Merz Pharma GmbH & Co. KGaA v Allergan Inc.

It was discovered in the 1930s that stimulation of a motor neuron causes the release of a chemical called acetylcholine and that acetylcholine in turn causes skeletal muscles to contract. In this way acetylcholine was identified as a neurotransmitter. It is present in the lumen of small clear vesicles in a specialised synapse between the muscle and the neuron called the neuromuscular junction. The release of acetylcholine causes the activation of a receptor present on the surface of the muscle, which in turn allows the entry of calcium and triggers the contraction.

The various botulinum toxin serotypes have different molecular weights but they all contain a neurotoxic component which was known to be responsible for the toxic effect. This component has general and functional similarities in the case of each serotype. It has a molecular weight of approximately 150 kDa and is comprised of a long polypeptide chain of approximately 100 kDa (the H-chain) and a short polypeptide chain of approximately 50 kDa (the L-chain). The H-chain and L-chain are joined by a disulfide bridge and non-covalent forces. The H-chain is responsible for high affinity binding to the presynaptic membrane and the translocation into the cytoplasm of the L-chain. It does this by forming a pore in the membrane. The L-chain can thus pass through the membrane and into the cytoplasm.

Once the L-chain has been translocated into the cytoplasm, it interferes with the machinery that controls the release of acetylcholine at the neuromuscular junction.

The neurotoxic component of each serotype works in a slightly different way, although each has the same ultimate effect. The neurotoxic components are all zinc endopeptidases. An endopeptidase is an enzyme that cleaves a protein or polypeptide at an internal bond. The zinc acts as a catalyst in the reaction.

Release of acetylcholine from the vesicles requires the presence of three synaptic proteins called SNAREs (soluble NSF attachment receptors). So if any one of them is cleaved at any site, the release of the acetylcholine will be blocked. The three different synaptic SNAREs are cleaved by the different serotypes of botulinum toxin, each serotype cleaving at a different site. Serotypes B, D, F and G cleave VAMP (synaptobrevin), serotypes A, C and E cleave SNAP-25 and serotype C cleaves Syntaxin-1.

I did not understand there to be any dispute that all of these matters were common general knowledge at the priority date, and some of them were discovered by the work of Dr Schiavo.

As I have indicated, the neurotoxic component of botulinum toxin occurs naturally in non-covalent associationwith other proteins, the neurotoxin-associatedproteins, or NAPs.

NAPs include a non-toxic, non-haemagglutininprotein and several haemagglutinin proteins (“HAs”). Each neurotoxin component serotype is complexed to varying NAPs. The association of different NAPs with the neurotoxic component generatesa range of toxin complexes having sizes from 300 to 900 kDa. Type A, for example, has three different size complexes (referred to as LL, L and M). The molecular weight of the M complex is 300-350 kDa, that of the L complex is 450-500 kDa and that of the LL complex is about 900 kDa. Professor Dolly produced the following schematic representation of the composition and approximate relative sizes of the botulinum toxin complexes found for each of the serotypes A-F. The type G serotype was known as of 1993. However, at that time, it had not been properly characterised. Later characterisation has revealed that it consists of L and M complexes.

Professor Dolly also explained that although repeated purification of the type A crystalline toxin LL complex can give a single molecular weight composition, extensive variations arise with less purified samples due to the presence of different proportions of M, L and LL forms. Heterogeneity is also a feature of the toxin complexes obtained for most of the other serotypes.

Both Professor Dolly and Dr Schiavo agreed that NAPs were known to stabilise the neurotoxic component and prevent loss of neurotoxicity in the harsh conditions of the gastrointestinal tract.

As described in the Patent, it had been known for very many years before the priority date that botulinum toxin could be obtained in commercial quantities by establishing and growing cultures of Clostridium botulinum.

Obtaining the 150 kDa neurotoxic component from the 900 kDa complex was, however, a more problematic task. It requires a relatively difficult process which takes several days. It is not a simple purification procedure (due to the toxicity of the molecule) and involves several chromatographic separations. It was described in a paper published in 1982: Preparation and Characterisation of Homogeneous Neurotoxin Type A for Clostridial Botulism by Tse et al in European J. Biochem 122, 493-500.

