Federal Court of Australia

Novartis AG v Pharmacor Pty Limited [2022] FCAFC 58

ORDERS

THE COURT ORDERS THAT:

1.    The application for leave to appeal be refused.

2.    The applicants pay the respondent’s costs of the application for leave to appeal.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

YATES AND MOSHINSKY JJ:

1    This is an application for leave to appeal from a decision of the primary judge on a matter of practice and procedure. The application for leave to appeal was heard together with the appeal (if leave were granted). At the conclusion of the hearing, after a short adjournment, the Full Court announced its decision and made orders that the application for leave to appeal be refused, and that the applicants pay the respondent’s costs of the application for leave to appeal. The Court indicated that reasons would be published in due course. The following are our reasons for joining in the orders.

2    In the proceeding at first instance, the respondent (Pharmacor) filed a cross-claim seeking the revocation of the two patents in suit (the Patents). Broadly, the claims in the Patents are addressed to, or include, a method of treating relapsing-remitting multiple sclerosis in which a daily dose of 0.5 mg of fingolimod is administered orally. In November 2021, Pharmacor filed its expert evidence in support of the cross-claim. This included an affidavit of one neurologist, Associate Professor John King. In February and March 2022, the applicants (Novartis) filed their expert evidence in response. This included affidavits of four neurologists: Professor Michael Barnett; Professor Gavin Giovannoni; Professor Fred Lublin; and Professor Pamela McCombe. Pharmacor objected to Novartis calling four neurologists and applied to the primary judge for an order under r 5.04, item 16 of the Federal Court Rules 2011. Rule 5.04(1) provides that, at any hearing, the Court “may make directions for the management, conduct and hearing of a proceeding”. Rule 5.04(3) provides that, without limiting subrule (1), the Court may make a direction mentioned in the table. Item 16 of the table is a direction in relation to “[t]he number of expert witnesses to be called”.

3    The matter was listed for a case management hearing before the primary judge, who is the docket judge for the matter. The parties filed brief outlines of submissions in advance of the hearing. The case management hearing took place on 18 March 2022. The transcript of that hearing was before the Full Court. On the premise that the applicants’ main neurological expert witness was Professor Barnett, the primary judge ruled, in summary, that the evidence of Professor Giovannoni and Professor McCombe should be excluded, and that the evidence of Professor Lublin other than factual evidence (and associated background and contextual evidence) should be excluded. In relation to Professor Lublin’s evidence, Novartis was given time to identify the parts of his evidence to be excluded on the basis of the view expressed by the primary judge. During the hearing, Novartis requested reasons for the primary judge’s ruling, and the primary judge said that the reasons were in the transcript (T41). On 22 March 2022, the primary judge made orders including the following:

1.    The Applicants are not permitted to rely at trial on the affidavit of Professor Giovannoni affirmed on 4 March 2022 (Giovannoni Affidavit), the affidavit of Professor McCombe affirmed on 24 February 2022 (McCombe Affidavit), or the affidavit of Professor Lublin affirmed on 26 February 2022 (Lublin Affidavit), other than those parts of the Lublin Affidavit indicated by the Court at the case management conference on 18 March 2022 to be outside its rulings (Exempt Lublin Evidence).

4    On 22 March 2022, Novartis provided its proposed redactions of the Lublin affidavit to Pharmacor, and Pharmacor indicated that it did not disagree with the proposed redactions. The redacted version of the Lublin affidavit was then provided to the primary judge. The material before the Full Court includes a copy of the Lublin affidavit marked up to show the redacted portions. This shows that the following paragraphs were not excluded: paragraphs 6 to 21 (dealing with Professor Lublin’s qualification and experience); paragraphs 56 to 69 (dealing with the Phase III fingolimod multiple sclerosis (MS) trials); and paragraphs 76 to 104 (relating to various matters, including doctors’ scepticism about including the 0.5 mg dose in the Phase III trials, the trial results, and the FDA’s involvement in relation to fingolimod research). The excluded parts of the affidavit are: substantially all of paragraphs 22 to 42 (dealing with clinical trials in general and MS clinical trials in particular); substantially all of paragraphs 43 to 55 (dealing with the fingolimod transplant trials and the Phase II fingolimod MS trial); and paragraphs 70 to 75 (dealing with certain Novartis press releases relied on by Pharmacor in its cross-claim, and other documents).

