A once steady stream of preliminary injunction (PI) applications in Australian pharmaceutical patent litigation came to a halt in around 2021. It took four years for patentees to once again 'enter the fray', with two applications in the Federal Court, and an appeal to the Full Court from one of those decisions, heard in the second half of 2025.
First, in September 2025, Justice Rofe dismissed a PI application brought by Regeneron and Bayer seeking to prevent Sandoz from launching its aflibercept biosimilar AFQLIR®, in Regeneron Pharmaceuticals, Inc. v Sandoz Pty Ltd [2025] FCA 1067. MinterEllison acted for Sandoz in that proceeding, and in Regeneron/Bayer's subsequent appeal to the Full Court.
By contrast, in December 2025, Justice Burley granted a PI sought by Janssen in Janssen Pharmaceutica NV v Juno Pharmaceuticals Pty Ltd [2025] FCA 1538, restraining Juno from launching its generic paliperidone palmitate products.
The decision to restrain Juno marks a shift in the Court's recent approach, insofar as it 'bucks the trend' of the Federal Court declining to grant the majority of PI applications made over the last decade.
This article compares and contrasts these two decisions, and distils the key takeaways for pharmaceutical companies in Australia.
1. Injunction refused: The decision to allow Sandoz to launch
Infringement
Regeneron (the patentee) and Bayer (the exclusive licensee) alleged infringement of a dosing regimen patent relating to the administration of aflibercept (EYLEA®), a top 5 PBS-listed drug used in the treatment of eye disorders such as diabetic macular oedema. The PI application was high stakes, as refusing the injunction would open the door to a $500M annual market in Australia for Sandoz and subsequent biosimilar entrants.
Regeneron and Bayer argued that they had a strong prima facie case on infringement, which was disputed by Sandoz. The parties' position on infringement revolved around whether Sandoz's Product Information for AFQLIR instructed ophthalmologists to administer the drug in a way that would infringe the patent, and whether Sandoz had reason to believe that ophthalmologists would direct the use of the drug in such a manner.
Justice Rofe held that the statements in Sandoz's Product Information did not fall within the claims of the patent. Her Honour found that the phrase "one injection per month" (the wording used in Sandoz's Product Information) does not amount to an instruction to administer aflibercept "2 to 4 weeks" after the preceding dose (as claimed in the patent), and similarly, "every two months" does not equate to "8 weeks". That is, her Honour construed the claims strictly, noting that the patentee had chosen to specify weekly intervals without qualification.
On that basis, the Court held that Sandoz would not have reason to believe that ophthalmologists would infringe the patent. This was bolstered by Sandoz's unanswered expert evidence that most ophthalmologists use a different treatment protocol, as a matter of practice, to the regimen set out in the Product Information.
Justice Rofe was unpersuaded by Sandoz's cross-claim for invalidity based on lack of novelty, inventive step and support at the preliminary stage.
Balance of convenience
Justice Rofe found that the balance of convenience did not favour granting a PI, given the weak prima facie case on infringement. Her Honour recognised that Sandoz's launch would trigger a 25% reduction in the price paid by the Commonwealth under the Pharmaceutical Benefits Scheme (PBS) which, together with further reductions and discounting thereafter, represented significant financial harm to Regeneron/Bayer.
However, a crucial factor was that Regeneron/Bayer were already shifting the existing aflibercept market by encouraging ophthalmologists to switch patients from 2mg (the dosage in dispute) to 8mg (for which there is presently no contemplated biosimilar competition).
Her Honour also recognised the relevance of Sandoz's first mover advantage, bolstered by the idiosyncrasies in the biosimilar market for aflibercept. Sandoz led evidence that ophthalmologists are typically conservative and reluctant to switch drugs, meaning that significant time would be required to educate and familiarise practitioners to convince them to switch to a first biosimilar.
Finally, Justice Rofe accepted that broad injunctive relief would be disproportionate in the circumstances, given the weak infringement case. Relevantly, Sandoz had highlighted the large proportion of non-infringing use when ophthalmologists (a) used a different treatment protocol, or (b) administered aflibercept to treat other eye disorders not covered by the patent.
Regeneron and Bayer appealed her Honour's PI decision to the Full Court, which was heard on 29 October 2025, but the dispute settled prior to the Full Court delivering its judgment in the appeal.
2. Injunction granted: The decision to restrain Juno Pharmaceuticals
Infringement
In Juno, Janssen alleged infringement of a dosing regimen patent, this time in relation to paliperidone palmitate products used in the treatment of schizophrenia. Janssen's allegations centred around instructions in Juno's Product Information for VALINO MONTHLY and PALJUNA MONTHLY. In this case, however, Justice Burley held that the Product Information did contain instructions to infringe the Janssen patent.
Notably, the PALJUNA Product Information instructed that patients who have never taken oral paliperidone should establish tolerability by initiating doses of Janssen's (i.e. the patentee's) INVEGA product first, which should be administered in accordance with the INVEGA Product Information (whose instructions fall within the claims of the patent).
On that basis, Juno argued that it was not 'authorising' infringing conduct, because the treatment regimen would initially proceed by the instructions given in Janssen's Product Information. However, Justice Burley considered that the PALJUNA Product Information would still result in Juno being responsible for facilitating, procuring or persuading infringement of the patent, notwithstanding that Juno does not itself supply the loading doses.