As Dr Schiavo explained, the procedures for isolation of the neurotoxic component were published and generally known by 1993, but there were very few places worldwide where the procedure was actually undertaken. To Dr Schiavo’s knowledge, these included the laboratories of Drs DasGupta (University of Wisconsin), Sugiyama/Schantz (also at the University of Wisconsin), Habermann (University of Giessen), Melling (Porton Down), Sakaguchi (Public Health Research Institute, Japan), Niemann (Federal Research Centre, Tübingen) and Popoff (Institut Pasteur). As I relate hereafter, it had also been undertaken by Dr Frevert at the Battelle Institute.

Professor Bhatia explained in his evidence the various muscular disorders for which botulinum toxin has been used. The following is a summary. Cranial-cervical dystonias: postural disorders which may involve spasm in the muscles of the face, shoulders, neck, trunk and limbs. Cervical dystonia (also known as spasmodic torticollis or just torticollis):an irresistible turning movement of the head that becomes more persistent, so that eventually the head is held continually to one side, with the patient often suffering considerable pain. Blepharospasm: an involuntary tight contraction of the eyelid. Hemifacial spasm: contraction of facial muscles on the affected side. To this I would add strabismus: a squint like disorder caused by hyperactivity of one or more of the muscles controlling eye position.

Professor Bhatia and Professor Dolly explained the history of the clinical use of botulinum toxin type A. It started in the early 1970s when Dr Alan Scott, an ophthalmologist at the Smith-Kettlewell Eye Research Foundation in San Francisco, began to investigate the use of botulinum toxin type A as a non-surgical treatment for strabismus. Using animal models, he found that the injection of a small amount of the type A toxin complex corrected the condition.

The samples used by Dr Scott were from a batch (“79-11”) prepared by Dr Schantz in 1979 at the University of Wisconsin. This material had been twice crystallised and shown to meet stringent conditions of purity, stability and reproducible biological activity.

Based upon the work using animal models, the FDA granted Dr Scott permission to use this treatment for strabismus in human volunteers. In 1989 the FDA approved botulinum toxin complex type A (batch 79-11) as an orphan drug for strabismus and blepharospasm. In the same year Allergan acquired the rights to distribute Dr. Scott’s botulinum toxin type A for these ophthalmic conditions. In the early 1990s Allergan acquired substantially all the assets of Dr Scott’s company, including all his rights in the product. It then began clinical trials to assess the potential of the product for other indications. Shortly thereafter it renamed the product BOTOX.

Simultaneous development of botulinum toxin type A had been ongoing at Porton International in the UK and their product DYSPORT was licensed by the Medicines Control Agency for blepharospasm and hemifacial spasm in 1990.

Following its initial success in treating strabismus, use of botulinum toxin type A began to interest other neurologists, who viewed it as a possible form of treatment for other muscle spasm disorders. By way of example, in 1985 a team led by Dr Tsui at the University of British Columbia published the results of a pilot study on the use of botulinum complex in the treatment of torticollis: A pilot study on the use of botulinum toxin in spasmodic torticollis, Can J Neurol Sci 1985; 12: 314-6. This was one of the first reported uses of botulinum toxin type A for this type of muscle spasm disorder.

Dystonia is a brain disorder, thought to be due to the dysfunction of a part of the brain called the basal ganglia, resulting in excessive or inappropriate muscular contractions of a part of the body (focal dystonia) or more widespread (generalised dystonia). Cervical dystonia or torticollis is a focal dystonia of the neck muscles which, as I have mentioned, is characterised by involuntary posturing of the head away from its normal central position.

The majority of patients with torticollis suffer pain. Often the pain is localised to the area of muscle causing the abnormal neck posture, but there can also be pain as a result of changes caused by the torticollis to non-affected, surrounding muscles.

There is no cure for torticollis. The primary aim of symptomatic treatment with botulinum toxin is to weaken or reduce excessive or inappropriate activity of the affected muscles to improve the condition.

The resultant reduction in muscle activity leads to improvement in patients' torticollis in around 90% of cases. Improvement in the torticollis generally also leads to a reduction in pain.

Botulinum toxin rapidly became the leading treatment for torticollis. The existing alternatives of drug treatment, primarily with anticholinergics or surgery, were less attractive options for clinicians.