5    Novartis has applied for leave to appeal from paragraphs 1, 22, 23 and 24 of the orders made by the primary judge on 22 March 2022. Paragraphs 22, 23 and 24 deal with costs. In support of its application for leave to appeal, Novartis relies on an affidavit of John Collins, a partner of Clayton Utz, the solicitors acting for Novartis, dated 24 March 2022. In opposing the application, Pharmacor relies on an affidavit of Benjamin Miller, a partner of Maddocks Lawyers, the solicitors acting for Pharmacor, dated 29 March 2022.

6    Novartis has provided a draft notice of appeal. This contains three grounds, which are in substance as follows (omitting particulars):

(a)    the primary judge erred in finding that Novartis is not permitted to rely at trial on the affidavit of Professor Giovannoni and parts of the affidavit of Professor Lublin, on the ground that that evidence involved “overlapping witnesses” (T18-19) or, further or alternatively, “overlapping” or “doubling up” with the affidavit of Professor Barnett (T3, T41);

(b)    further or alternatively, to the extent (if at all) that it was open to the primary judge to find that any part of Professor Giovannoni’s affidavit and the excluded parts of Professor Lublin’s affidavit were “substantially duplicative” of other evidence filed by Novartis, or there was otherwise a discretionary basis available on which to make paragraph 1 of the primary judge’s orders in respect of that evidence, then his Honour erred in finding (if his Honour did so find) that discretionary factors weighed in favour of making paragraph 1; and

(c)    further or alternatively, the primary judge erred in finding that Novartis not be permitted to rely at trial on the excluded parts of Professor Lublin’s affidavit.

7    The principles applicable to an application for leave to appeal from an interlocutory decision are well established. In general, the tests to be applied are: (a) whether, in all the circumstances, the decision is attended with sufficient doubt to warrant its being reconsidered by the Full Court; and (b) whether substantial injustice would result if leave were refused, supposing the decision to be wrong: Decor Corporation Pty Ltd v Dart Industries Inc (1991) 33 FCR 397 at 398-400 per Sheppard, Burchett and Heerey JJ.

8    Novartis submits that the evidence the subject of paragraph 1 of the orders made on 22 March 2022 is not “substantially duplicative” of other evidence filed by Novartis in the proceeding. Novartis submits that the primary judge approached the exercise of power on the basis that differences in witnesses’ background and expertise, and the actual content of their evidence, are not relevant considerations.

9    Novartis submits that the practical effect of the primary judge’s order is to force Novartis to limit its defence of the cross-claim to one that only answers the case as framed by Pharmacor’s evidence and witnesses, and to prevent Novartis from advancing any alternative positive case based on different material, supported by evidence of witnesses with materially different knowledge and backgrounds.

10    Novartis no longer seeks to rely on the affidavit of Professor McCombe that was excluded by the primary judge. It is therefore unnecessary to consider that affidavit.

11    Novartis makes the following submissions about the evidence of the other neurologists:

(a)    Professor Barnett is an Australian neurologist who had some involvement with a Phase III fingolimod clinical trial, but not with its Phase II trials or otherwise with fingolimod’s clinical development before the priority date (which is in June 2006). He addresses a hypothetical “task” similar to that undertaken by Associate Professor King relating to the development of a new treatment for MS.

