Regeneron Pharmaceuticals, Inc. v Sandoz Pty Ltd [2025] FCA 1067 | Case summary | Zoe

CTHFCA

Regeneron Pharmaceuticals, Inc. v Sandoz Pty Ltd [2025] FCA 1067

[2025] FCA 1067

Federal Court of Australia|2025-09-03|Before: Rofe J

View original source

General information, not legal advice.

At a glance

Source facts

Court

Federal Court of Australia

Decision date

2025-09-03

Before

Rofe J

Catchwords

**PRACTICE AND PROCEDURE** - patents - interlocutory injunction application - prima facie case - infringement - validity of patents - balance of convenience - application refused

Source

Original judgment source is linked above.

Catchwords

**PRACTICE AND PROCEDURE** - patents - interlocutory injunction application - prima facie case - infringement - validity of patents - balance of convenience - application refused

Judgment (159 paragraphs)

[1]

For the reasons I set out below, I consider that the balance of convenience does not lie in favour of the grant of interim relief sought by the Applicants. Accordingly, I dismiss the interlocutory application for injunctive relief.

[2]

The Applicants rely on evidence from seven witnesses:
- Mark Davies Daniell is a senior consultant ophthalmologist at the Royal Victorian Eye and Ear Hospital (RVEEH) and Clinical Professor (Prof) of Ophthalmology at the Centre for Eye Research, University of Melbourne. Prof Daniell is the immediate past president of the Royal Australian and New Zealand College of Ophthalmologists (RANZCO) and has held numerous professional memberships and positions in societies, councils and authorities. Prof Daniell obtained a MBBS and Master of Surgery degrees from the University of Melbourne in 1986 and 1992, respectively. Prof Daniell gives evidence on the treatment of neovascular AMD and diabetic macular oedema () using intravitreal injections of VEGF antagonist, and how the landscape for the treatment of those conditions will change if a biosimilar to one of the anti-VEGF reference drugs became available in Australia. The Applicants rely on one affidavit affirmed by Prof Daniell.

Sandoz relies on the evidence of five witnesses:

Sanjeewa Sudarshan Wickremasinghe is a consultant ophthalmologist in the Medical Retina Clinic at the RVEEH in Melbourne, a Medical Retina Specialist at Retina Specialists Victoria and a Clinical A/Prof in the Department of Ophthalmology at the University of Melbourne. A/Prof Wickremasinghe obtained his MBBS degree from the University of London in 1996 and a Doctor of Medical Science degree from the University of Melbourne in 2015. A/Prof Wickremasinghe gives evidence about the construction of the 599 Patent and the prior art documents relating to the lack of prima facie case. A/Prof Wickremasinghe gives further evidence on his current prescribing and treating practises in relation to Wet AMD, DME, Myopic CNV and RVO. Sandoz relies on two affidavits affirmed by A/Prof Wickremasinghe.
Robert Kelvin Cooper is a Partner at MinterEllison, the solicitors for Sandoz. Mr Cooper provides evidence on prior disclosures of the amino acid comprising SEQ ID NO:2 of the 599 Patent, which Sandoz relies on for its want of prima facie case. Mr Cooper also gives evidence regarding Sandoz's launch plans for AFQLIR® and the Applicants' promotional materials of EYLEA® 8mg. Sandoz relies on two affidavits affirmed by Mr Cooper.
Dr Louisa Maria King is a Patent and Design Searcher at KingSearch. Dr King obtained her Bachelor of Science degree and a Doctor of Philosophy in organic chemistry from the University of Sydney, in 1987 and 1992, respectively. Dr King provides evidence on how she would have located the amino acid sequence for aflibercept at the relevant priority date. Sandoz relies on one affidavit affirmed by Dr King.
Andrew Murray Ross is a Forensic Accountant and Partner at HKA. Mr Ross has more than 35 years' experience in the preparation of expert reports and the provision of expert evidence as a forensic accountant. Mr Ross gives evidence responsive to Mr Stone. Sandoz relies on one affidavit sworn by Mr Ross.
Timothy Behrens is the Head of Growth at Sandoz. Mr Behrens has significant experience overseeing and forecasting the Australian biosimilar market for biosimilar products and is responsible for the proposed launch of the Sandoz Aflibercept Products. Mr Behrens gives evidence on the development of biosimilar drugs to reference biologics, the biologics market and more specifically the anti-VEGF market. Sandoz relies on one affidavit affirmed by Mr Behrens.

[6]

(a) Wet macular degeneration (Wet AMD): due to abnormal blood vessels growing under the retina, blood and fluid leakage into the retina causes damage to the macula, a specialised area in the centre of the retina, which is responsible for sharp, central vision. Wet AMD is also referred to by certain experts as neovascular age-related macular degeneration AMD (nvAMD).

(b) Choroidal neovascularisation secondary to pathologic myopia (Myopic CNV): caused by abnormal blood vessels growing from the choroid into the sub-retinal space, and into the retina, which can leak and damage the retina.

(c) Diabetic macular oedema (DME): whilst not always caused by angiogenesis, DME occurs as a complication of diabetic retinopathy (DR), which may occur in two stages - non-proliferative or proliferative DR. Non-proliferative DR refers to the early stage of DR and is characterised by blood vessel leakage, whereby capillaries in the retina are damaged and rendered more permeable from microaneurysms, venous beading and intraretinal microvascular abnormalities. The damaged capillaries leak fluid into the macula causing its swelling and thickening. This process is otherwise known as an oedema, which contributes to the blurring of the central vision. Proliferative DR is considered more severe and occurs when the level of ischemia of the retina is such that angiogenesis occurs from increased levels of VEGF. Like Wet AMD, the new and fragile blood vessels can leak or cause haemorrhage or retinal detachment, leading to impairment or loss vision.

(d) Retinal vein occlusion (RVO): refers to the blockage of retinal veins, which constricts blood leaving the eye. As in DME, anti-VEGF agents are used in RVO to reduce blood and fluid leakage in the eye, rather than to address angiogenesis. The visual impairment can be further divided into blockage in the main vein which takes blood out of the eye - central retinal vein occlusion (CRVO) - or blockage in one of the tributaries of the main vein limiting damage to a specific area - branch retinal vein occlusion (BRVO).

[28]

In some cases, a patient may collect the anti-VEGF injection from a chemist or pharmacy located nearby the ophthalmologist's clinic. In such circumstances, the prescription is sent by the ophthalmologist to the dispensing pharmacy. However, these cases are very rare. Mr Behrens deposes that EYLEA®, for instance, is rarely collected by a patient from a retail pharmacy.
The mode of supply, dispensing and administration of VEGF antagonists differs from the ordinary meaning of "retail" supply which involves patients collecting their own prescriptions from retail pharmacies such as Chemist Warehouse or Terry White. The corollary of this important distinction is that, insofar as VEGF antagonists are concerned, pharmacists are unable to swap the Applicants' EYLEA Aflibercept Products for the Sandoz Aflibercept Products. The election between VEGF products occurs almost exclusively at the prescription stage. In other words, only a prescribing ophthalmologist may decide whether to prescribe EYLEA® or the Sandoz Aflibercept Products.
A further unique feature of the aflibercept market is the fact that treatment rates for each of the indications is close to 100% of patients. This means that there is not a significant pool of patients who are undiagnosed, or diagnosed but not seeking treatment, or diagnosed and not complying with treatment requirements. Patients who notice their eyesight being compromised tend to seek treatment promptly.
Another idiosyncrasy of the VEGF antagonist market is that each of the intravitreal injections available in the market (i.e., aflibercept, ranibizumab, and faricimab) use slightly different devices for the administration of the drug. Each vial and/or PFS for each drug is slightly different. In the context of a PFS, ophthalmologists need to become familiar with using each type of syringe. For example, the EYLEA® PFS is substantially different from the PFS for LUCENTIS®, and from the PFS for administering VABYSMO®. Differences include the pressure required to be applied to the plunger, needle gauge, ease of use and familiarity.
In the context of an intravitreal injection, it is critical that the device used enables accurate, efficient and consistent application without introducing infection. Ophthalmologists administer VEGF antagonists to a high volume of patients daily. It is therefore imperative that the ophthalmologist is familiar and comfortable with the device to ensure safe and efficient treatment.

