clozapine’s phenyl rings with a thiophene ring.
[21] Dr. Tupper, after visiting the library and determining that no such compounds had previously been made, worked out ways to synthesize them. The end result was the family of compounds covered by the ‘687 patent.
[22] The ‘687 patent was filed in 1975 and issued to Lilly in 1980. Its inventors were Dr. Chakrabarti and Dr. Tupper, both of whom worked at Erl Wood, Lilly’s research facility in Sussex, United Kingdom. The ‘687 patent described a “novel class of compounds” called “thienobenzodiazapines” with a three-ring chemical structure, similar to clozapine’s. The patent asserted that this family of compounds had displayed useful central nervous system activity in animal tests, and had potent neuroleptic, sedative, relaxant and anti-emetic properties. They showed good CAR-CAT separation. These properties, the patent stated, rendered the compounds useful in the treatment of mild anxiety states, and certain kinds of psychotic conditions such as schizophrenia. Further, the compounds had a high therapeutic index (meaning that there was a wide margin between the effective dose and a gross toxic effect) and were effective across a broad dosage range (from 0.1 mg/kg/day to 10 mg/kg/day).
[23] The focus of the ‘687 patent was clearly on the compounds themselves – their constituents, their structure and the processes by which they could be made. Still, the patent specifically asserted that the compounds’ utility lay in their potential use in the treatment of central nervous system disorders, including schizophrenia. Dr. Ian Pullar testified that Lilly was hopeful that the class of compounds described in the’687 patent would be effective in treating both the positive and negative symptoms of schizophrenia, and have low EPS liability. One of the vast number of compounds covered by the ‘687 patent (15 trillion) was olanzapine. In fact, it fell within a group of the “most preferred compounds” of the invention, although it was not specifically identified.
[24] Over the years following the filing of the ‘687 patent, Lilly scientists worked at bringing some of the compounds of the invention to market. A few dozen were synthesized and tested in vitro. Dr. Chakrabarti published a paper in 1980 that gave data on 76 of the ‘687 compounds, including their CAR and CAT values. From that study, flumezapine and ethyl flumezapine, which had been specifically identified in the ‘687 patent, looked promising. A few others also looked favourable, but Dr. Chakrabarti noted that the “profile of activity needs further development of this class of compounds” (D-39, at p. 883).
[25] Lilly began tests on ethyl flumezapine, but this work was wound down in 1978 after dog studies showed the compound caused a reduction in white blood cells, just as clozapine had been known to do, which was the major side effect sought to be avoided. At that point, Lilly turned its attention to flumezapine. Dog studies on flumezapine did not show any problem with white blood cells, although other problems were detected – weight loss, anemia and elevated prolactin. Still, in due course, in 1981, Lilly was granted permission by the U.S. Food and Drug Administration (FDA) to administer flumezapine to healthy volunteers, and then to begin clinical trials with patients experiencing schizophrenia.
[26] Lilly halted its clinical trials on flumezapine in April 1982 after receiving reports of elevated liver enzymes and a muscle enzyme called creatine phosphokinase (CPK) in some patients. Lilly passed on those reports to the FDA who asked Lilly to discontinue treating patients with flumezapine.
[27] Lilly decided not to resume the clinical trials on flumezapine, even though there were signs that it was an effective antipsychotic. Investigators “were very impressed with the efficacy of the drug, as well as the significant absence of extrapyramidal side effects . . .” (D-84). Lilly could have changed the clinical trial protocols, for example, by reducing the maximum dose, or monitoring patients more closely for liver enzymes and CPK. Dr. Paul Leber, an FDA official who was involved in the discussions about flumezapine at the time, testified that the FDA did not halt Lilly’s development of flumezapine:
It was simply an assertion that, in the current state, they should not do further clinical testing until they submitted new reports to us and we reviewed them. Then we would explain to them what they could or could not do.
[28] However, continuing with flumezapine would have required considerable time and effort. Lilly would have had to persuade the FDA to allow it to continue clinical trials. While the project team felt that studies of flumezapine should be continued, Lilly management concluded that further investment in flumezapine was unwarranted. Lilly discontinued, but did not completely abandon, its request to have flumezapine approved. Still, in effect, flumezapine was a tainted product.