(b)    Professor Giovannoni is a Professor of Neurology in the United Kingdom, who also holds qualifications in pharmacy. He has extensive experience in clinical trials for MS therapies, including the use of animal models in pre-clinical studies, and participated in advisory board discussions with Novartis regarding fingolimod. Unlike Professor Barnett and Associate Professor King, he discusses a body of work published before the priority date reporting on studies of fingolimod in transplant patients (the Transplant Studies). His hypothetical “task” evidence responds to a narrower task than that addressed by Professor Barnett and Associate Professor King, because the task contains an assumption that fingolimod had been identified as a candidate by the research group. Professor Giovannoni had express regard to what he would have understood and taken from the Transplant Studies in the context of that task.

(c)    Professor Lublin is a Professor of Neurology in the United States. He has knowledge of US clinical trials, FDA processes, and was on advisory boards to Novartis before and after the priority date. He explains that he was likely provided with the Transplant Studies in one of his roles in the Phase II and III clinical advisory boards. He does not give evidence in relation to a hypothetical “task”.

12    Novartis submits that the evidence described above (in particular that of Professor Giovannoni) is necessary to enable it to advance a different theory as to the notional skilled team (namely, one that includes a neurologist with the background and level of experience of Professor Giovannoni) and the hypothetical task (namely, the narrower question addressed by Professor Giovannoni).

13    It is convenient to deal first with the factual aspects of Professor Lublin’s evidence. In its submissions to the primary judge, Novartis submitted that, to a large extent, Professor Lublin is a witness of fact. This was recognised by the primary judge in the order that he made, which did not exclude significant portions of Professor Lublin’s evidence, on the basis that these portions were of a factual nature. In its oral submissions before the primary judge, Novartis indicated that the heart of Professor Lublin’s factual evidence was paragraphs 56 to 69, which relate to the Phase III fingolimod MS trials (T31-32). The primary judge did not exclude this evidence. Further, the primary judge gave Novartis the opportunity to identify earlier parts of Professor Lublin’s affidavit that should not be excluded, namely parts that dealt with Professor Lublin’s qualifications and experience, and parts that gave “content” to his later evidence, including “if he needs to provide a personal explanation of some aspects of the clinical trials that inform or explain” the evidence at paragraphs 56 to 69 (T36, T49). As indicated above, following the case management hearing on 18 March 2022, Novartis prepared a redacted copy of Professor Lublin’s affidavit that it considered reflected the view expressed by the primary judge at the hearing. Pharmacor did not disagree with the proposed redactions, and the redacted copy was provided to the primary judge. However, it appears to us that Novartis may have excised more than was necessary to give effect to the primary judge’s view at the case management hearing. The redacted copy excises substantially all of paragraphs 22 to 55 of the affidavit. Arguably, at least, a good deal of this material is factual material of a background nature that is necessary to inform or explain the evidence at paragraphs 56 to 69. In our view, it would be open to Novartis to apply to the primary judge to have a substantial part of paragraphs 22 to 55 treated as necessary background and contextual material, and therefore not to be excluded.

14    We now turn to the evidence of Professor Giovannoni and the other parts of Professor Lublin’s evidence.

15    Insofar as Novartis submits that this evidence is not substantially duplicative of Professor Barnett’s evidence, we do not accept that submission. Each of the three experts has the title Professor of Neurology and has expertise in MS. Each: (a) discusses what was generally known regarding MS and clinical trials at the priority date; and (b) considers the Novartis media release dated 6 April 2006, the Medical News Today article dated 10 April 2006 and the Neurology Today article dated 16 May 2006 (the Articles), which are relevant to Pharmacor’s obviousness/novelty case. Each of the experts expresses substantially the same opinions to the following effect:

(a)    that the Articles indicate that Novartis hoped the benefits of the 1.25 mg dose (rather than the 0.5 mg dose) would be confirmed in the Phase III clinical trial;

(b)    that the 0.5 mg dose would have been expected to have little or no clinical efficacy; and

(c)    that it would have been expected that the 0.5 mg dose was included in the Phase III clinical trial in order to better understand the dose-response relationship.