[30]

The principles concerning the grant of an interlocutory injunction are not controversial. An applicant for an interlocutory injunction must show that it has a prima facie case "in the sense that if the evidence remains as it is there is a probability that at the trial of the action" it will obtain the injunctive relief sought and that the balance of convenience favours such injunctive relief: see Samsung Electronics Co Ltd v Apple Inc (2011) 217 FCR 238 at [53]-[55], [60]-[67] (Dowsett, Foster and Yates JJ).
Regarding the first question, a prima facie case does not require that an applicant for interlocutory injunction show that it is more probable than not that it will succeed at trial. It is sufficient that the applicant shows a sufficient likelihood of success to justify in the circumstances the preservation of the status quo pending the trial: Beecham Group Ltd v Bristol Laboratories Pty Ltd (1968) 118 CLR 618 at 622-3, 625 (Kitto, Taylor, Menzies and Owen JJ), cited with approval in Australian Broadcasting Corporation v O'Neill (2006) [2006] HCA 46; 227 CLR 57 at [65] (Gummow and Hayne JJ). The threshold of sufficient likelihood of success recognises the serious consequences for the respondent where the interlocutory relief is granted: Samsung at [51] (Dowsett, Foster, Yates JJ).
The requisite likelihood of success for establishing a prima facie case is not to be considered in isolation and varies depending on the nature of the rights that are being asserted, and the harm likely to flow to the applicant unless the injunction sought is granted.
Where an interlocutory injunction is sought for threatened patent infringement and the alleged infringer alleges the patent is invalid, the applicable principles are found in the summary by the Full Court in Sanofi-Aventis Deutschland GmbH v Alphapharm Pty Ltd (2019) 139 IPR 409 at [8]-[10] (Jagot, Yates and Moshinsky JJ). At [10], their Honours explained that the apparent strength of the parties' substantive cases will often be an important consideration to be weighed in the balance of convenience. Further, at [14], they explained that "a sufficiently strong case of invalidity may well qualify the conclusion that there is a prima facie case of infringement at all. [...] What is relevant is the strength of the case that the claim or claims said to be infringed are invalid".
Regarding the balance of convenience, the relevant considerations under the exercise of the discretion were summarised by the Full Court in Samsung at [66] as follows:

[32]

The 599 Patent claims a priority date of 13 January 2011. The claims alleged to be infringed are 1, 3, 4, 5 and 12 (Asserted 599 Claims). The parties accept that the post-Raising the Bar form, being that amended by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth), of the Act applies.
The 599 Patent describes the field of invention as relating to the field of therapeutic treatments of eye disorders and, more specifically, the administration of VEGF antagonists to treat eye disorders caused by or associated with angiogenesis.
Various eye disorders associated with pathological angiogenesis are described at [0002], including Wet AMD, DME and CRVO. At [0003], the 599 Patent notes that FDA-approved treatments of angiogenic eye disorders such as AMD and CRVO include the administration of an anti-VEGF antibody called ranibizumab (LUCENTIS® Genentech, Inc.) on a monthly basis by intravitreal injection. Despite methods for treating eye disorders using VEGF antagonists being mentioned in prior patents, the 599 Patent notes that "there remains a need in the art for new administration regimens for angiogenic eye disorders especially those which allow for less frequent dosing while maintaining a high level of efficacy".
The invention is summarised in [0005] as methods for treating angiogenic eye disorders comprising sequentially administering multiple doses of a VEGF antagonist to a patient over time. The sequential administration of a VEGF antagonist is specified as comprising a single initial dose of a VEGF antagonist, followed by one or more secondary doses of the VEGF antagonist, followed by one or more tertiary doses of the VEGF antagonists. In the same paragraph, the Specification notes that the "[...] inventors have surprisingly discovered that beneficial therapeutic effects can be achieved in patients suffering from angiogenic eye disorders by administering a VEGF antagonist to a patient at a frequency of once every 8 or more weeks, especially when such doses are preceded by about three doses administered to the patient at a frequency of about 2 to 4 weeks."
In the same paragraph, the Specification states that "thus, according to the methods of the present invention, each secondary dose of VEGF antagonist is administered at least 8 weeks after the immediately preceding dose". One advantage of such a dosing regimen is identified as allowing for less frequent dosing, as "for most of the course of treatment (i.e., the tertiary doses), it allows for less frequent dosing (e.g., once every 8 weeks) compared to prior administration regimens for angiogenic eye disorders which require monthly administrations throughout the entire course of treatment. See, e.g., prescribing information for LUCENTIS® [ranibizumab], Genentech, Inc.)".
The wording of the Specification does not specify or limit the identity of the VEGF antagonist to which the method applies. Indeed, the Specification states at [0007] that:

[38]

At [0019], the Specification notes that the methods of the invention may comprise administering to a patient any number of secondary and/or tertiary doses of a VEGF antagonist.
From [0021] to [0022], the term "VEGF antagonists" means "any molecule that blocks, reduces or interferes with the normal biological activity of VEGF" and is said to include "molecules which interfere with the interaction between VEGF and a natural VEGF receptor. Specific exemplary VEGF antagonists include anti-VEGF antibodies, anti-VEGF receptor antibodies, and VEGF receptor-based chimeric molecules (also referred to herein as "VEGF-Traps")."
At [0030], the Specification describes each dose of the dosing regimen as a "therapeutically active amount". A therapeutically active amount is defined as a "dose of VEGF antagonist that results in a detectable improvement in one or more symptoms or indicia of an angiogenic eye disorder, or a dose of VEGF antagonist that inhibits, prevents, lessens, or delays the progression of an eye disorder". In the case of an anti-VEGF antibody or a VEGF receptor-based chimeric molecule such as VEGFR1R2-FcC1(a), a therapeutically effective amount can be from about 0.05 mg to about 5 mg.
At [0032], the Specification specifies, using the number of weeks as a parameter, when the claimed invention begins to demonstrate efficacy. More specifically, it discloses that the methods of the invention demonstrate efficacy within 104 weeks of the initiation of the treatment regimen. In the same paragraph, the Specification expressly notes that the measurement of the 104 weeks begins with the initial dose administered at week 0. Efficacy is then defined by reference to the Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity charts. For AMD, CRVO and DME, "efficacy" means where "the patient exhibits a loss of 15 or fewer letters on ETDRS visual acuity chart". In certain embodiments, "efficacy" also means a gain of one or more letters on the ETDRS chart.
At [0034], the Specification discloses that the VEGF antagonist used in all Examples in the 599 Patent is "a dimeric molecule having two functional VEGF binding units" with each functional binding unit comprising an "Ig domain 2 from VEGFR1 fused to Ig domain 3 from VEGFR2, which in turn is fused to the hinge region of a human IgG1 Fc domain VEGFR1R2-FcC1(a); encoded by SEQ ID NO:1". The paragraph further specifies that for the purposes of the Examples, the method of using the specific VEGF antagonist encoded by SEQ ID NO:1, referred to as "VEGFT", the dosing regimen is to be measured by a parameter where "monthly" dosing is equivalent to dosing once every four weeks.

[53]

The interval between doses injected into the same eye should not be shorter than one month.

Advice on treatment initiation and maintenance of therapy specific to each patient population is described in the section below. Once optimal visual acuity is achieved and/or there are no signs of disease activity, treatment may then be continued with a treat-and-extend regimen with gradually increased treatment intervals to maintain stable visual and/or anatomic outcomes. If disease activity persists or recurs, the treatment interval may be shortened accordingly. Monitoring should be done at injection visits. The monitoring and treatment schedule should be determined by the treating ophthalmologist based on the individual patient's response. If visual and anatomic outcomes indicate that the patient is not benefiting from continued treatment, AFQLIR should be discontinued.

• Treatment of neovascular (wet) age-related macular degeneration (wet AMD)

AFQLIR 2 mg treatment is initiated with one AFQLIR 2 mg injection per month for three consecutive months, followed by one injection every two months.

Based on the physician's judgement of visual and/or anatomic outcomes, the treatment interval may be maintained at two months or further extended using a treat-and-extend dosing regimen, by increasing injection intervals in 2- or 4-weekly increments while maintaining stable visual and/or anatomic outcomes. If visual and/or anatomic outcomes deteriorate, the treatment interval should be shortened to a minimum of four weeks based on anatomical and/or visual outcomes.

[...]

• Treatment of diabetic macular oedema (DME)

AFQLIR 2 mg treatment is initiated with one AFQLIR 2 mg injection per month for five consecutive months.

Following the initiation period and based on the physician's judgement of visual and/or anatomic outcomes, the treatment interval may then be maintained at an injection every two months or further individualised, such as with a treat-and-extend dosing regimen, by increasing injection intervals in 2- or 4-weekly increments while maintaining stable visual and/or anatomic outcomes. If visual and/or anatomic outcomes deteriorate, the treatment interval should be shortened accordingly. Treatment intervals shorter than 4 weeks or longer than 4 months have not been studied (see Section 5.1 Pharmacodynamic properties, Clinical trials).

(Emphasis added in italics.) (Emphasis in original in bold.)

[57]

48. In wet AMD, this consists of administering injections every 4 weeks until the patient has achieved their best vision and their retina is free from fluid. These doses are referred to as "loading doses". Following achievement of that stability, I then extend the interval between each injection. I would most commonly extend the interval by 2 weeks at each visit. For example, if the patient's macula is dry after the last injection at a 4 week interval, I will administer the next injection at a 6 week interval, and then at 8 weeks, 10 weeks and 12 weeks, and so on, until the patient reaches a point where fluid returns or their vision decreases. Once that happens, I decrease the interval to the interval at which they were last stable. For example, if there was a recurrence at 14 weeks, I might try an interval of 10 weeks or 12 weeks and maintain that interval for a few doses before challenging the patient with a higher interval again, once their disease is stable and they are suitable to extend again.

49. The patient's suitability to extend a dosage interval (whether for the first time, or after decreasing an interval) is based on an assessment of the amount of fluid in the patient's macula, their level of vision, and how long their vision has remained that way. It also depends on my and the patient's willingness to extend out their injections and the risk of doing so. For example, if the patient's other eye has good vision, there is less at risk in challenging the eye I am treating. However, if their other eye has poor vision, I would take a more cautious approach to extending dosage intervals for the eye I am treating.