[29] Dr. Pullar, who was the chairman of the flumezapine project team, described this as a “black time” at Erl Wood. Yet, the project team felt there was enough promise within the ‘687 compounds that they went looking for another candidate. There was pressure coming from Lilly management to show that the substantial corporate investment in developing an antipsychotic would pay off. Erl Wood, established in 1967, had only produced one drug in its 15-year existence that had actually made it to market.
[30] Within a few weeks of the discontinuation of flumezapine, Dr. Tupper and his colleague, Mr. Terrence Hotten, synthesized seven more compounds, one of which was olanzapine. Dr. Tupper felt that, given that flumezapine had gone quite far in its development, the focus should be on methyl compounds, not ethyl. Ethyl flumezapine had been an abject failure. At first, Dr. Pullar did not think that olanzapine would be a good choice for development because it did not show particularly dramatic potency in animal tests. But the rest of the team favoured olanzapine based on its overall performance on an array of animal and in vitro tests. Dr. Pullar now feels glad he was outvoted and, naturally, is proud of his association with a drug that treats many patients effectively. As he said, the team “carried out very good research in order to put [olanzapine] on the market”. Dr. Tupper expressed similar sentiments and cited numerous prizes the Lilly scientists had received for their work.
[31] So, by 1983, Lilly was satisfied that olanzapine showed potential as an antipsychotic. Studies continued and Lilly’s hopes were confirmed by further preliminary results. Beginning in 1986, Lilly gave olanzapine to healthy volunteers and, in 1989, started clinical trials in patients. By 1990, bringing olanzapine to market became a top priority for Lilly. It recognized a market opportunity, given that clozapine was about to be reintroduced, and new drugs, such as risperidone, were about to come on stream. A patent for olanzapine was filed in the United Kingdom in 1990 and in Canada in April 1991. Lilly was granted its Canadian patent, the ‘113, in 1998.
[32] By the time it filed the ‘113 patent, Lilly had received the results of its healthy volunteer studies, as well as some preliminary data from its clinical trials. It had also concluded a six-month study in dogs. The patent mentions these studies and provides some general information about what they disclosed.
VI. The ‘113 Patent
[33] The ‘113 patent proclaims a number of advantageous qualities for olanzapine. These can be grouped into two main categories. First, the ‘113 patent identifies certain advantages of olanzapine over the other compounds from the ‘687 patent. Second, the ‘113 patent boasts the superiority of olanzapine over other known antipsychotic drugs used in the treatment of schizophrenia and related conditions.
(a) Olanzapine’s advantages over the other ‘687 compounds
[34] The ‘113 patent says that olanzapine displays “surprising and unexpected properties” as compared to flumezapine and other related compounds. This broad statement involves a direct comparison between olanzapine and the other compounds of the ‘687 patent. There are four such comparisons in the patent.
[35] The patent notes that Lilly developed flumezapine to the point where it was administered to 17 patients. However, as mentioned above, due to an “unacceptably high incidence” of elevated liver enzymes (SGPT and SGOT), as well as creatine phosphokinase (CPK), Lilly discontinued clinical trials of flumezapine after discussions with the FDA. Later, the ‘113 patent states that patients treated with olanzapine experienced “a low incidence of only mild and transient elevation of liver enzymes”, and CPK levels were lower than with flumezapine. In my view, the ‘113 patent asserts the superiority of olanzapine over flumezapine in respect of both liver enzymes and CPK levels in patients receiving therapeutic doses.
[36] Early in the patent, the inventors explain that many antipsychotic drugs cause EPS. The patent also states that “[i]n clinical trials with flumezapine two of the patients showed the emergence of extra pyramidal side effects . . .” Immediately thereafter, it states “[w]e have now discovered a compound which possesses surprising and unexpected properties by comparison with flumezapine and other related compounds”. Then, further on in the patent, the inventors state that olanzapine “is less likely to induce extrapyramidal side effects in the clinic.” Taken together, I read these statements in the ‘113 patent as an assertion of the superiority of olanzapine over flumezapine with respect to its EPS liability.