In 1989, Dr Alan Scott wrote a chapter entitled Clostridial Toxins as Therapeutic Agents for a book called Botulinum Toxin and Tetanus Toxin, published by Academic Press. He referred to the use of botulinum toxin for a variety of applications but noted in his conclusions that the use of the toxin needed to be refined. Among the more important issues that had to be resolved were (1) establishing the optimum dosing and injection schedule for obtaining beneficial results; (2) developing procedures to minimise adverse effects such as diffusion of toxin and paralysis of healthy muscle; and (3) finding techniques to eliminate or overcome the unwanted effects of toxin injection, such as antibody formation.

The potency of a botulinum toxin is frequently expressed in terms of units, where a unit is understood to be the amount of toxin that would kill 50% of a group of 18 to 20 female Swiss-Webster mice weighing about 20 grams each. Both BOTOX and DYSPORT were used by clinicians and there was general agreement that DYSPORT was less potent that BOTOX. It was found that clinical applications using DYSPORT required approximately 3-5 times more units of activity than those using BOTOX. This was taken into account by clinicians when switching between them.

Dry mouth and dysphagia (difficulty with swallowing) are side effects of treatment with botulinum toxin if the neck muscles are injected. This is so whether the patient is treated with BOTOX or DYSPORT.

Professor Bhatia explained in evidence, and I accept, that originally very high dosages (up to 1,000 units) of DYSPORT were administered and these gave rise to a high rate of side effects. It was thought that these side effects were caused by the toxin diffusing from the site of the injection or climbing up into the central nervous system. But by 1991 or 1992 it was found the dose could be substantially reduced, with a corresponding reduction in side effects and without loss of efficacy. By 1993 Professor Bhatia was advising his patients that the risk of side effects occurring was around 3 to 5%.

It was suggested by Allergan that DYSPORT had a greater propensity to induce such side effects than BOTOX. I am not satisfied that this was so. It was not Professor Bhatia’s experience or understanding. Further, the suggestion was not supported by any head to head clinical studies before the priority date. In so far as there was an understanding that patients treated with DYSPORT suffered a high incidence of side effects, it was thought to be attributable to the dosing regime adopted. As Professor Bhatia explained, by 1993 dysphagia was very much less of a problem. What clinicians were then worried about was the patients who were not responding to treatment – the secondary non responders.

Clinical trials and reported treatments identified a group of secondary non-responders (5-10% of those treated) who initially responded well to treatment with botulinum toxin type A but later became refractory. By the early 1990s these secondary non responders were becoming a real cause for concern.

The issue arose in patients treated for movement disorders, but not those treated for opthalmological complaints. In the former, the muscles concerned are that much larger, requiring a higher dose of toxin.

It was thought that the cause of the problem was the generation by patients of antibodies to the toxin – so neutralising its medicinal effects. As Professor Bhatia explained, treatment with botulinum toxin was in many ways a departure from normal clinical practice as it is very unusual to inject foreign proteins. It is more common in vaccinations where, for example, tetanus toxoid is actually used to vaccinate against tetanus. Therefore, the formation of neutralising antibodies (so effectively vaccinating the patient) is always a concern if foreign proteins are used as a treatment rather than a vaccination.

Professor Bhatia explained the concern was twofold. First, the clinicians simply did not know whether the percentage of those patients affected would grow. Treatment of movement disorders with botulinum toxin was still in its formative stages and clinicians were still feeling their way to a certain extent. Secondly, the results of treatment with botulinum toxin had raised levels of patient expectation and clinicians were very concerned that the formation of neutralising antibodies could lessen its effectiveness in a higher percentage of patients. If patients became refractory to treatment, the psychological damage could be large. They would be forced to live with their dystonia or return to drug treatment or undergo surgery, running the risk of life-affecting side effects.

The issue was obviously a very live one at the Madison conference. Following the conference, Dr DasGupta edited a collection of papers and reviews in a book entitled Botulinum and Tetanus Neurotoxins, published by Plenum Press in 1993. One chapter, written by Paul Greene and Stanley Fahn of the Columbia-Presbyterian Medical Center in New York, entitled Development of Antibodies to Botulinum Toxin Type A in Patients with Torticollis treated with Injections of Botulinum Toxin Type A explained that treatment of torticollis required higher doses of toxin than for blepharospasm and that at these doses patients had been recognised with antibodies at sufficiently high concentrations to block all clinical effects. The authors noted that patients developing resistance tended to have more frequent injections of higher doses but that other factors, such as concentration and volume of toxin, number of injection sites and purity of toxin might have influenced the development of resistance.