[...]

53. After the initial doses, and then a dose at typically 6 weeks, approximately:

(a) 40% of my wet AMD patients receive their doses between 8 and 10 weeks apart;

(b) 30% of my wet AMD patients are "stuck" at receiving a dose between 4 to 6 weeks apart, as their disease deteriorates when attempting to extend further than that interval; and

(c) 30% of my wet AMD patients can tolerate an interval of 12 to 16 weeks between doses.

54. Additionally, sometimes patients effectively extend their dosage themselves. For example, if their next dose is scheduled at an 8-weekly interval, but they reschedule that appointment for a week later, they have extended their dose to 9 weeks. If they tolerate that extension, I will likely extend again to 10 weeks for their next dose. However, if the 9-week interval prompts additional disease activity, then I will reduce the interval to 8 weeks again. I estimate that this unintentional extension occurs at least once for approximately 5% of the patients that I treat for wet AMD.

55. Further, notwithstanding the best intentions to treat patients at the precise interval intended (e.g. 2, 4, 6 or 8 weeks) often that is not possible due to practical realities such as patient inconvenience, doctor availability, illness and public holidays. Across any given patient's treatment course, there will invariably be at least some deviations from the intended treatment interval of at least several days.

(Emphasis added.)

[60]

The principles of claim construction are well-established as set out in Jupiters Ltd v Neurizon Pty Ltd [2005] FCAFC 90; (2005) 65 IPR 86 at [67] (Hill, Finn and Gyles JJ); Décor Corporation Pty Ltd v Dart Industries Inc [1988] FCA 399; (1988) 13 IPR 385 at 400 (Sheppard J); and Flexible Steel Lacing Co v Beltreco Ltd [2000] FCA 890; (2000) 49 IPR 331 at [70]-[81] (Hely J). It suffices to note that the claims are to be construed "[...] from the perspective of a person skilled in the relevant art as to how such a person, who is neither particularly imaginative nor particularly inventive (or innovative), would have understood the patentee to be using the words of the claim in the context in which they appear [...] in the light of the common general knowledge before the priority date": Product Management Group Pty Ltd v Blue Gentian LLC [2015] FCAFC 179; (2015) 116 IPR 54 at [35] (Kenny and Beach JJ).
Claims are also to be construed in the context of the specification as a whole. However, it is not legitimate to narrow or expand the boundaries of the monopoly as delineated by the words of a claim via the addition to those words glosses drawn from other parts of the specification. Where terms are unclear, those terms may be defined by reference to the body of the specification, but the body of the specification cannot be used to change a clear claim for one subject matter into a claim for another and different subject matter.
It is well-established that a patent specification should be given a "purposive" construction, not a "purely literal", "meticulous verbal analysis" or a "too technical or narrow" construction. A specification should be read in a practical and common-sense way: see Glass Hardware v TCT Group Pty Ltd (2024) 182 IPR 1 at [87] (Yates, Charlesworth and Rofe JJ), and cases cited therein.
To apply a "purposive" construction does not justify extending the patentee's monopoly to the "ideas" disclosed in the specification: at [40] of Blue Gentian (per Kenny and Beach JJ). Instead, its application should be guided by considerations including those addressed by Hoffmann LJ when his Honour explained "purposive construction" in Kirin-Amgen Inc v Hoechst Marion Roussel Ltd (2004) 64 IPR 444 at [33]-[34], a passage which has been endorsed by the Full Court including in Blue Gentian at [41] and Tramanco Pty Ltd v BPW Transpec Pty Ltd [2014] FCAFC 23; (2014) 105 IPR 18 at [174] (Nicholas J). The foregoing was fleshed out by the Full Court in Glass Hardware at [88]:

[64]

The relevant claim integer 1.3 is "wherein each secondary dose is administered 2 to 4 weeks after the immediately preceding dose [...]".
All discussion of periods of time between doses, as pertaining to the dosing regimen of the claimed invention, in the Specification is measured in terms of weeks, from 2 weeks to 104 weeks. Whilst there is description of time periods between administration by reference to months, these descriptions are strictly confined to the Examples.
At [0005], the Specification describes an advantage of the dosing regimen of the invention as being that for most of the course of treatment (i.e., the tertiary doses) it allows for "less frequent dosing (e.g., one every 8 weeks)" compared to prior regimens which required monthly administration.
At [0009], the Specification notes that aflibercept was approved by the FDA in November 2011, for the treatment of patients with Wet AMD, with a recommended dose of 2 mg administered by intravitreal injection every 4 weeks for the first three months, followed by 2 mg administered by intravitreal injection once every 8 weeks. The dosing regimen is again described via the parameter of "weeks" to measure the period between doses. It is apparent that the term "months" is used only as a descriptor of the duration of the dosing regimen, not the time periods between each dose within the dosing regimen itself.
Figure 1 shows an exemplary dosing regimen of the invention. In that dosing regimen, two secondary doses are administered at weeks 4 and 8 respectively, and at least six tertiary doses which are administered once every 8 weeks thereafter (i.e. at weeks 16, 24, 32, 40, 48, 56). The doses are depicted in Figure 1 as falling precisely on the week marker. There is no depiction of any margin for error, or +/- any days from the week mark.
At [0017], the Specification describes one exemplary embodiment of the invention wherein each secondary dose is administered 2 to 4 (e.g., 2, 2½, 3, 3½, or 4) weeks after the immediately preceding dose. The examples of what constitute "2 to 4 weeks" do not contemplate any period outside that range, including outside the upper limit of 4 weeks such as 4½ weeks. The discussion in [0018] of another embodiment speaks of the secondary doses being administered 4 weeks after the preceding dose "(i.e., at week 4 and at week 8)". I have emphasised the use of "at" rather than "in", which suggests a precision in dosing on the 4-week mark, rather than in or around the fourth week.
At first blush, the precision of the patentee's reference to time periods in weeks sits slightly awkwardly with the evidence regarding ophthalmologist practice given by A/Prof Wickremasinghe. In the context of discussing the 8 weeks integer, his evidence was that in the case of patients being maintained on 8 weekly dosing, as a matter of practical reality, in his experience none of those patients would be administered aflibercept exactly every 8 weeks (i.e., every 56 days). Given the exigencies of life, illness and public holidays, patients will often receive a dose at a different interval such as after 7 weeks and 5 days, after 8 weeks and 4 days or after 9 weeks. Across any given patient's treatment course, A/Prof Wickramasinghe considered that there will invariably be at least some deviations from the intended treatment interval of at least a few days.
The Applicants contend for a purposive construction in which 4 weeks is equivalent to a month. In particular, they refer to [0034] which notes that "[f]or the purposes of the following Examples "monthly" dosing is equivalent to dosing once every four weeks". Importantly, as I noted above, this comment is expressly restricted to the Examples rather than being inclusive of the description of the invention and its embodiments. Notwithstanding that comment, and a discussion of the prior art dosing regimen in terms of "monthly administrations", the patentee deliberately chose to use weeks rather than months in the claims, and not to adopt qualifiers such as "at least" or "from" or "around" in relation to the specified number of weeks.
Adopting a commonsense construction, my provisional view is that the "2 to 4 weeks" integer requires the secondary dose to be administered in or within the time period range of 2 to 4 weeks after the immediately preceding dose. Claim 1 claims a range of time in which the secondary dose is administered after the immediately preceding dose, being within 2 to 4 weeks. The examples given or the "id est" specification (meaning, "that is") of what is meant by that range are all within the range, capped at four weeks. There is no discussion of any secondary dose being given after 4 weeks. There is no reference to "at least 2 to 4 weeks" or other qualification to that range in the Specification. The two-week range of "2 to 4 weeks" provides sufficient flexibility to allow for illness, public holidays or other exigencies, so that the dose will be within that specified window, without the need for inferring any +/- days outside that window. The flexibility afforded by the range "2 to 4 weeks" resolves the apparent tension between the patentee's choosing of the term "weeks" and A/Prof Wickremasinghe's evidence on practical administrative realities. Indeed, what appeared to be overparticular diction with respect to the word "weeks" is ameliorated by the range of two weeks provided by "2 to 4 weeks".
The Sandoz PI provides that for Wet AMD, secondary doses of AFQLIR® are one AFQLIR® 2 mg injection per month for three consecutive months, followed by one injection every two months.
The Applicants contend that the Sandoz PI itself treats 4 weeks as equivalent to one month. The Sandoz PI expressly refers to "[t]reatment intervals greater than four months (16 weeks) [...] between injections" as not having been studied. Importantly, the reference of weeks pertains to hypothetical treatment intervals not falling within the guideline dosing regimen of AFQLIR®. The Sandoz PI does, on the other hand, provide that the intervals between doses comprising the dosing regimen "should not be shorter than one month". The reference to "a minimum of four weeks" is reserved for circumstances where "visual and/or anatomic outcomes deteriorate", necessitating greater precision by reference to "weeks" rather than "months". Rather than grounding a general implication that the Sandoz PI treats

[67]