[37] The ‘113 patent also mentions a dog study in which olanzapine was compared with ethyl olanzapine, another of the ‘687 compounds. The patent reports the outcome of the study as indicating “that four out of eight dogs showed a significant rise in cholesterol levels, whereas the compound of the invention did not show any rise in cholesterol levels” at a dosage of 8 mg/kg. To understand the scope and nature of this comparison, one must examine the dog study closely. Several experts gave testimony in this area. Their evidence is discussed in greater detail below. For present purposes, I interpret the statement in the ‘113 patent about cholesterol levels in dogs as an assertion of the superiority of olanzapine over ethyl olanzapine in terms of its liability for elevating cholesterol in humans. Ethyl olanzapine raises cholesterol; olanzapine does not.
[38] Therefore, the ‘113 patent declares olanzapine to be superior to the ‘687 compounds due to:
• lower incidence of liver enzyme elevations compared to flumezapine;
• lower CPK levels than flumezapine;
• lower EPS than flumezapine; and
• no increase in cholesterol compared to ethyl olanzapine.
(b) Olanzapine’s advantages over other antipsychotic drugs
[39] The ‘113 patent describes olanzapine’s function as an antagonist at various brain receptors (including dopamine at the D-1 and D-2 receptors, the 5HT-2 receptor and noradrenergic q-receptors). These characteristics suggest that olanzapine has potential as a drug with relaxant, anxiolytic or anti-emetic properties, and might be useful in treating psychotic conditions, including schizophrenia.
[40] In addition, the patent states that olanzapine shows “surprising and excellent results” in experimental screens and clinical trials. Further, olanzapine shows a “high level of activity in the clinical evaluation of psychiatric patients suffering schizophrenia” at doses lower than were expected based on animal models. The patent describes an open study in which six out of eight patients who had taken olanzapine for at least two weeks showed between 66% and 87% improvement on a well-recognized scale for measuring symptoms of schizophrenia (the BPRS scale). Further, preliminary results from three ongoing clinical trials confirmed a high level of efficacy at low doses (2.5 and 5.0 mg).
[41] These statements are followed by the broadest and stoutest assertion about olanzapine in the ‘113 patent:
Overall, therefore, in clinical situations, the compound of the invention shows marked superiority and a better side effects profile than prior known antipsychotic agents . . .
[42] Because of its placement – just after a description of particular side effects and olanzapine’s comparative advantages – and because of the introductory words of the sentence (“overall, therefore”), I interpret this assertion as being a general contention about the advantages of olanzapine in respect both of its efficacy and the particular side effects discussed in the preceding passages. It is not a statement asserting the superiority of olanzapine in respect of all possible side effects. In my view, a fair interpretation of the patent is that it asserts the superiority of olanzapine in respect of the side effects specifically identified in it, most importantly, the ones that presented the greatest concern to schizophrenia patients – EPS and agranulocytosis.
[43] However, I also believe that a skilled reader would infer from the broad declaration about olanzapine’s superiority with respect to side effects that the inventors had conducted sufficient tests to be able to have a fair idea about what olanzapine’s side effects liability was. This was not a warranty that worrisome side effects could not later come to light. It was a general statement intended by the inventors to assure skilled readers that olanzapine appeared genuinely to represent a significant advance in neuropsychopharmacology, the treatment of schizophrenia with antipsychotic agents.
[44] Therefore, with that implied assertion in mind, in my interpretation of the ‘113 patent, the advantages coming within this broad statement of olanzapine’s superiority over other antipsychotic drugs include:
• high level of efficacy at low doses;
• lower elevation of prolactin;
• lower EPS liability; and
• no alteration of white blood cell count.
[45] There is also an implied comparison with respect to liver enzymes, given the linkage made between flumezapine and chlorpromazine in that area. The patent says that “in respect of its tendency to raise liver enzyme levels, flumezapine is similar to chlorpromazine, an antipsychotic which has long been in use but whose safety has been called into question.” However, when I analyze the ‘113 as a selection patent below, I will deal with the assertion about liver enzymes within the discussion of the comparison with the ‘687 compounds, given the direct comparison with flumezapine.