Similarly Dr Habermann wrote in the ‘Foreword’ to the same publication:

“Contrary to the previously held view, the long-term outcome of repeated injections of toxin challenges the immune system of many patients which respond with antibody protection, thus decreasing the efficacy of treatment. Immunization is inherent to the injection schedule, and so it can be well investigated. In order to achieve controlled destruction and to avoid immunization as far as possible, only highly purified, undenatured toxins should be used. Botulinum toxins should be as free as possible from immunologically cross-reacting agents and from inadvertently toxoided material. Both contaminants will contribute to immunogenicity but not to paralysis.”

And Dr Scott himself wrote in the ‘Clinical Preface’ to that publication:

“The immune response problem will be overcome by more potent preparations (less protein), and other serotypes of botulinum toxin are on the way as alternatives for the few patients (less than 40) documented to produce antibodies to botulinum toxin type A.”

Dr Scott wrote again in the ‘Foreword’ to a book called Therapy with Botulinum Toxin by Jankovic and Hallett (1994):

“What can we expect in the future? The development of circulating antibodies that prevent effectiveness of the toxin will be avoided by more potent preparations containing less antigenic toxin protein”

I accept the evidence of Professor Bhatia that, by 28 December 1993, there was a clinical impetus for a formulation which addressed the issue of preventing secondary non-response. As he explained, a number of routes forward presented themselves. In those patients with suspected secondary non-response higher doses were tried in case the non-response was because they were not receiving adequate amounts. A reduction in dosage was also considered to reduce the perceived risk of antibody formation, as well as any unwanted side effects, such as dysphagia. Formulations based on other serotypes were considered. Finally, but importantly, there was consideration of a purer toxin. From a clinical point of view, as with any injectible, the fewer foreign proteins injected into the body the better.

As I have explained, it was common ground that the neurotoxic component could be isolated and its use as a vaccine or probe was proposed by Tse in 1982. It was also common ground that the NAPs were understood to be important for stabilising the toxin complex in the gastrointestinal tract. There was, however, a substantial issue between the parties as to whether it was conventional wisdom that the isolated neurotoxic component was so unstable that it would be unsuitable for making a formulation for clinical use in treating muscle spasm.

Particular reliance was placed by Allergan on an article entitled Properties and Use of Botulinum Toxin and Other Microbial Neurotoxins in Medicine by Drs Schantz and Johnson, Microbiol. Rev., March 1992. They wrote:

“The crystalline type A toxin contains 16.2% nitrogen and, as far as is known, is composed only of biologically active amino acids (32. 207) for both the neurotoxin and the non-toxic proteins. The isoelectric point of the crystalline type A toxin is pH 5.6. Under slightly acidic conditions, pH 3.5 to 6.8, the neurotoxic component of 150,000 M_r is bound noncovalently to the non-toxic proteins in such a manner as to preserve or help stabilize the second and tertiary structures upon which toxicity is dependent. Under slightly alkaline conditions (>pH 7.1) and in the blood and tissues of animals and humans, the neurotoxin is released from the toxin complex.”

And a little later:

“The biological activity (toxicity) of the toxin, like many other biologically active proteins, is due to the spatial or conformational structure of the neurotoxin molecule (173, 182). The nontoxic proteins bound to the neurotoxin apparently play an important role in maintaining the toxic shape of the neurotoxin. Careful handling of purified toxin is therefore important for maintenance of stability. Botulinum toxin type A is readily denatured by heat at temperatures above 40ºC, particularly at alkaline pH. Solutions of the toxin lose toxicity when bubbles form at the air/liquid interface causing stretching and pulling of the neurotoxin out of its toxic shape (173). This denaturation also takes place in an atmosphere of nitrogen or carbon dioxide. Dilution to extremely low concentrations (nanograms per millilitre) also tends to decrease the stability of the neurotoxin, but this can be prevented by diluting with a buffered solution (at pH 6.8 or below) containing another protein such as gelatin and certain albumins such as bovine or human serum albumin. When the pH is raised above 7.3, the neurotoxin is liberated, which is very labile. Because of its lability the neurotoxin is not practical for medical applications.