The relevant claim integer 1.4 is "wherein each tertiary dose is administered 8 weeks after the immediately preceding dose [...]".
The Specification describes the administration of tertiary doses other than at precisely 8-week intervals. At [0017], the Specification speaks of a tertiary dose being administered "at least 8 (e.g. 8, 8½, 9, 9½, 10, 10½, 11, 11½, 12, 12½, 13, 13½, 14, 14½, or more) weeks after the immediately preceding dose".
At [0020] of the Specification, a further embodiment of the invention is described wherein each subsequent tertiary dose is administered "from 8 to 12". Examples from 8 to 12 weeks are again given (e.g. 8, 8½, 9, 9½, 10, 10½, 11, 11½, 12). The Specification also notes in the same paragraph that the frequency of administration may be adjusted during the course of treatment depending on the needs of the individual patient.
The Applicants refer to public literature and ophthalmologist practices to urge a purposive construction in which 8 weeks (or 56 days) is equivalent to two months. I note that two months will range from 59 days (February to March) to 62 days (July to August), both of which are in excess of 56 days and/or 8 weeks.
Adopting a commonsense construction, my provisional view is that the "8 weeks" integer requires the tertiary dose to be administered 8 weeks after the immediately preceding dose. I consider that "8 weeks" will encompass 8 weeks +/- one day on either side to accommodate for public holidays, the exigencies of life and other delays including those of administrative nature. The Specification contemplates periods between administration of tertiary doses of "at least" 8 weeks or "from" 8 weeks and expressly refers to the prior regimen for VEGF antagonist administration requiring "monthly" administration. However, the patentee deliberately chose to claim "8 weeks" without adding any qualification to that period.
Whilst the published literature and ophthalmologist practice may suggest a more flexible approach to periods of time in VEGF antagonist dosing regimens, the patentee has chosen to use the specific language of 8 weeks without qualification.
Relevantly, the Full Court in Glass Hardware observed at [96]:

[73]

The Applicants submit that, the combined effect of s 117 as applied on the present facts is such that the threat of infringement is the supply of the Sandoz Aflibercept Products along with their corresponding Sandoz PI. The Applicants contend that in circumstances where the relevant product is one that may be used for both infringing and non-infringing uses, the operation of s 117(1) is to be read with s 117(2)(b) of the Act.
To enliven the infringement by supply provision under s 117(1), the Applicants refer to "use" under s 117(2)(b) and s 117(2)(c) of the Act. Regarding s 117(2)(b), the Applicants' position is that the Sandoz PI demonstrates that Sandoz (the supplier) had reason to believe that ophthalmologist (the person) would put the Sandoz Aflibercept Products to use in a manner that infringes the Asserted 599 Claims. Similarly, the Applicants' s 117(2)(c) claim is that the Sandoz PI are "instructions" or, in the alternative, an "inducement" to use the Sandoz Aflibercept Products in a manner that would infringe the method of the Asserted 599 Claims.
In response to the Applicants' s 117(2)(c) infringement by supply case, Sandoz advances three arguments. First, it contends that the Sandoz PI does not constitute "instructions" within the meaning of s 117(2)(c). Sandoz refers to the permissive rather than instructive language of the Sandoz PI. For the treatment of Wet AMD, the Sandoz PI states that "the treatment interval may be maintained at two months"; similarly, for the treatment of DME, the Sandoz PI states that "the treatment interval may then be maintained at an injection every two months". Sandoz relies on Allied Pumps Pty Ltd v LAA Industries Pty Ltd [2023] FCA 1457; (2023) 179 IPR 1 at [709] (Downes J) to characterise the Sandoz PI as a document specifying a particular use, which it argues is to be distinguished from a document instructing a person to engage in that use. Sandoz also relies on the "overarching statement" of the Sandoz PI, which stipulates that "[t]he monitoring and treatment schedule should be determined by the treating ophthalmologist based on the individual's response" (the Overarching Statement).
Secondly, Sandoz submits that even if the statements are an instruction, such statements do not instruct ophthalmologists to use the Sandoz Aflibercept Products on an infringing basis. In so doing, Sandoz refers to the above Overarching Statement along with other statements in the Sandoz PI which account for the individualisation and adjustment of treatment protocols by ophthalmologists, as appropriate.
Thirdly, Sandoz submits that the Sandoz PI is not "given to" the ophthalmologist as required by s 117(2)(c) of the Act. Sandoz says that it will not proactively distribute the Sandoz PI to ophthalmologists or patients, and that any such distribution will be upon the request, which is rare. The only document which Sandoz says will be "given to the" ophthalmologist is the Consumer Medicine Information (CMI), which Sandoz says does not contain any of the Sandoz PI statements that are impugned by the Applicants.
Regarding s 117(2)(b) of the Act, Sandoz submits that the evidence does not establish a prima facie case that a reasonable person in Sandoz's position would have reason to believe that ophthalmologists would engage in infringing use, by reason of the fact that most ophthalmologists adopt a treat-and-extend dosing regimen which is outside the claimed method.

[77]

Notwithstanding Sandoz's counterarguments regarding the permissive language of the Sandoz PI and the Overarching Statement, I accept, at this preliminary stage, the Applicants' submissions that the indications in the Sandoz PI constitute "instructions" within the meaning of s 117(2)(c) of the Act. In Interpharma, Kenny J considered whether the words "may be initiated" in the product information of a generic drug, i.e., the "Amended Generic Products PI", used in conjunction with statements as to the use of a "loading dose" of dexmedetomidine constituted an "instruction" to use a general recommended loading dose of dexmedetomidine for sedation of initially intubated patients during treatment in an intensive care setting.
In that case, Interpharma argued that the inclusion of the word "may" in the Amended Generic Products PI caused the statement regarding the initiation with a loading infusion to fall short of an instruction within the meaning of s 117(2)(c). Interpharma relied on evidence from a clinician that the replacement of the words "is generally" with the word "may" in the Amended Generic Products PI informed him and other clinicians "that the administration of dexmedetomidine using a loading dose or bolus is not a matter of routine" and that the words "may be" would cause to "pause and think more about the risks of the drug versus the benefits". In contrast, Hospira relied on the majority's observations in AstraZeneca AB v Apotex Pty Ltd [2014] FCAFC 99; (2014) 226 FCR 324 at [427] (Besanko, Foster, Nicholas and Yates JJ) and Jagot J's analysis in Beadcrete Pty Ltd v Fe Yu (t/as Jewels 4 Pools) (No 3) [2013] FCA 187; (2013) 100 IPR 188 to contend that it was sufficient that the Amended Generic Products PI indicated that dexmedetomidine can be administered in a loading dose to constitute an instruction for the purpose of s 117(2)(c): see Interpharma at [268]. Further, Hospira relied on the evidence of two experts who expressly disagreed with Interpharma's experts' understanding of the Amended Generic Products PI, asserting that "[t]he Amended [Generic Products] PI [...] provide[d] instructions on how to administer a loading dose, should the clinician do so".
Justice Kenny held at [274] that the Amended Generics Products PI did not discourage clinicians from giving a loading dose or warn against that use. On the contrary, the Amended Generic Products PI "expressly instruct[ed] a clinician that a loading dose of the Generic Products may be administered for the initiation of ICU sedation "if needed"". That the statement regarding the giving of a loading dose was to be qualified by a physician's judgment did not preclude her Honour's finding at [280] that the relevant statements were "instructions" within the meaning of s 117(2)(c).
Given the similarity of permissive language between the Sandoz PI and the Amended Generic Products PI in Interpharma, I reject Sandoz's submission that the permissive language of the Sandoz PI is sufficient to render the relevant statements not "instructions" within the meaning of s 117(2)(c). As in Interpharma, that the Overarching Statements qualify the instructions by allowing the ophthalmologist to use the Sandoz Aflibercept Products as they deem appropriate does not preclude the fact that the relevant statements in the Sandoz PI provide instructions for use.
Ultimately, whether the Sandoz PI or the approved indications amounts to an instruction for the purposes of s 117(2)(c) is a question for the court. The evidence of expert witnesses may be of assistance but is not necessarily determinative: Neurim Pharmaceuticals (1991) Ltd v Generic Partners Pty Ltd (No 5) [2024] FCA 360 at [289] (Nicholas J). Whilst it may be the case that significant variation and customisation of the dosing regimen may occur in the use of the Sandoz Aflibercept Products due to an ophthalmologist's judgment and intervention, if one of such uses infringes the 599 Patent, infringement is established under s 117(2)(c). This issue was considered in Beadcrete, where the patent in suit claimed inter alia glass beads of a certain size averaging between 1.5 mm and 4.5 mm. The respondents there supplied glass beads in various colours, wherein the green and ice blue glass beads were found to have had an average size that fell within the relevant claim. In those circumstances, the use of green and ice blue glass beads, either in isolation or in combination with each other, would have infringed the patent in suit. At [126], Jagot J explained that the "instructions" in the respondent's sales materials were:

[106]