16    Both Professor Barnett and Professor Giovannoni were asked to respond to the whole of Associate Professor King’s affidavit.

17    Both Professor Barnett and Professor Giovannoni were instructed to address a hypothetical “task”: see Professor Barnett’s affidavit, paragraph 66 and Professor Giovannoni’s affidavit, paragraph 60. While there is a difference between the tasks (in that Professor Giovannoni was instructed to assume that fingolimod had been identified as a drug that the notional research group was interested in), this difference does not appear to have practical significance as Professor Barnett identified fingolimod as a promising candidate: see paragraphs 80 to 82 of Professor Barnett’s affidavit.

18    Insofar as Novartis submits that the practical effect of the exclusion of this evidence is to limit Novartis to responding to the case framed by Pharmacor, as distinct from advancing a different positive case, we are not satisfied that this is the practical effect of the order. A large part of this submission turned on the Transplant Studies. These studies are already part of Novartis’s expert evidence through the affidavit of Professor Carl Kirkpatrick, a Professor of Pharmacy Practice, dated 9 March 2022. To the extent that Novartis seeks to have a neurologist express opinions about the Transplant Studies, Novartis accepted before the primary judge that Professor Barnett could express opinions about them (T40-41). The primary judge said that it was open to Novartis to apply to adduce further evidence from Professor Barnett about this. No such application has yet been made, but it remains open to Novartis to make such an application. We note that Novartis submits that Professor Giovannoni was aware of the Transplant Studies at the priority date, while Professor Barnett was not. However, the significance of this may be overstated. Professor Giovannoni states at paragraph 61 of his affidavit that he became aware of the transplant trials through “confidential advisory board discussions with Novartis staff”. In paragraph 63 of his affidavit, Professor Giovannoni puts his “actual personal knowledge” of the trials to one side in addressing the hypothetical “task”. Thus, the evidence given by Professor Giovannoni is not premised on the notional research group having his personal knowledge of the trials. Further, the factual evidence relating to the transplant trials in Professor Lublin’s affidavit could be the subject of an application to the primary judge as discussed at [13] above.

19    Insofar as Novartis submits that there are differences in the background and experience of the three experts, this may be accepted, but it does not suggest error by the primary judge. The differences in background and experience of Professors Barnett, Giovannoni and Lublin are not sufficient to justify the calling of multiple expert witnesses of the same discipline in the circumstances of this case. As the primary judge noted, there are invariably differences between experts’ backgrounds. Further, as the primary judge recognised, the notional person is not an avatar for experts whose testimony is accepted; rather, it is a tool of analysis guiding the Court in determining obviousness by reference to expert and other evidence: AstraZeneca AB v Apotex Pty Ltd (2015) 257 CLR 356 at [23] per French CJ.

20    Further, in our view, the approach taken by the primary judge was consistent with modern case management in patent litigation and the purpose expressed in Pt VB of the Federal Court of Australia Act 1976 (Cth): see Australian Securities and Investments Commission v Australia and New Zealand Banking Group Ltd (2019) 139 ACSR 52 at [12]-[15] per Allsop CJ.

21    For these reasons, we are not satisfied that the order made by the primary judge is attended with sufficient doubt to warrant its being reconsidered by a Full Court, or that substantial injustice would result if leave were refused, supposing the decision to be wrong.

22    It was for the above reasons that we joined in the order refusing leave to appeal at the conclusion of the hearing.

Associate:

Dated:    8 April 2022

REASONS FOR JUDGMENT

BEACH J:

23    I joined in the order refusing leave to appeal substantially for the reasons given by Yates and Moshinsky JJ. But there are two matters that it is appropriate to elaborate on. The first matter concerns the general question of over-lapping expert evidence to be called by a party. The second and more specific matter concerns a modified approach that is open for the primary judge to take concerning the evidence of Professor Frederick Lublin, a professor of neurology and the Director of the Centre for Multiple Sclerosis at the School of Medicine, Mount Sinai Hospital, New York.