And finally:

“Clinical use of pure neurotoxin compared with toxin complexes. Most recent information concerning the structure and pharmacology of botulinum toxin has been obtained with purified neurotoxins, but it is unlikely that these will be used in a clinical setting. The toxin complexes are much more stable than neurotoxins and can be diluted and formulated with retention of toxicity. Pure neurotoxins can be kept for several weeks to months in solution in the cold but are inactivated on dilution, formulation, and drying. No clinical trials on primates have been performed with purified neurotoxins.”

Dr Schiavo was cross examined on this paper and accepted that the neurotoxic component is (and was understood to be) more labile than the toxin complex. But he also said that there was perceived to be no difficulty in preserving the activity if the neurotoxic component was kept under the right conditions. He explained that it was well known that it had to be kept at low temperature and in a buffered solution – but this was no bar to medical applications. By 1993 it was being used by many workers in the field. Indeed, he used it in experiments in 1991 and 1992, and for two years from 1994 his co-workers used it for human treatment as reported in an article entitled: Botulinum neurotoxins: mechanism of action and therapeutic applications,Mol. Med. Today, 1996, Oct, 2 (10): 418-24.

There are other instances too. At a paper presented at the Madison conference entitled Enzyme linked immunosorbent assays (ELISAs) to detect Botulinum toxins using high titre rabbit antisera (and later included by Dr DasGupta in his publication Botulinum and Tetanus Neurotoxins) Drs Ransom and Lattuada described their work in storing low concentrations of botulinum toxins and neurotoxins (ie the neurotoxic component) for several years and continued:

“Many clinicians have cited the instability of dilute aqueous solutions of botulinum toxins to be a major obstacle to human medical use. Thus, the storage of stable neurotoxin coupled with accurate dispensing as described above may have application in the medical community. However, for human injections, the fetal bovine serum component in the diluent must be replaced with a non-immunogenic protein stabilizer such as sterile human albumin. Sterile neurotoxin can be stabilized by storing at -25ºC in a 50% glycerol diluent (concentrated about 100 times the intended use dilution). Just before use, an aliquot of the stable toxin concentrate can be withdrawn with a Hamilton micro-syringe and diluted to the appropriate concentration in sterile saline. The amount of glycerol in the final injection would be minimal (0.0005 ml or less). ”

Similarly, Dr Peter Hambleton of Porton Down described the parenteral injection of the toxic component in a review entitled Clostridium botulinum toxins J Neurol 1992; 239:16-20. But he expressed no concerns about stability.

It is also notable that the Patent itself does not suggest that any special measures are necessary to preserve or protect the neurotoxic component. Indeed the measures it describes are entirely standard for the toxin complex.

Many of these matters were put to Professor Dolly. When asked about the Hambleton paper, he accepted that there was no big problem in administering the neurotoxic component to mice. But he suggested that there might be problems in subjecting the neurotoxic component to a process of formulation and, in particular, to a process of lyophilisation. But as to that, he acknowledged that he had no experience of formulation and had no clear idea what those problems might be.

I have no doubt that in 1982 members of CAMR at Porton Down were hesitant about using the neurotoxic component. I also have no doubt that it is more difficult to handle than the toxin complex. However, I do not accept that it is so difficult to handle that it cannot be used to make a medicament for the treatment of torticollis. Indeed, if it were the Patent would be clearly insufficient. Nor do I accept that it was the general perception of researchers working with botulinum toxin in 1993 that it could not be used for that purpose.

The notional addressee is the person skilled in the art who comes to the reading of the specification with the common general knowledge of the art. He is a person with practical knowledge and experience of the kind of work in which the invention was intended to be used.

In this case, the experts were agreed that the Patent is addressed to a team of people involved in the development of botulinum toxins for the treatment of medical disorders. The team would have required expertise in microbiology, biochemistry, pharmacology, handling of toxic compounds, toxicology, clinical administration and pharmaceutical formulation. Typically the team would include clinicians (who have direct experience of clinical use) and research scientists (who formulate products based on issues identified by the clinicians). The skilled team therefore comprises, in effect, one of the established groups working in the area of botulinum toxins as at the end of 1993 together with clinicians with practical and direct experience of the clinical use of these toxins.

I must now summarise the matters of common general knowledge in December 1993 and deal with the outstanding areas of dispute.