Further, the parties disputed the degree of adherence by health care professionals, and patients, to treatment regimes recommended in Product Information. However, Justice Burley found for the purpose of the PI hearing that:
- a "substantial proportion" of those who prescribe and administer INVEGA do so by closely following the instructions;
- a "small" percentage of patients who fail to take their medicine as prescribed would fall outside of the claims; and
- a "small but non-trivial" cohort of patients would not infringe, as the Product Information allows for dosing in ranges outside of the claims for patients with specific characteristics, such as mild renal impairment.
On that basis, Justice Burley considered there was a strong prima facie case on infringement, which was not displaced by Juno's arguments that the patent is invalid for want of inventive step or false suggestion.
Balance of convenience
Justice Burley considered that the balance of convenience weighed in favour of granting a PI, noting that the apparent strength of the parties' substantive prima facie cases is often determinative.
Although the typical immediate 25% PBS price reduction did not apply here (a 26.1% reduction had previously been applied to paliperidone palmitate), his Honour recognised that generic entry would lead to irreversible price discounting (even if no mandatory price disclosure reductions would be applied before patent expiry on 17 December 2028).
Further, even if additional generic entrants did not launch before patent expiry (which Justice Burley considered possible rather than probable) and without considering the claimed negative impact on sales of Janssen's other INVEGA products with different dosages (which was said to be "overstated"), his Honour recognised that Janssen would suffer a "very substantial" loss of revenue.
Justice Burley also acknowledged that Juno was seeking to obtain the "first mover advantage" but noted that the extent and duration of such a benefit was speculative.
Importantly, Justice Burley considered that the difficulty in calculating damages to Juno on the undertaking as to damages if it was wrongly restrained, would be greater than the task of calculating Janssen's damages for patent infringement if the PI was not granted.
Finally, Justice Burley noted the patent's imminent expiry date, being 17 December 2028. His Honour expected that a substantive trial could take place in the second quarter of 2027, such that the proceeding and any appeal should be resolved ahead of patent expiry.
3. Key learnings
Lack of inventive step continues to be a difficult ground to successfully defeat a PI
In Sandoz, inventive step arose in the context of Sandoz's argument that the prior art inherently disclosed the amino acid sequence of aflibercept and not, for example, in reliance on expert evidence on a 'hypothetical task'. However, Justice Rofe's rejection of the novelty argument meant that inventive step also failed. Her Honour further noted that, because this was a PI hearing, she did not have the benefit of cross-examination of experts.
In Juno, Justice Burley referred to Justice Rofe's comments on this issue in Sandoz, noting that a challenge based on obviousness is a "complex process of fact finding and legal analysis", requiring "the forensic task of considering the expert evidence". Crucially, his Honour noted that at this stage of the matter, lack of inventive step "is ill suited [as an invalidity ground]…to demonstrate weakness in the prima facie case of infringement".
Displacing a prima facie case on infringement requires a compelling non-infringement argument or a strong lack of novelty / internal validity case
While the Sandoz and Juno cases both focused on dosage regimes and the content of the generic/biosimilar's Product Information, the Court reached contrasting conclusions on the prima facie case on infringement. Those findings impacted not only the strength of the prima facie case on infringement, but also weighed significantly in the balance of convenience. Whilst the proportionality of broad injunctive relief was a material consideration for Justice Rofe, this was given less importance by Justice Burley in Juno.
The position of the expert witnesses on infringement issues also appeared to differ. Sandoz's evidence on the relevant alternative treatment protocol was left effectively unanswered by Regeneron/Bayer's experts. By contrast, the judgment reveals this issue was keenly disputed between Juno and Janssen's experts. This highlights the importance of robust evidence directed to the nuances of the parties' competing infringement positions.
As has been demonstrated in previous decisions where a PI has been refused, a strong lack of novelty or internal invalidity ground is often the most effective basis to materially undermine a patentee's prima facie case on infringement, given that these grounds can be more readily assessed on the face of the prior art document or the patent specification.
A unique or "X factor" point may sway the balance of convenience
In both decisions, the strength of the infringement case weighed significantly in the balance.
Obtaining a first mover advantage for a biosimilar – and in particular aflibercept given its specialised administration – was more recognised by Justice Rofe, particularly in light of Regeneron/Bayer's weak prima facie infringement case.
Conversely, Justice Burley emphasised that any first mover advantage accruing to Juno would flow from likely infringing conduct. In weighing the balance, his Honour accepted that Juno had an "arguable" invalidity case, but that "it does not rise to the level that it diminishes the strength of the infringement arguments".
In both decisions the Court recognised that it is more difficult to assess the damage to a party that is wrongly restrained from launching. However, in Juno, Justice Burley noted this would only "tend to tilt the scales a little", whereas it held more weight in Sandoz, in circumstances where the non-infringement case was stronger.
Further, in both cases, the typical balance of convenience factors (such as price discounting and loss of revenue, potential reduction in goodwill, the first mover advantage, difficulty of calculating damages) were considered. However, as Sandoz demonstrates, an additional unique factor, for example if the patentee itself is shifting and/or diminishing the market, can be material to the balance of convenience assessment.
Additionally, whilst this factor did not ultimately tip the balance in Juno, a further important consideration is whether a final decision can be handed down before patent expiry. If there is only a short time left to run on a patent, there is scope to argue that a PI would be akin to final relief.
In light of the many PI applications being refused by the Court since 2018, the question remains whether Juno is an anomaly or the tide is beginning to recede on defeating PIs. Adding to that tide is Justice Downes' decision to grant a PI against Pharmacor in
AstraZeneca AB & Anor v Pharmacor Pty Limited, handed down on 16 February 2026. Stay tuned for our follow-up article reporting on the highly anticipated decision.
Contact
Rebecca Pereira, Jonathan Kelp and Robert Cooper