(c) The ‘113 patent’s claims
[46] The claims in issue here are the following:
· Claim 3: Olanzapine
· Claim 6: The use of olanzapine for the manufacture of a drug for the treatment of schizophrenia.
· Claim 13: A pharmaceutical composition comprising olanzapine and a pharmaceutically acceptable diluent or carrier.
· Claim 14: A pharmaceutical composition in capsule or tablet form containing 0.1 to 20 mg of olanzapine.
· Claim 15: A pharmaceutical composition in capsule or tablet form containing 0.5 to 10 mg of olanzapine.
· Claim 16: A pharmaceutical composition in capsule or tablet form containing 2.5 to 5 mg of olanzapine and a pharmaceutically acceptable diluent or carrier.
VII. Is the ‘113 patent a valid selection patent?
(a) The requirements for a valid selection patent
[47] As discussed, the earlier ‘687 patent covered olanzapine, as well as a large number of other related compounds. By contrast, the ‘113 patent deals with olanzapine alone. In these circumstances, patent law considers the ‘113 to be a “selection patent”. A selection patent is valid if it discloses to the public something new and useful in exchange for a further monopoly on the already-patented compound. In other words, the question is whether the selected compound truly represents an invention that merits a separate and free-standing monopoly. An “invention” under s. 2 of the Patent Act is a “new and useful . . . composition of matter, or any new and useful improvement in any . . . composition of matter”. Just as with any other kind of patent, then, a selection patent must disclose an invention. What sets selection patents somewhat apart is that the inventor must disclose an invention over and above what was disclosed in the prior patent – the “genus” patent – covering the selected compound.
[48] Justice Marshall Rothstein recently summarized the requirements for selection patents in Apotex Inc. v. Sanofi-Synthelabo Canada Inc., 2008 SCC 61, [2008] 3 S.C.R. 265, relying on the well-known precedent of Re I.G. Farbenindustrie A.G.’s Patents (1930), 47 R.P.C. 289 (Ch. D.) where the following principles were set out:
1. There must be a substantial advantage to be secured or disadvantage to be avoided by the use of the selected members.
2. The whole of the selected members (subject to “a few exceptions here and there”) [must] possess the advantage in question.
3. The selection must be in respect of a quality of a special character peculiar to the selected group. If further research revealed a small number of unselected compounds possessing the same advantage, that would not invalidate the selection patent. However, if research showed that a larger number of unselected compounds possessed the same advantage, the quality of the compound claimed in the selection patent would not be a special character. (At para. 10.)
[49] Therefore, in order to uphold the ‘113 patent as a valid selection patent, I must be satisfied that olanzapine has an advantage over the other compounds of the ‘687 patent. Further, that advantage must be substantial and somewhat peculiar to olanzapine. Finally, the patent must clearly describe olanzapine’s substantial and special advantage. Justice Rothstein said that “the specification of the selection patent [must] define in clear terms the nature of the characteristic which the patentee alleges to be possessed by the selection for which he claims a monopoly” (at para. 114).
[50] Justice Rothstein described a selection patent as containing “only a bit of the subject matter of the original genus patent because that bit does something better than and different from what was claimed in the genus patent” (at para. 100). Does olanzapine do something better than and different from its ‘687 classmates?
(b) Analysis of the ‘113 patent as a selection patent
[51] In interpreting the patent, I must do so from the perspective of a person skilled in the relevant art and be guided by the expert evidence presented by the parties. A person skilled in the art for present purposes would possess a conglomeration of knowledge and experience in medicinal chemistry, toxicology, psychiatry, and pharmacology, as well as a capacity to interpret data from animal studies and appreciate their relevance to the treatment of human disease.
[52] As mentioned, there is a general assertion in the ‘113 patent that olanzapine is superior to the class of compounds covered by the ‘687 patent. It says that olanzapine displays “surprising and unexpected properties” as compared to flumezapine and other related compounds. As mentioned, there are four examples of superiority provided in the patent, in respect of two ‘687 comparator compounds (flumezapine and ethyl olanzapine). The patent says that olanzapine is better and different from those two compounds in the following respects:
(i) Olanzapine has lower elevations of liver enzymes than flumezapine,
(ii) Olanzapine has lower elevations of CPK than flumezapine;
(iii) Olanzapine has less EPS liability than flumezapine;
(iv) Olanzapine does not elevate cholesterol; but ethyl olanzapine does.