I am satisfied that the following were common general knowledge:

Allergan suggested that the compositions of BOTOX and DYSPORT were matters of common general knowledge. Both products were, of course, well known. It was also well known that they differed in their potency and hence that they must have had different compositions. However, Allergan contended that those in the art also knew the details of their respective compositions and, in particular, that DYSPORT comprised a mixture of LL, L and M complexes and so had a significant proportion of complex (the M species) which was devoid of HA proteins. BOTOX, on the other hand, was known to be a pure product, consisting exclusively of the 900 kDa LL complex. As will be seen, this is relevant to the obviousness issue because Allergan contended that it was known that DYSPORT had fewer HA proteins, that this caused an increase in side effects and that knowledge of this would have deterred a skilled person from making a composition containing just the neurotoxic component and hence no HA proteins at all.

Allergan’s position was supported by the evidence of Professor Dolly who knew the composition of both products. But it was clear from his cross examination that he knew the composition of DYSPORT from his work at Porton Down. As for BOTOX, he explained that this was well known to comprise the crystalline 900 kDa complex purified by Dr Schantz.

Professor Dolly was able to point to only one paper describing the composition of DYSPORT, namely the 1992 paper by Hambleton to which I have referred in paragraph [71] above. This does not identify DYSPORT by name but says the product made by CAMR comprises M and L type complexes. It does not, however, give their relative proportions. Moreover, according to Professor Dolly, DYSPORT in fact contains a proportion of LL complex too.

Neither Dr Schiavo nor Professor Bhatia knew the compositions of BOTOX or DYSPORT. Dr Schiavo accepted that if he were to set about the task of producing a competitive product in 1993 he would try and find out what they were. However I am not satisfied that the composition of DYSPORT was available although I think it likely that he would have been able to find out the composition of BOTOX in the light of the extensive literature devoted to the work of Dr Schantz. In all the circumstances I do not accept that it was common general knowledge that DYSPORT comprised a significant proportion of M complex and hence significantly fewer HA proteins than BOTOX.

Professor Dolly also suggested that there was a general perception that the NAPs, and the HA proteins in particular, were necessary to prevent diffusion and so minimise side effects. There was no dispute that that HA proteins were known to bind to exposed sugar residues. Professor Dolly said in his report that the interaction of the HA proteins with the carbohydrates on the surface of the muscle cells could help to retain the complex at the site of the injection, thereby minimising diffusion. In cross examination he was unable to point to any paper where this was suggested. Further, in 2003 he was named on a paper entitled Intramuscular injection of ¹²⁵I-botulinum neurotoxin-complex versus ¹²⁵I- botulinum-free neurotoxin:time course of tissue distribution, Toxicon 42 (2003) 461-469, in which the authors reported that there was no significant difference in diffusion between the complex and the isolated neurotoxic component and that most of the neurotoxin remained at the site of the intramuscular injection regardless of the form. The authors proffered the view that this was not surprising in the light, inter alia, of the reported 1986 work of Professor Dolly that binding of the toxin to the nerve terminal membrane was mediated by the larger subunit of the neurotoxic component. Dr Schiavo believed that any binding of the HA proteins constituted only a minor component of the entire binding ability of the neurotoxin. In the light of all this evidence I do not accept that it was a matter of common general knowledge in 1993 that the HA proteins were necessary to minimise side effects by preventing or inhibiting diffusion of the toxin away from the site of the injection.

Finally, and as I have already mentioned, it was common ground that it was known that the HA proteins stabilised the toxin complex in the hostile environment of the gut. However, it had also been reported by Dr Hambleton in his 1992 paper that when administered parenterally the specific toxicity of the isolated neurotoxic component was greater than that that of toxin complexes. Overall, and for the reasons I have given, I do not accept that it was conventional wisdom that the instability of the isolated neurotoxic component made it unsuitable for formulation into a medicament.

The title of the Patent is “Use of the neurotoxic component of botulinum toxin for treating muscle–associated pain”. It is a divisional of a divisional. The Original Application (WO 95/17904) was filed on 16 December 1994 with a priority claim to 28 December 1993 and matured into EP 0 737 074 on 1 August 2001. Application 99203920.6 was divided out and filed on 23 November 1999 and matured into EP 1 005 867 on 3 December 2003. The application for the Patent was divided out and filed on 14 July 2003. It has a deemed priority date of 28 December 1993. The relationship between the various patents and patent applications is illustrated in the following diagram helpfully provided by Allergan:

The specification begins with the identification of the field of the invention. Paragraph [0001] explains that the invention provides novel uses for treating various disorders and conditions with botulinum toxins. Importantly, the invention provides medicaments useful in relieving pain related to muscle activity or contracture. It is therefore said to be of advantage in the treatment of, for example, muscle spasm, lower back pain, myofascial pain, pain related to spasticity and dystonia, sports injuries, and pain related to contractures in arthritis.

