Secondary patent protects Sanofi-Aventis' blockbuster drug leflunomide

1 September 2011

Once their compound patents have expired many blockbuster drugs remain protected by secondary 'method of treatment' patents. The validity and enforceability of these secondary patents can be less certain in many jurisdictions.

In Sanofi-Aventis Australia Pty Ltd v Apotex Pty Ltd (No.3) [2011] FCA 346, the Federal Court found that the secondary patent protecting Sanofi-Aventis' blockbuster leflunomide drug was valid and infringed – a surprising outcome, at least initially, based on the facts of the case.

Significance of this case

  • This case illustrates the value of a carefully drafted secondary patent. Sanofi-Aventis' method patent directed to the treatment of psoriasis has effectively given it a further 10 year monopoly on the use of leflunomide to treat PsA (a related but different condition to psoriasis) following expiry of the compound patent.

  • Much of this case turned on rather unique facts. While PsA and psoriasis are today understood to be related diseases, in 1993 they were considered to be distinct. So in the absence of any explicit prior art teaching that leflunomide could treat psoriasis or PsA (in which case it would also be treating psoriasis), the Patent was novel. The fact that, during the term of the Patent, it was discovered that the use of leflunomide to treat PsA and psoriasis also treated psoriasis was relevant to infringement, but not novelty or inventive step.

  • The case arguably shows a willingness of the Federal Court to uphold a secondary patent and apply a rigorous enablement test to prior art. Broad general disclosures may not be enough to invalidate a later Patent – the prior art must actually teach a particular feature to a skilled person (or instil a reasonable expectation of success in trying it) in order for that feature to be anticipated.


Sanofi-Aventis markets in Australia a pharmaceutical product called ARAVA®, which contains as its active ingredient leflunomide. ARAVA® is approved for the treatment of rheumatoid arthritis (RA) and psoriatic arthritis (PsA), but, importantly, is not approved for 'the treatment of psoriasis that is not associated with manifestations of arthritic disease'.

Sanofi-Aventis owns two relevant patents

  • Australian patent no. 529341 covering the pharmaceutical compound leflunomide (341 Patent) which expired in 2004
  • Australian patent no. 670491 covering a method of preventing or treating psoriasis using leflunomide (Patent), which expires in 2014 and was the subject of this proceeding.

In July 2008, Apotex obtained regulatory approval for its generic version of leflunomide for the treatment of the same conditions as ARAVA®. In response, Sanofi-Aventis commenced proceedings in the Federal Court alleging that Apotex's product infringed the Patent. Apotex counterclaimed alleging that the Patent was invalid for lack of novelty, inventive step, fair basis, utility, sufficiency and for not being a 'manner of manufacture'.

The Court made some very interesting findings on novelty and inventive step which are discussed below.


Before assessing validity, the Court construed claim 4 of the Patent (the principal claim at issue) at length, which reads:

'A method of preventing or treating… psoriasis, which comprises administering to a recipient an effective amount of a pharmaceutical composition containing as an active ingredient a compound of formula I [including leflunomide] …'

Both parties presented a number of competing constructions. For example, for the claim to be infringed, did the doctor administering the drug need to (i) intend subjectively, or (ii) intend objectively, to treat psoriasis? Alternatively, is intention irrelevant and the claim infringed where (iii) in the course of treating PsA, psoriasis is incidentally treated, or (iv) psoriasis is as a matter of fact treated? Apotex favoured construction (ii) and Sanofi-Aventis favoured construction (iv).

The Court decided in favour of Sanofi-Aventis, by applying an orthodox claim construction and finding that because the claim makes no reference to subjective or objective intentions, the purpose of the administration is irrelevant.

The importance of the evidence

The evidence on psoriasis, RA and PsA was determinative in this case. It explains what, at first blush, appears to be an odd result — that Apotex's product (approved for treating RA and PsA, but not psoriasis alone) infringed a Patent directed to the treatment of psoriasis.

The finding was based on evidence that almost all PsA sufferers have, or are likely to develop, psoriasis.  The current accepted view is that PsA and psoriasis are related autoimmune diseases – both caused by an abnormal inflammatory response that, in the case of psoriasis, leads to an excessive production of skin cells.  Administering Apotex's product to treat PsA will also treat or prevent psoriasis.

This finding in turn raises the question: if treating PsA also treats psoriasis, how is the Patent novel over the earlier use of leflunomide to treat PsA?

This potential inconsistency is explainable by the evidence on the state of developments in 1993, the priority date of the Patent.  Jagot J accepted that, at that time, PsA and psoriasis were considered distinct, unrelated disorders.

Novelty and inventive step

Apotex alleged that three documents anticipated the Patent, including the earlier 341 Patent owned by Sanofi-Aventis. Apotex's main argument was that the Patent lacked novelty under a 'reverse infringement test'. If the Patent is infringed by administering leflunomide (for any purpose) which in fact treats psoriasis; then the use of leflunomide (for any purpose) prior to the Patent must anticipate the Patent.