Sandoz submits that each of Dixon and Adis anticipate the claimed invention because the SEQ ID NO:2, or the amino acid sequence of aflibercept was, at the priority date, "inherent" in each of the prior art documents. Sandoz contends that, before the priority date, aflibercept was known and that a skilled addressee could have, if they wished, located the amino acid sequences in the public USP Dictionary or INN List.
Sandoz's contention is not entirely based on the skilled reader resorting to other documents outside Adis or Dixon to provide the missing sequence information. Rather, Sandoz submits that there is anticipation where the prior art describes something in different terms than what is claimed, but where carrying out the directions in the prior art "will inevitably result in something being made or done which [...] would constitute an infringement of the patentee's claim [...]" (Inevitability Argument): Novozymes A/S v Danisco A/S [2013] FCAFC 6; (2013) 99 IPR 417 at [145] (Jessup J), citing General Tire and Rubber Co v Firestone Tyre and Rubber Co Ltd (1971) 1A IPR 121 at 138 (Sachs LJ); see also Mylan Health Pty Ltd v Sun Pharma ANZ Pty Ltd [2020] FCAFC 116; (2020) 279 FCR 354 at [84] (Middleton, Jagot, Yates, Beach and Moshinsky JJ).
Sandoz further relies on AstraZeneca to support its Inevitability Argument. Sandoz accepts that where a patentee claims a second medical use of a known substance, as the Applicants do in this case, the claimed use may be novel notwithstanding that said second medical use inevitably arises from use according to a first medical use. However, Sandoz submits that the novelty of such second medical use claims must arise from the purpose of the use. In this case, Sandoz contends that given that the purpose of use for treatment of retinal and angiogenic eye disorders was disclosed in Dixon and Adis before the priority date, the Asserted 599 Claims which claim such previously disclosed second medical use or purpose cannot be novel or inventive.
Sandoz also contends that it is a "hard fact" that aflibercept has the SEQ ID NO:2 in the claims. As such, Sandoz submits that the practical inability of a skilled person to obtain the proprietary aflibercept compound at the priority date is not fatal to its anticipation case. Sandoz relies on the proposition that the novelty analysis is hypothetical, being concerned with what would happen if the process were carried out, not the doing, or the ability to do it. The focus is on the content of the information: Novozymes A/S [177] (Jessup J). Sandoz argues that because there is only one compound which is 'aflibercept', the use of that one compound according to Adis or Dixon must be anticipatory.
Novozymes A/S concerned a process for preparing a foodstuff using an enzyme to produce an emulsifier from a fatty acid ester and in which another functional ingredient was produced from a second constituent of the composition. One of the prior art documents cited by the alleged infringer, the "Novo Patent", described the use of an enzyme which "inherently possessed" dual phospholipase/lipase activity for baking bread. Further, evidence showed that an emulsifier and a second functional ingredient were produced by the enzyme. However, this dual activity was not explicitly disclosed in the document.
The Full Court in Novozymes A/S, considered that where a patent claims a product with certain characteristics, it is not necessary for the prior art to describe those characteristics expressly in order for it to provide novelty-defeating disclosure, if by carrying out the prior art description the claimed product would inevitably be obtained. Justice Jessup (with whom Greenwood and Yates JJ agreed) observed the following:

[110]

The difference between the present case and Novozymes A/S is that in that case, the starting enzyme, the phospholipase derived from Fusarium oxysporum, was described and characterised. Such description and characterisation included the amino acid sequence of the phospholipase of interest. Working the process directed in the relevant example, the use of that starting enzyme inevitably involved every integer of the claimed process as the enzyme inherently had the dual activities required.
Section 7(1) of the Act provides that an invention is taken to be novel when compared to the prior art base unless it is not novel in light of certain kinds of information, each of which must be considered separately unless the person skilled in the art would treat them as a single source of information. Here, the prior art information must be made publicly available in a single document before it can be novelty defeating.
In contrast to Novozymes A/S, Adis and Dixon lack the information as to the amino sequence of aflibercept. The dosing regimen in Dixon cannot be used or applied to treat an angiogenic eye disorder (so as to infringe or satisfy the reverse infringement test) without the person skilled in the art first being in possession of a VEGF antagonist composed of inter alia SEQ ID NO:2 or aflibercept. The person skilled in the art has the name "aflibercept" but has no information on its amino acid sequence. Without that information, the skilled person is unable to work the directions disclosed in Adis or Dixon. It is not merely a matter of the practicality in sourcing reagents.
The issue in this case is that the identity and structure of aflibercept, including its primary structure, is not disclosed in any of the cited prior art documents. Further, that information would not otherwise be known to the skilled person without impermissible reference to other public sources of information. That the "identity and structure" of the molecule "goes to the heart of the disclosure and information" disclosed by the prior art was affirmed by the Full Court in Sandoz AG v Bayer Intellectual Property GmbH [2024] FCAFC 135; (2024) 183 IPR 309 at [117] (Yates, Burley and Downes JJ). As the Full Court observed at [118], "[t]his is an information deficiency [...] which constrains the skilled person in the hypothetical". The Full Court's comments, although made in the context of inventive step, are apposite to the novelty case here. Without the essential integer of the amino acid sequence of aflibercept, the skilled person is unable to work the directions in Adis or Dixon, without impermissible resort to other sources of information. Sandoz's attempt to plug the information gap in Adis and Dixon with its inherency and inevitability argument must fail at this preliminary stage.

[112]

Sandoz made no oral submissions on its lack of inventive step case and relied on its written submissions. Sandoz's position in its written submissions is that the Asserted 599 Claims do not involve any "step" over the prior art which can be characterised as inventive. Sandoz submits that the mere inclusion of the aflibercept amino acid sequence in a claim does not render it inventive; it simply characterises the same active ingredient or API. In other words, Sandoz's obviousness case is premised on a definition of the claimed invention as the mere characterisation of aflibercept.
In contrast, the Applicants submit that the claimed invention is to be defined as the pre-clinically and clinically corroborated dosing regimen which allows for inter alia the extension of monthly doses to 8-week doses beyond a certain tertiary dose. In oral submissions, Senior Counsel for the Applicants described the "heart of the invention" as "the move from four-weekly dosing to eight-weekly dosing".
The Applicants contend that even if it was common general knowledge that an extension of time between doses was desirable, without the sequence for the drug disclosed in the prior art, the skilled addressee could not have taken the development of the drug forward. The Applicants argue that even if the person skilled in the art could add the aflibercept sequence to the prior art, the skilled addressee would not be directly led to try the claimed dosage regimen, or to do so with the requisite expectation of success.
In seeking to distinguish from AstraZeneca AB v Apotex Pty Ltd [2015] HCA 30; (2015) 323 ALR 605 (French CJ, Kiefel, Gageler, Keane and Nettle JJ), the Applicants noted that the present dosing regimen of VEGF antagonist is not one that could safely be made about safety and efficacy. Further, distinguishing from Sandoz AG, the Applicants submit that the claimed dosing regimen in this case had not been identified at the outset as the "ideal dosage regimen".
For the present purposes, it is neither necessary nor desirable to finally define the invention. This is in part because I am not assisted by the benefit of the examination and cross-examination of the experts, and because the fact that Sandoz's lack of inventive step argument must, at this stage, fail is dispositive. In essence, the inventive step case of the interlocutory injunction application rises or falls with the inherent novelty case. At present, there is no evidence that each of Adis, Dixon and SEQ ID NO:2 as found in the USP Dictionary and INN List were part of the common general knowledge. Alternatively, the current evidence does not appear to suggest that the skilled person would read Adis or Dixon together with the reference documents. A/Prof Wickremasinghe deposes that he "would have located the amino acid sequence for aflibercept had [he] wanted to do so" and that he "would have expected this information to be located in an online database or a patent document". However, this evidence is, on itself, insufficient to establish that the skilled person would have read any of the cited prior art together with the aflibercept amino acid sequence available at the public dictionaries. The necessity for further and better evidence in this regard is highlighted in light of A/Prof Symons' evidence to the contrary. A/Prof Symons deposes that he is "not sure what online databases A/Prof Wickremasinghe is referring to", stating that he "was unaware of any databases, online or otherwise, which could be used to identify the amino acid sequence of a drug".
In the event that, at trial, Sandoz is able to establish that the SEQ ID NO:2 was part of the common general knowledge, or that the skilled person would read Adis or Dixon together with the reference documents containing the amino acid sequence of aflibercept, I consider that Sandoz may have a compelling lack of inventive step case based on the reasoning of the Full Court in Sandoz AG, wherein the entry of a first-in-class molecule into preclinical trials was held to be sufficient to render the relevant claimed invention obvious in the absence of anticipated hurdles.
In this regard, and on the evidence before me, I reject the Applicants' attempt to distinguish Sandoz AG, along with its corresponding submission which characterises the present dosing regimen as not or less than "ideal" in light of the cited prior art documents. Dixon and Adis both demonstrate that before or at the priority date, aflibercept was being tested for treatment of angiogenic eye disorders according to the claimed dosing regimen, in Phase III clinical trials. In other words, the dosing regimen as claimed by the 599 Patent had already undergone pre-clinical trial and Phases I-II of clinical trial. Without more, it does not appear to be open to the Applicants to argue that, considering Dixon or Adis, the skilled person would have an expectation other than that the development process for aflibercept treatment regime would be routine or straightforward.