[53] As direct comparisons with members of the ‘687 patent, these assertions are obviously relevant to the determination whether the ‘113 is a valid selection patent. However, in my view, the comparisons in the ‘113 patent between olanzapine and “prior known antipsychotic agents” are also relevant here. In my view, reading the ‘113 patent as a whole, the skilled reader, aware of the ‘687 patent, would interpret the alleged superiority of olanzapine over other antipsychotic drugs on the market as being another major advantage of olanzapine over the other ‘687 compounds. The ‘113 patent refers to two of the ‘687 compounds and discusses their disadvantages. But it is clear that neither of them had been used for the treatment of schizophrenia or any other condition. By contrast, according to the ‘113 patent, not only could olanzapine be used for that purpose, it was, “overall”, markedly superior to, and had a better side effects profile than, other drugs on the market. As discussed above, the comparisons with other antipsychotic agents are:
(i) high level of efficacy at low doses;
(ii) lower elevation of prolactin;
(iii) lower EPS liability; and
(iv) no alteration of white blood cell count.
[54] According to Dr. Guy Goodwin, in addition to clozapine, the relevant drugs for these comparisons in 1991 would include chlorpromazine, haloperidol, flupenthixol, and perphenazine. In his view, these drugs provided choices but they presented similar side effects, most particularly EPS. Elevated prolactin was sometimes an issue, as well, but in Dr. Goodwin’s view, this rarely caused acute problems.
[55] Therefore, the first step I must take is to decide whether one or more of the asserted advantages of olanzapine was known to exist, or was soundly predicted, at the time the ‘113 patent was filed in 1991. Second, I must decide whether at least one of them could be considered a substantial advantage over the ‘687 compounds and somewhat peculiar to olanzapine. And, if so, the third question is whether the disclosure of that substantial and special advantage in the ‘113 patent was adequate. If I decide any one of them in the negative, I must find the ‘113 patent to be invalid.
(c) Were the alleged advantages of olanzapine known or soundly predicted in 1991?
[56] To understand whether the alleged advantages of olanzapine over the ‘687 class of compounds were known or soundly predicted, one must appreciate what was known about each of flumezapine, ethyl olanzapine and olanzapine in respect of the four relevant parameters at the time the ‘113 patent was filed in April 1991. Detailed results from the flumezapine clinical trial and surrounding documents were available. With respect to ethyl olanzapine, toxicology tests and the results of the dog study were available.
[57] For olanzapine, results of a four-week, open-label clinical trial (called E001) involving ten patients were available prior to the filing of the ‘113 patent. In addition, four studies of olanzapine in healthy volunteers had been completed. In three of them, four subjects had been given olanzapine, either in a single dose (Studies HGAA and HGAB) or over a two-week period (HGAC). In the fourth (E002), eight subjects were given 10 mg/day of olanzapine for a week. The ‘113 patent refers to preliminary results from other clinical trials that were ongoing in 1991, but I heard little evidence about what, if any, advantages over the ‘687 compounds could be gleaned from those trials. There were some early indications of olanzapine’s efficacy, but that is not one of the alleged advantages over the ‘687 class that was set out in the ‘113 patent.
[58] To repeat, a selected compound is an invention, and separately patentable, if the patentee can establish that, as of the filing date, the selection had a substantial and peculiar advantage over the previously-patented family of compounds, or if that advantage could be soundly predicted.
[59] A sound prediction must meet four criteria. First, there must be a factual basis for the prediction. Second, the inventor must have an articulable and sound line of reasoning from the factual basis to the desired outcome. Third, the inventor must set out in the patent the factual basis and the line of reasoning he or she was applying. Fourth, of course, the predicted outcome must be borne out by subsequent testing. An incorrect prediction cannot be a sound one. (See Apotex v. Wellcome, above, at para. 70, 76.)