The Court disagreed and said that the 'reverse infringement test' is just 'one way of expressing the broader principle that, if a finding of anticipation is to be made, all that is "essential to the invention must be read out of the prior publication" (Hill v Evans)'. Further, this test cannot be applied literally to paper anticipations as the hypothetical infringement depends on someone actually doing or making what the publication describes or suggests.

In light of the expert evidence, none of the cited prior art disclosed the use of leflunomide to treat PsA (the conflicting evidence of Professor Brooks was dismissed on the basis that he was a not a typical non-inventive worker, see below). While the prior art arguably taught the use of leflunomide to treat RA, the occurrence of psoriasis in patients with RA (as opposed to PsA) is a mere coincidence (no higher than in the general population, at around 2% of people). Therefore, in nearly all cases (98%), treating RA sufferers with leflunomide will not 'treat or prevent psoriasis' and the Patent is therefore novel under the 'reverse infringement test'.

Some of the Court's comments on Sanofi-Aventis' earlier 341 Patent are also worth noting. The Court observed that, taking the evidence as a whole, it did not support construing the 341 Patent as 'in fact teaching the administration of leflunomide to treat any disease in humans' [emphasis added]. Even if it did teach a method of treating rheumatic complaints (around 150 diseases) the disclosure is 'too broad and lacks sufficient clarity to deprive the Patent of novelty'. One wonders if  the parties' submissions would have been different if the 341 Patent was still in force?

The Court's approach to inventive step was similar to its novelty assessment. Lack of knowledge about the cause of PsA would not have indicated to the skilled person that leflunomide was an effective treatment for PsA. Nor would it lead them to try the claimed method with any expectation of success. The Patent was therefore inventive.


Based on Court's construction of claim 4, the finding of infringement was a relatively straight-forward matter. Apotex's product, when used to treat or prevent PsA, would in almost all cases also treat or prevent psoriasis. On that basis it infringed the Patent. Apotex was liable under section 117 of the Patents Act 1990 (Cth) because they supplied the product to doctors, pharmacists and patients and either had reason to believe the product would be used in an infringing manner (s117(2)(b)); or provided instructions directing use in an infringing manner (s117(2)(c)).

It is worth noting that s117(2)(b) only applies in situations where 'the product is not a staple commercial product'. Jagot J disagreed with Apotex's argument that it did not infringe under s117(2)(b) because the active ingredient, leflunomide, is a staple commercial product with a range of potential uses.  Following established case law, she noted that a 'staple commercial product' means a product that 'not only can be used in a variety of ways but is also traded for use in various ways'. In this case 'the product' was Apotex's pharmaceutical product (not just the active ingredient). The product information document for Apotex's product discloses only its use as an oral medicament for the treatment of RA and PsA. It is therefore not a staple product supplied commercially for a variety of uses; and, accordingly, Apotex infringed the Patent under s117(2)(b).

Jagot J also noted that Sanofi-Aventis did not suggest that the supply of leflunomide to treat RA alone would infringe the Patent. This is potentially significant because in June 2010 Apotex filed a new regulatory approval for its product for the treatment of RA only. Arguably the supply of Apotex's product under this new approval would not infringe the Patent — for the same reasons as the previous disclosure of leflunomide to treat RA did not invalidate the Patent.

Misleading and deceptive conduct

Sanofi-Aventis also argued that Apotex engaged in misleading and deceptive conduct under the Trade Practices Act 1974 (Cth) (now embodied in the Australian Consumer Law) by representing to doctors, pharmacists, patients and the Therapeutic Goods Administration (TGA) that its product could be used to treat PsA but failing warn them of actual or potential Patent infringement. 

With respect to the TGA, the representation arose out of the s 26B certificate required under the Therapeutic Goods Act.  It required Apotex to certify a good faith belief, on reasonable grounds, that its product would not infringe a valid claim of a patent.  Jagot J held that the certificate did not breach the Trade Practices Act as the representation related to Apotex's state of mind.  Sanofi-Aventis had not proved that the statement was inaccurate.

In contrast, material supplied to doctors, pharmacists and patients was silent on patent infringements.  Jagot J held that this conveyed a representation that doctors were free to prescribe, pharmacists were free to dispense and patients were free to use the product. This was false, and likely to be misleading and deceptive, because they were not free to do so — such conduct would infringe the Patent.

Choosing the right expert

Jagot J also continued a recent Federal Court trend of closely considering the suitability of experts. She noted that Professor Brooks was 'ahead of the pack' and did not reflect the typical non-inventive worker. This follows recent comments by Kenny J on experts, in particular that the practical 'hands-on' skills of an industrial expert were favoured over those of an academic.

This highlights the importance of correctly identifying and retaining an expert as soon as possible in the context of patent litigation. Not only must an expert have the right technical skills (often a very limited pool of candidates), but they must also agree with your case and have a suitably robust personality to withstand the rigours of cross-examination.

Author(s) John Fairbairn, Dennis Schubauer