[116]

The broad mechanism of action whereby VEGF antagonists block VEGF's action and prevent the formation of new blood vessels, thereby inhibiting the angiogenic properties of VEGF was discussed in the Specification. This mechanism was known by each of the experts and would have been part of the common general knowledge of the person skilled in the art at the priority date.
The Specification at [0002] discusses eye disorders associated with pathological angiogenesis, giving as examples AMD, DME, CNV and CRVO, along with the mechanisms by which those diseases manifest in the eye. Further in that paragraph, the Specification notes that the release of VEGF contributes to increased vascular permeability in the eye and inappropriate new vessel growth. Thus, inhibiting the angiogenic-promoting properties of VEGF appears to be an effective strategy for treating angiogenic eye disorders.
Aflibercept was not the first VEGF antagonist used to treat angiogenic eye disorders. The Specification gives the example of a FDA-approved anti-VEGF drug, ranibizumab (LUCENTIS®) at [0003]. At [0004], the Specification lists five United States patents relating to methods of treating eye disorders using VEGF antagonists. Further discussion of angiogenic eye disorders is found at [0025]. These disorders are said to be "any disease of the eye which is caused by or associated with the growth or proliferation of blood vessels or by blood vessel leakage".
A/Prof Symons' evidence was that he understood that "growth or proliferation of blood vessels" refers to angiogenesis and "blood vessel leakage" refers to vascular permeability. Each of Wet AMD, DR, DME and CRVO is caused by or is associated with angiogenesis and/or vascular permeability.
A/Prof Symons' evidence was that by the priority date, it was known that VEGF played a key role in the regulation of abnormal and normal angiogenesis and the induction of vascular permeability, in both cancers and other neovascular disorders such as nvAMD. He was aware at the priority date that VEGF antagonist or anti-VEGF therapies were being used and/or investigated to treat various retinal disorders such as nvAMD, DME, DR and RVO due to their therapeutic effects on angiogenesis and/or vascular permeability.
At the Priority Date, A/Prof Symons was aware of the anti-VEGF drug ranibizumab (LUCENTIS®) and that it was the primary method of treatment for nvAMD and was approved by the FDA for this indication in 2006 and by the TGA in 2007. He was aware from his own clinical experience that LUCENTIS® was the most effective and widely used treatment available for nvAMD (and later in 2012, for DME) in both Australia and the United States. A/Prof Wickremasinghe's evidence was that LUCENTIS® had been available in Australia since around 2009.
The claims are limited to treating angiogenic eye disorders, with claim 4 being specifically limited to an eye disorder selected from five specific angiogenic eye disorders and claim 5 being limited to age related macular degeneration. While A/Prof Wickremasinghe deposed that in his experience the treatment of DME is less predictable than that of Wet AMD because factors that influence its progression are multifactorial, rather than being primarily driven by VEGF, there was no evidence of any angiogenic eye disorder that did not respond to treatment with a VEGF antagonist.
On the evidence as it currently stands, I consider that there is support for the claims to the treatment of angiogenic eye disorders using the VEGF antagonist, aflibercept.

[119]

When used in the context of a claim to a method of treating human patients, I consider that the phrase "a method for treating" imports, at a minimum, an expectation that the dose of the pharmaceutical compound being administered as the treatment is an amount that is safe and efficacious. A/Prof Symons considered that "there is an inherent limitation in claim 1 that the dose required to be administered must be safe and effective".
At [0030] the Specification notes that the amount of VEGF antagonist administered to the patient in each dose is, in most cases, a "therapeutically effective amount". The Specification describes the phrase "therapeutically effective amount" as meaning "a dose of VEGF antagonist that results in a detectable improvement in one or more symptoms or indicia of an angiogenic eye disorder, or a dose of a VEGF antagonist that inhibits, prevents, lessens, or delays the progression of an angiogenic eye disorder". Examples of "therapeutically effective amount" dose are provided from about 0.05mg to about 5mg.
Efficacy is discussed at [0032] of the Specification and in certain embodiments "efficacy" means a gain of one or more letters on the ETDRS visual acuity chart from the time of initial treatment.
At [0009], the Specification notes that Aflibercept in the form of EYLEATM was approved by the FDA for use in November 2011 with a recommended intravitreal dose of 2mg for the treatment of patients with nvAMD or Wet AMD. At [0043], the Specification discloses that the patients in Study 1 who received 2mg doses of VEGFT (aflibercept) (2Q8) every 4 weeks to week 8 and then every 8 weeks (with sham injection at the interim 4-week visit) in the nvAMD Phase III trial discussed in Example 4, achieved an improvement in visual acuity at 52 weeks compared to baseline.
On the limited evidence as it currently stands, I consider that there is support for the claims to a method of treatment of angiogenic eye disorders using a safe and effective amount of the VEGF antagonist, aflibercept. What constitutes a safe and effective dose of aflibercept could be determined by a person skilled in the art using only routine techniques on the basis of the information in the Specification, supplemented by their common general knowledge.

[130]

The EYLEA Aflibercept Products are presently listed on the F1 of the PBS Schedule.
As a consequence of the operation of the price reduction regimes, subject to the exercise of ministerial discretion, the listing of the Sandoz Aflibercept Products on the PBS Schedule in the Launch Scenario will result in the movement of the EYLEA Aflibercept Products to the F2, and at least a 25% price drop and price disclosure associated price reductions for all of Bayer Australia's PBS listed products, i.e., EYLEA® 2mg and 8mg vials and PFS, even though there is no 8mg Sandoz Aflibercept Product.
Mr Behrens gave an example of where the PBS price drop had been reversed where the supply of the pharmaceutical item which triggered the PBS price drop could no longer be guaranteed by the manufacturer. On 30 July 2020, the PBS price drop was taken not to have applied to the MERCAPTOPURINE-LINK® brand of mercaptopurine (an oral tablet containing 50mg of mercaptopurine monohydrate). On the same date, mercaptopurine was moved from F2 back to F1. Ms Zha's evidence was that she considered that the statutory price reduction was generally irreversible. Increases to PBS prices are very rare and Ms Zha had never seen a First New Brand (FNB) statutory price reduction be reversed absent a failure to supply a guaranteed brand , which she has only seen occur once.
Ms Zha considered that once the EYLEA Aflibercept Products move from F1 to F2, certain special agreements between the Applicants and the Government will cease. The repercussions are that the effective price, as reduced by the FNB statutory price reduction, will become publicly available and therefore referable by international reference pricing regimes. Additionally, there may be further price drops, as Sandoz may be prompted to enter pricing negotiations with the Government. Under the s 85C of the NH Act, if there are two or more listed brands of a pharmaceutical item (as will be the case for the Launch Scenario) the Minister has an obligation to ensure the Approved Ex-Manufacturer Price (AEMP) of each listed brand of a pharmaceutical item is the same, before entering into a price agreement with a responsible person in relation to a brand of that pharmaceutical item under s 85AD(1) of the NH Act. Ms Zha gave the example that if the listing of the Sandoz Aflibercept Products were to trigger a 25% price reduction, Sandoz may offer the Government a 30% price reduction. The Government would then ask Bayer Australia to offer the same 30% reduced AEMP for the matching EYLEA® pharmaceutical items.
Although the Minister has a discretion not to reduce the PBS price, or reverse the price drop, there was no evidence that the Minister had ever done so where a generic product has launched with supply. I consider it unlikely that the Minister would reinstate the F1 AEMP price of EYLEA® if the Sandoz Aflibercept Products were later removed from the market if Sandoz was ultimately restrained from selling them.
Ms Zha deposes that further price disclosure related reductions may also occur. The magnitude of these reductions depends on the market share of the biosimilar and originator products and the level of incentives provided by the biosimilar and originator suppliers. Unlike the 25% price drop, the Department of Health does not appear to have any discretion to reverse these subsequent price decreases.

[133]

The evidence before me suggests that EYLEA® 2mg and EYLEA® 8mg are distinct markets. As Ms Zha deposes, the launch of EYLEA® 8mg did not trigger a FNB statutory price drop on EYLEA® 2mg because "8mg aflibercept is not bioequivalent to 2mg aflibercept". That the two dosages consist of separate markets is also axiomatic considering the significant marketing efforts engaged by the Applicants to encourage EYLEA® 2mg consumers to move to EYLEA® 8mg.
Sandoz contends that the Applicants are cannibalising their own EYLEA® 2mg market by actively transitioning prescribing ophthalmologists and patients across to EYLEA® 8mg. Sandoz argues that by that conduct, the Applicants are destabilising the status quo.
[Redacted.]
At present, the Applicants do not have an EYLEA® 8mg product in the form of PFS in Australia. [Redacted]. Prof Daniell deposes that "pre-filled syringes are generally preferred over vials by ophthalmologists due to their relative convenience", and that he expects that "switching from Eylea 2mg to Eylea 8mg would accelerate significantly if Regeneron/Bayer enters the market with an Eylea 8mg pre-filled syringe product". [Redacted.]
A sample advertisement for the EYLEA® 8mg product spruiked the benefits of the 8mg against Bayer Australia's own EYLEA® 2mg product for Wet AMD patients. Based on the evidence before me, I consider that ophthalmologists are being encouraged to "upgrade" their Wet AMD patients currently on EYLEA® 2mg to the "superior" EYLEA® 8mg product.
Whilst it is a legitimate commercial tactic for the Applicants to seek to transition the market for their product from EYLEA® 2mg to EYLEA® 8mg in order to mitigate what they see as the impact of biosimilar entry, if Sandoz is enjoined from entering the market until late 2027 or beyond, and Bayer continues with, and potentially accelerates its transitioning process, the market will have irredeemably changed from its current composition. The EYLEA® 2mg market would be considerably smaller than it is now, and the potential share, in terms of numbers of prescriptions, of that dwindled market that Sandoz might obtain when it is finally able to enter the market will be correspondingly smaller.
On the Launch Scenario, Sandoz submits that any loss of EYLEA® 2mg and EYLEA® 8mg profit and/or market share will not be wholly attributable to the launch of the Sandoz Aflibercept Products, as some shrinking of the EYLEA® 2mg market will be due to the Applicants' own cannibalisation of its 2mg market by transitioning patients to EYLEA® 8mg.