(i) Olanzapine vs. flumezapine – liver enzymes
[60] Several patients on flumezapine experienced a rise in liver enzymes during its clinical trial. In all cases, the levels returned to normal either after the drug was discontinued or while still on drug. The project team characterized the liver enzyme rises as “mild”.
[61] Dr. Ronald Diamond carried out a comprehensive analysis of the data available for flumezapine and olanzapine. He found that several persons taking olanzapine experienced liver enzyme elevations. Investigators raised concerns about liver enzyme elevations in all of the olanzapine studies conducted before the ‘113 patent was filed. Some subjects were withdrawn from studies on that basis. While it is difficult to compare the flumezapine and olanzapine data sets (different doses, different patients, different time frames, different drug potencies), Dr. Diamond concluded that there was no evidence that olanzapine was superior to flumezapine in terms of its effect on liver enzymes. He also noted that many antipsychotic medications (e.g., chlorpromazine, clozapine, flumezapine, olanzapine and any of the other “-zines”) cause transient elevations of liver enzymes that are not of any clinical significance. Indeed, the ‘113 patent specifically refers to chlorpromazine’s effect on liver enzymes. Yet, it was well-known at the time that chlorpromazine’s tendency to cause these elevations was not clinically significant. Chlorpromazine was in wide-spread use.
[62] Dr. Alan Young observed that some of the patients and healthy volunteers who had been given olanzapine had some elevations of the liver enzyme SGPT and, less frequently, SGOT. He felt that these elevations could be aptly described as mild and transient, as the ‘113 patent does. However, he did not make any comparison to the flumezapine data. In fact, Lilly was aware, based on the healthy volunteer studies, that “both flumezapine and [olanzapine] have illustrated a propensity to cause elevations of liver enzymes” (D-435, p. 15).
[63] The evidence shows that olanzapine was not known in 1991 to have any superiority over flumezapine with respect to liver enzymes; nor was there a factual basis for a prediction that olanzapine would possess an advantage in that respect. Further, there is no evidence of a line of reasoning that would sustain a sound prediction that olanzapine’s liver enzyme liability would be lower than flumezapine’s. Neither a factual basis nor a line of reasoning appears in the ‘113 patent. We now know that olanzapine has a certain liability to raise liver enzymes. The olanzapine product monograph states that, in clinical trials, olanzapine “was associated with elevation of hepatic transaminases, primarily ALT (SGPT)”. However, none “manifested clinical symptomatology associated with liver impairment”. In other words, elevations of liver enzymes are generally not a health concern.
(ii) Olanzapine vs. flumezapine – CPK
[64] We know that concern about flumezapine’s risk for elevating CPK is what brought its clinical trial to a halt and led Lilly to start working on olanzapine. Clearly, flumezapine’s potential to elevate CPK was a serious concern both to Lilly and the FDA. As a muscle enzyme, CPK can be a harbinger of serious conditions such as neuroleptic malignant (NMS) or rhabdomyolysis.
[65] The evidence is not at all clear that flumezapine was, in fact, responsible for the elevations of CPK observed during its clinical trial. Dr. Diamond provided convincing testimony that the CPK results were more a product of the site where the elevations were seen than a consequence of administering flumezapine. He found it surprising that all of the CPK elevations were seen at a single site, not across the various clinical trial locations. He also noted some peculiarities about that site:
• One patient had high CPK even before being given flumezapine, yet his CPK levels were not checked again until the nineteenth day of treatment, when they had spiked to 5,500. They then dropped to less than 1,000 a few days later, at which point the treatment was discontinued.
• One patient was on a high dose of flumezapine for 20 days before his CPK level jumped to 5,000. It went down a few days later after the dose was reduced from 35 mg/day to 20 mg/day. The patient remained on flumezapine for a total of 57 days.
• One patient on a 20 mg dose of flumezapine had a CPK level of 6,300 on the 22nd day of the study. He was taken off the study ten days later.
• One patient had a high level of CPK at a 10 mg dose on the 10th day of the study. The patient remained on flumezapine and his CPK level reduced to normal within a week (this patient was removed from the study after being diagnose