24    As to the question of over-lapping expert evidence in the context of patent litigation, four points should be made.

25    First, the default position is that a party should not adduce expert evidence from more than one expert in any single discipline, absent telegraphing its intention to do so to the opposite party and the Court at the earliest opportunity.

26    Now it may be accepted that the concept of a discipline can be broad. Take the field of chemistry. You have well-recognised divisions of physical, inorganic or organic chemistry. And even within those divisions you have sub-divisions such as synthetic, analytical or computational. Take the field of medicine. You can have non-exhaustive possibilities such as the academic, the clinical and the experimental. Clearly, experts in the same general field but falling within separate recognised divisions or sub-divisions in that field should be regarded as experts in separate disciplines to which the default position does not apply. A modicum of common sense informed by the context of the technical matters in issue between the parties needs to be exercised. But if there is a doubt, this should be raised with the other party and the Court at the earliest opportunity.

27    Second, if the default position is to be departed from, it should be raised with the other party and the Court and justified before the expert evidence is filed and served. Justifications that may be convincing may require establishing, inter-alia, that:

(a)    different questions have been posed to different experts, although this will then require an explanation as to why that was necessary;

(b)    different experts have substantially different knowledge and experience before the priority date, although this may then entail an inquiry as to why both are necessary, particularly the one with lesser knowledge or experience before the priority date; but it may be that such knowledge and experience does not only divide on a time spectrum, but is more a question of saying that each may be relevantly knowledgeable and experienced as at the priority date but have significantly different knowledge bases and experience as at the priority date which difference then substantially bears upon the opinions expressed; or

(c)    one of the experts is giving direct factual evidence perhaps with a modicum of hearsay, and not only as a foundation for an opinion admissible under s 79 of the Evidence Act 1995 (Cth); in other words, the expert is in substance or also a witness of direct fact.

28    But to establish any such difference may only be a necessary but not sufficient condition for an acceptable justification. One also needs to show that as a result of any such difference the expert evidence is not substantively duplicative, although this may follow in most cases from the difference.

29    Third and generally, the Intellectual Property Practice Note (IP-1) makes it plain that proper case management requires ensuring the effective use of expert evidence (cls 5.2 and 10.2). Unnecessary and potentially duplicative evidence is to be eschewed (cls 6.1 and 6.2). The Expert Evidence Practice Note (GPN-EXPT) is consistent with these themes (cl 6.1(a)). And the Court has a smorgasbord of powers to achieve these objectives.

30    Fourth, even if the parties seek to luxuriate in an agreed position of each relying upon multiple over-lapping experts in the same discipline, the docket judge is likely to have little tolerance. Hence the importance of raising the matter with the Court at the earliest opportunity irrespective of the parties’ agreed position. Emmett J said in Novartis AG v FH Faulding & Co Ltd [2004] FCAFC 254 at [7] that “an imaginative approach to case management in complex technical cases such as patent cases should be encouraged”. I agree, but we are not even close here to considering or imposing innovative solutions. Little imagination needs to be used to appreciate that substantially duplicative expert evidence will not be indulged with insouciance.

31    Let me turn then to the evidence of Professor Lublin and make the following points.

32    Now because his Honour gave no separate reasons for his ruling ex tempore or otherwise, it is unclear as to the precise power that he was exercising. Clearly though, he was not exercising power under rule 5.04 of the Federal Court Rules 2011 (Cth). Moreover, he was not engaged in a voir dire or any other exercise under s 192A of the Evidence Act given that his ruling concerning Professor Lublin was not about relevance or more generally admissibility; in other words it had been assumed that the excluded evidence of Professor Lublin was admissible. Further, Professor Lublin’s evidence had not been filed and served in breach of any court order. Now it is suggested that his Honour was exercising more general case management powers. But I need not linger on the power question because I refused leave to appeal concerning the ruling on Professor Lublin’s evidence on the second limb of the leave question; no substantial injustice had been shown.