[144]

The Applicants submit that even if the loss arising from the United States's international reference pricing is too uncertain, the position with respect to Taiwan is "far less amorphous [...] than in the USA".
Ms Zha gives evidence of Taiwan's international reference pricing system, noting that in Taiwan, EYLEA® prescriptions are reimbursed under the Taiwan's National Health Insurance (NHI) system. Ms Zha is employed by Bayer Consumer Care, and her affidavit provides scant basis for her possessing any specialised knowledge of Taiwan's pharmaceutical pricing system and the relevant legislative and regulatory schemes.
According to Ms Zha, the legislation and regulation governing the NHI system contain rules for setting prices of "pharmaceutical items" that are reimbursed under the NHI system. According to Ms Zha, on the basis of a translation of the relevant statutory provisions into English, one such rule is that prices of pharmaceutical items reimbursed under the NHI system are adjusted based on the price of those items reported in ten reference countries (A10 Reference Countries).
Ms Zha gives evidence that in Taiwan, EYLEA® is categorised as a Category III drug because it is not currently the subject of Taiwan patent protection for the API. Ms Zha says that that categorisation enlivens art 20 of the Regulations Governing the Adjustment of Drug Prices under the National Health Insurance (Tw), which purportedly stipulates that EYLEA®'s price ought to be determined as follows: in the first year EYLEA® becomes a Category III drug, 115% of the weighted average selling price of other aflibercept in the same dosage form (Group Weighted Average Price) or via the lowest EYLEA® price in the A10 Reference Countries (Lowest International Reference EYLEA Price); in the fourth year and every subsequent three years, 115% of the Group Weighted Average Price or the Lowest International Reference EYLEA Price, and; in all other years that the drug remains a Category III drug, 115% of the Group Weighted Average Price.
According to Ms Zha, the above price reductions are set to take effect in September 2026 and again in 2027 for EYLEA® 2mg, and March 2027 for EYLEA® 8mg. [Redacted.]
Even presuming that Taiwan's drug price-setting regime is as per Ms Zha's understanding, the Applicants have not provided corroborating evidence of sufficient certainty to support the Lowest Reference Price Assumption. Additionally, there is no evidence as to whether the Taiwan market currently has any biosimilar aflibercept product or whether any biosimilar aflibercept product may become available by September 2026 or thereafter. This information appears to be particularly relevant considering that EYLEA®'s pricing in Taiwan may also be determined by way of the Group Weighted Average Price.
For those reasons, and from the limited material before me, I consider that the Applicants' purported loss arising from Taiwan's international reference pricing ought to be given limited to no weight in the balance of convenience assessment.

[146]

At [153] of Sanofi, Burley J noted the prospect that a price reduction in Australia could also trigger commensurate price reductions in other countries which adopt international reference pricing. However, Burley J considered that there was insufficient evidence as to the nature, size and consequences of the relevant "International Price Referencing" for it to play a significant role in his assessment of the balance of convenience. The Applicants have sought to remedy any evidentiary deficiency of that kind by way of filing evidence from Ms Zha and Mr Gritzmann, including on information and belief from others.
However, in the case of the United States, the fate of the Executive Order and how it might be applied is currently uncertain, particularly as there are already biosimilars to aflibercept on the market. I consider that the purported position in the United States is so highly speculative that it can play no significant role in my assessment of the balance of convenience.
In relation to both the United States and Taiwan, there is no evidence as to the market circumstances in any of the other reference countries, whether there are already biosimilars on the market or not, and their applicable regulatory and pricing schemes. It is entirely unknown whether there might be circumstances (such as the launch of a biosimilar) in those countries that may result in any of them being the lowest price at the relevant time, or if not the lowest, then the second lowest to Australia, by a matter of a minute amount.
Finally, the Applicants refer to losses by Regeneron, Bayer Australia and Bayer Consumer Care. It does not appear that any of the threatened losses in the United States or Taiwan are asserted to be suffered by any of the Applicants directly. Ms Zha says she is informed of matters in her affidavit by an employee of Bayer Taiwan Company Ltd. There is no evidence, other than at a very high level, of the relevant Bayer group and Regeneron corporate structures and financial arrangements (including transfer pricing arrangements), whereby it would be possible to ascertain whether and how losses to Bayer Taiwan Company would flow to the Applicants. On the present evidence, these threatened losses appear to be suffered by companies that are sufficiently remote so as to be considered losses accruable by third-parties.

[148]

The Applicants contend that if Sandoz enters the market, there is a real risk of job losses at Bayer Australia and Bayer Consumer Care. The team responsible for the success and growth of EYLEA® in Australia is currently around thirteen staff, comprising nine full-time equivalent (FTE) employees in sales roles and six FTE employees acting as medical advisors and medical liaisons.
Mr Rock's evidence is that if Sandoz is not restrained from listing the Sandoz Aflibercept Products on the PBS, there is a real risk that at least some of those employment positions will be lost. Further, the Second and Third Applicants purport that they will need to cut marketing, commercialisation and support activities for EYLEA® (its second-best selling pharmaceutical product after Xarelto®), including education and patient support programs, research funding and partnerships with industry and patient support societies. All of this will be to the detriment of patients, ophthalmologists and researchers.
However, Ms Zha's evidence is that if the Applicants' application to restrain the launch of the Sandoz Aflibercept Products is unsuccessful, the Second and Third Applicants will likely [redacted] EYLEA® 8mg PFS was recommended for listing of treatment of Wet AMD and DME during the November 2024 PBAC meeting.
Given the idiosyncrasies of the VEGF antagonist market I discussed above, I consider that for Bayer Australia and Bayer Consumer Care to accelerate the transition of patients and ophthalmologists from the EYLEA® 2mg to the "upgraded" EYLEA® 8mg would require two things. First, it would require a substantial education program which advertises and promotes the benefits of the higher dose EYLEA®. Secondly, it would require [redacted]. In those circumstances, it is likely that the employees of the Second and Third Applicants currently conducting marketing and support activities for EYLEA® 2mg, would or could pivot to conducting similar activities focussed on EYLEA® 8mg [redacted]. I also consider it unlikely that Bayer Australia and Bayer Consumer Care would cease sponsoring research and educational events for ophthalmologists at a time they are seeking to transition ophthalmologist and patients to the EYLEA® 8mg product, of which there is currently no approved biosimilar in Australia.

[149]

Sandoz submits that if it is the first biosimilar entrant to the market, all other things being equal, the general reluctance to switch between biosimilars means that the head start afforded to Sandoz as the first entrant is far more significant than for small molecule generics. Sandoz asserts that its first mover advantage has the potential to springboard long lasting impacts on market share. Afterall, evidence suggests that once a patient is switched to a biosimilar, they are unlikely to be switched to another.
Sandoz contends that it has a significant head start, as the earliest any other aflibercept biosimilar could enter the market with PBS listing is April 2026. Sandoz submits that its strategy of educating ophthalmologists prior to a December 2025 PBS launch is critical to the securement of market share. This is due to the extensive familiarisation and education required for ophthalmologists and patients to begin using AFQLIR® or consider switching to AFQLIR®. Thus, the longer the period between Sandoz's launch and the launch of the next biosimilar, the greater the first mover advantage and market share will be retained by Sandoz.
However, if Sandoz is prevented from launching its biosimilar aflibercept, by the time it is eventually able to launch, other pharmaceutical companies are likely to also be in a position to simultaneously launch their own competitor biosimilar.
Currently, the only other approved aflibercept biosimilar in Australia is Celltrion's EYDENZELT®. Should Sandoz be enjoined, I consider it unlikely that Celltrion will launch at risk given EYDENZELT®'s narrow indication, i.e., for visual impairment due to Myopic CNV. The very small Myopic CNV market share is unlikely to justify the cost of defending any potential infringement action or launching revocation proceedings. However, according to Mr Behrens, there are a number of substantial competitors (including Samsung Bioepis, Biocon, Amgen, Formycon AG and Apotex) who already have biosimilar aflibercept products which have been approved in foreign jurisdictions, including the EU, the United States and Canada. Amgen and Celltrion currently compete with Sandoz in Australia in relation to other biosimilar products such as adalimumab, rituximab, bevacizumab and trastuzumab.
It is noteworthy that Actor Pharmaceuticals Pty Ltd has recently filed proceedings to revoke the 599 Patent, suggesting that it may have an interest in launching an aflibercept biosimilar.
Further, as I alluded to above, there is also the possibility that the Applicants might launch their own authorised biosimilar product prior to Sandoz entering the market. This would effectively deny Sandoz any first mover advantage.
The law recognises that the first mover advantage is a benefit sought to be derived at the expense of the Applicants' asserted monopoly rights: see Otsuka Pharmaceutical Co Ltd v Generic Health Pty Ltd [2012] FCA 239; (2012) 291 ALR 763 at [178] (Yates J), cited with approval in Apotex Pty Ltd v Cipla Ltd [2017] FCA 1627 at [142] (Beach J). However, I consider that where the Applicants do not have a strong prima facie case on infringement, Sandoz's loss of first mover advantage remains a relevant factor to be considered in assessing where the balance of convenience lies. This is particularly so having regard to the idiosyncrasies of the biosimilars and VEGF antagonist markets, including the higher barriers of entry and the constraints of any biosimilar in gaining market share.

[151]

Ms Zha understands that the Department of Health seeks to cost-minimise drugs which are alternative treatments for the same indications, on the basis of PBAC advice that these treatments are alternatives, even if they are not biosimilar, bioequivalent or "a"-flagged. Ms Zha's evidence is that EYLEA® has been subject to cost-minimisation price reductions previously. By way of example, price reductions that have applied to LUCENTIS® have previously flowed on to EYLEA® on a cost-minimisation basis.
In the Launch Scenario, Ms Zha expects that cost-minimisation price reductions following any reduction in the cost of EYLEA® would flow on to VABYSMO®. The Applicants submit that Roche is likely to suffer loss arising from lost sales of VABYSMO® in relation to new patients for which they cannot recover. Novartis' LUCENTIS®, which competes with EYLEA® and VABYSMO® can also be expected to suffer similar loss in sales and lost profits due to price reduction. Of course, VABYSMO® and LUCENTIS® are also likely to lose market share as a result of the Applicants' transitioning of patients from the EYLEA® 2mg to the "upgraded" EYLEA® 8mg, if ophthalmologists are convinced of the benefits of EYLEA® 8mg such as less injections per year.
If Sandoz is permitted to enter the aflibercept market, the price reductions resulting from AFQLIR®'s launch will result in substantial benefit to the Commonwealth because it will pay less for reimbursement in relation to aflibercept prescriptions. Similarly, if Sandoz is prevented from entering the market, the Commonwealth will continue to pay the higher PBS reimbursement price. Sandoz notes that the governments of the United States and Taiwan will also benefit from any price reduction, if the Applicants' submissions as to international price referencing are correct, [redacted].
In addition to the cost-benefit analyses of the Launch Scenario above, Sandoz notes that in the Restraint Scenario, [redacted].
On both scenarios there are potential benefits to some third parties and detriments to others. For that reason, I place no weight on the effects on third parties.

[154]

Sandoz submits that even if there were a prima facie case for infringement, there is no prima facie case in respect of the final injunctive relief sought, i.e., restraining the exploitation of the Sandoz Aflibercept Products. Sandoz submits that the evidence is that overwhelmingly, the Sandoz Aflibercept Products will be used in a non-infringing way. That is because the typical Wet AMD and DME treatments, will not on the face of the evidence fall within the Asserted 599 Claims. Further, the Sandoz Aflibercept Products will also be used to treat other eye disorders that are not the subject of the Asserted 599 Claims, such as Myopic CNV and RVO. As most ophthalmologists prescribe and administer aflibercept for Wet AMD and DME on a treat-and-extend basis according to their clinical judgment and patient response to the drug, most administration regimens are likely to fall outside the claimed method of treatment.
The Full Court in AstraZeneca observed at [443] that if s 117(1) of the Act is engaged in circumstances where some but not all persons to whom the product is supplied will put it to an infringing use, then "some consideration of proportionality as between the extent of the infringing use that is forecast and the scope of any injunctive relief is warranted". The Full Court continued at [444] that "[i]t may be undesirable to impose a blanket restraint upon a supplier who has reason to believe that only some consumers, perhaps a very small minority, may put the product that is or may be supplied to them to an infringing use". In the same paragraph, their Honours concluded that "all other things being equal, the more difficult it is for the patentee to establish that there is a likelihood of widespread infringing use, the more difficult it should be for the patentee to obtain injunctive relief in the broad terms restraining any supply of the relevant product".
Sandoz relies on the relative proportion of infringing and non-infringing use to contend that it would be disproportionate for the court to grant injunctive relief that restrains the supply of the Sandoz Aflibercept Product.
In light of my conclusion that there is a real prospect that the use of the Sandoz Aflibercept Products in accordance with the Sandoz PI will not infringe the Asserted 599 Claims, I accept Sandoz's submission that the broad relief sought by the Applicants is disproportionate. It is therefore not necessary for me to consider what other injunctive relief, if any, might have been appropriate in lieu of that sought by the Applicants.

[158]

On the Launch Scenario, there is likely to be a 25% price drop that would affect all the EYLEA® 2mg products. The price drop is less certain regarding EYLEA® 8mg. While there may be further price discounting, any such discount is likely to be small when compared to the usual small molecule generic price discounts. This is due to the limited prospects of a competitor biosimilar entering the market any time soon, the limited opportunity - if any - for pharmacists to substitute EYLEA® with AFQLIR®, and the lack of any apparent benefit to ophthalmologists in substituting.
There will likely be some loss of market share for EYLEA® 2mg, but that loss will be capped given that there is no significant pool of undiagnosed patients or diagnosed but untreated patients. As such, Sandoz is likely to only capture new treatment-naïve patients and any patients transferring from LUCENTIS® or VABYSMO®. Furthermore, not all market shares would be lost to the Sandoz Aflibercept Products. In addition to the Second and Third Applicants' own migration of patients from the EYLEA® 2mg to EYLEA® 8 mg [redacted], VABYSMO® is increasing its market share. Since the recent launch of the PFS VABYSMO®, the market uptake of VABYSMO® is expected to increase. In the event that the Applicants were successful at trial, I consider that they are likely to regain the market share temporarily occupied by the biosimilar Sandoz Aflibercept Products due to inter alia the established reputation of EYLEA®.
The loss suffered by the Applicants on the Launch Scenario is principally loss of revenue due to the mandatory price drop following the listing and launch of the Sandoz Aflibercept Products. As explained above, I do not consider that there will be job losses or reduced sponsorship or research as consequence of the Launch Scenario. I further consider that the alleged losses in the United States and Taiwan are too speculative and too remote to be relevant factors in the present balance of convenience exercise. As such, I consider that the Applicants' loss is likely to be compensable by a damages award.
In contrast, if Sandoz is restrained, what would have happened in the market will be highly speculative. Further, Sandoz will lose the first mover advantage. In the circumstances of this case, the first mover advantage is likely to bestow a real and lasting advantage that Sandoz would likely retain over any subsequent biosimilar competitor to enter the market. Evidence suggests that patients are very unlikely to be switched between biosimilars.
On the evidence before me, the calculation of loss in either the Restraint Scenario or Launch Scenario is of comparable complexity, such that it is not possible at this provisional stage to determine with certainty which damages calculation exercise would be more complex. However, I consider that the calculation of the loss in the Restraint Scenario is likely to involve more unknowns in the relevant counterfactual scenarios. In the event that Sandoz enters the market and is subsequently found to infringe the 599 Patent, it will be open to the Applicants to pursue damages or an account of profits, or both. The repercussions to the market caused by the launch of the Sandoz Aflibercept Products will be relatively quantifiable, notwithstanding certain difficulties outlined by the Applicants.
An additional consideration is the changing market dynamic. In the Restraint Scenario, the possibility for dramatic and irremediable change in the market from the status quo during the period of any injunction is a relevant factor. This change would be substantially due to the Applicants' likely accelerated migration of patients across from the EYLEA® 2mg to 8mg. Further contributing to the change in market dynamic is the increasing market share captured by VABYSMO®, particularly since the launch of the PFS form. The combined effect of these changes to the market is such that the market for the Sandoz Aflibercept Products (i.e., new patients for 2mg aflibercept and patients considering switching to a 2mg aflibercept biosimilar) would be very different - and likely much smaller - to the current state. In the Restraint Scenario, and as a result of the loss of its first mover advantage, Sandoz will only achieve a smaller market share in a much smaller market.
In my view the balance of convenience does not favour the grant of an injunction to restrain the launch of the Sandoz Aflibercept Products.
To my consideration of the balance of convenience, I also add my conclusion as to the weakness of the Applicants' prima facie case on infringement given my provisional construction of integers 1.3 and 1.4. Taken together, these considerations lead me to conclude that there is insufficient likelihood of success in all the circumstances to justify the grant of the broad injunction sought. I am further fortified in this conclusion by the fact that, even if the Applicants were successful at trial, they may very well not succeed in establishing widespread infringing use such as to justify an injunction. This is because of the use of the Sandoz Aflibercept Products for indications other than Wet AMD and DME, and the practice of ophthalmologists to employ treat-and-extend regimens based on their clinical judgment and patient response, both of which will take uses of the Sandoz Aflibercept Products outside the scope of the claimed method.
Accordingly, the Applicants' application for interlocutory relief is dismissed.

Regeneron Pharmaceuticals, Inc. v Sandoz Pty Ltd [2025] FCA 1067

At a glance

Court

Decision date

Before

Catchwords

Source

Catchwords

Judgment (159 paragraphs)

Parties

Legislation Cited (4)

Cases Cited (51)