33    No irreparable injustice has been shown because on any appeal (if necessary) from his Honour’s ultimate judgment on the cross-claim alleging invalidity of the asserted claims of the two patents in suit, Novartis will have preserved to it the challenge it now makes to the exclusion of parts of Professor Lublin’s written evidence in chief.

34    Further, and even more persuasively, as his Honour’s ruling was interlocutory it can be revisited by him. Now should it be? That is a matter of course for his Honour. But let me make the following points.

35    First, Professor Lublin was a witness of fact (non–s 79 opinion evidence) on Phases II and III clinical trials concerning fingolimod, as his Honour appreciated. He also gave expert opinion evidence on Phase I trials. Fingolimod is an immunomodulatory agent with its mode of action being lymphocyte suppression and so thought to have advantageous properties to treat multiple sclerosis and in particular relapsing-remitting MS, which is an immune-mediated demyelinating disease involving the selective destruction of the myelin sheath around nerves. The Phase I trials concerned kidney transplant patients rather than MS patients. The Phases II and III trials concerned relapsing-remitting MS patients. Now his Honour excluded the evidence concerning the Phases I and II trials, but the Phase II trial evidence was non–s 79 opinion evidence, albeit lightly sprinkled with some hearsay. So the vice of duplicating s 79 opinion evidence did not arise. Moreover, Professor Lublin’s opinion evidence concerning the Phase I trials had not been duplicated elsewhere in the evidence that his Honour did not exclude concerning neurology.

36    Second, in order to appreciate how the Phase III clinical trials that Professor Lublin was addressing evolved, the background dealing with Phases I and II was relevant and significant evidence. I also note that the transplant articles which also relate to the Phase I trial in part are also going in through Professor Carl Kirkpatrick, a professor of pharmacology.

37    Third, Professor Michael Barnett, another neurologist, had no involvement in the US trials before the priority date that Professor Lublin was specifically addressing. He had little to say on the Phase I trials that Professor Lublin discussed. Moreover, putting to one side the different trials point, Professor Barnett had no involvement in any form of Phase II trials for fingolimod in relation to MS. Further, his work in Phase III trials concerning fingolimod seems to have substantially taken place after the priority date, and in any event were different to those that Professor Lublin’s evidence focused on; his Honour in part recognised this by not excluding Professor Lublin’s evidence on the Phase III trials at least.

38    Fourth, Professor Lublin’s comprehensive description of clinical trials and MS trials generally is not substantially duplicative of Professor Barnett’s evidence who deals with these general topics in snippets. Now a small part of it may have been contentious s 79 opinion evidence, but there was no substantial duplication.

39    Fifth, on the assumption that the evidence of Professor Gavin Giovannoni, another neurologist, is now excluded and given the previous four points, it is open to his Honour to revisit his ruling, particularly given that Professor Lublin’s evidence concerning the Phase I trials, the transplant studies and the Phase II trials is not duplicated by any neurologist’s evidence not excluded by his Honour. I should say that it is no sufficient answer that his Honour might entertain Professor Barnett supplementing his evidence on the transplant studies. That possibility does not address what Professor Lublin said on the Phases I and II trials. In any event, Novartis may wish to do so through Professor Lublin. I should say that on my perusal of the transcript of argument before his Honour on 18 March 2022, both parties were not as clear as they should have been concerning the different nature and quality of Professor Lublin’s evidence and any question concerning potential overlap, particularly if you assumed that Professor Giovannoni’s evidence was to be excluded.

40    In summary, his Honour can revisit his ruling if he so chooses and particularly given that Professor Lublin is to give evidence anyway. So, I was more against Novartis concerning the second limb rather than the first limb of the leave to appeal test concerning Professor Lublin’s evidence. And concerning the exclusion of Professor Giovannoni’s evidence, Novartis failed on both limbs.

41    For these reasons I refused leave to appeal.

Associate:

Dated:    8 April 2022