The background of the invention is set out in paragraphs [0002] to [0013]. It explains at the outset that botulinum toxins, in particular botulinum toxin type A, have been used in the treatment of a number of neuromuscular disorders and conditions involving muscular spasms, for example, torticollis. Treatment results in local paralysis and hence relaxation of the muscle afflicted by spasm.

Paragraph [0006] explains that the term botulinum toxin is a generic term embracing the family of toxins produced by the anaerobic bacterium Clostridium botulinum and that, to date, seven immunologically distinct neurotoxins had been identified. These have been given the designations A, B, C, D, E, F and G. This is of some importance in considering the scope of claim 5.

Paragraphs [0007] and [0008] explain the structure of the neurotoxic component in terms which are identical to those appearing in the Original Application. They read:

“[0007] The neurotoxic component of Botulinum toxin has a molecular weight of about 150 kilodaltons and is thought to comprise a short polypeptide chain of about 50 kD which is considered to be responsible for the toxic properties of the toxin, i.e., by interfering with the exocytosis of acetylcholine, by decreasing the frequency of acetylcholine release, and a larger polypeptide chain of about 100 kD which is believed to be necessary to enable the toxin to bind to the presynaptic membrane.

[0008] The “short” and “long” chains are linked together by means of a simple disulfide bridge. (It is noted that certain serotypes of Botulinum toxin, e.g., type E, may exist in the form of a single chain un-nicked protein, as opposed to a dichain. The single chain form is less active but may be converted to the corresponding dichain by nicking with a protease, e.g., trypsin. Both the single and the dichain are useful in the method of the present invention.)”

Paragraph [0012] records that botulinum toxin A is the toxin type generally utilised in treating neuromuscular conditions and is available commercially from Porton under the trade name DYSPORT and from Allergan under the trade name BOTOX.

Finally, paragraph [0013] explains that the objects of the invention are to provide novel treatments of neuromuscular disorders and conditions with various botulinum toxin types and to relieve pain.

The Patent then proceeds with the summary of the invention. Paragraph [0014] says:

“The present invention provides the use of the neurotoxic component of botulinum toxin for the manufacture of a medicament for the treatment of pain associated with muscle activity or contracture.”

Paragraph [0016] explains that while all the botulinum toxins appear to be zinc endopeptidases, the mechanism of action of different serotypes appears to be different. This is a recognition of the work of Dr Schiavo. Paragraph [0017] explains that it is expected to introduce the neurotoxic component directly by local injection.

The specification then proceeds to give a detailed description of the invention. As to administration, it says in paragraph [0021]:

“Preferably, the toxin is to be administered by means of intramuscular injection directly into a local area such as a spastic muscle, preferably in the region of the neuromuscular junction, although alternative types of administration (e.g., subcutaneous injection), which can deliver the toxin directly to the affected region, may be employed where appropriate. The toxin can be presented as a sterile pyrogen-free aqueous solution or dispersion and as a sterile powder for reconstitution into a sterile solution or dispersion.

There is then a description of stabilisation and formulation. Paragraph [0022] teaches that standard tonicity adjusting agents and stabilisers may be used and that the formulation may be preserved by means of a suitable pharmaceutically acceptable preservative such as paraben.

As to formulation, the specification explains that it is preferred that the toxin is formulated in unit dosage form and can, for example, be provided as a sterile solution in a vial or as a vial or sachet containing a lyophilized powder for reconstituting a suitable vehicle such as saline for injection. Importantly, in paragraph [0024] it says:

“In one embodiment, the neurotoxic component of Botulinum toxin is formulated in a solution containing saline and pasteurized human serum albumin, which stabilizes the toxin and minimizes the loss through non-specific adsorption. The solution is sterile filtered (0.2 micron filter), filled into individual vials and then vacuum-dried to give a sterile lyophilized powder. In use, the powder can be reconstituted by the addition of sterile unpreserved normal saline (sodium chloride 0.9% for injection).”

Paragraphs [0025] and [0026] detail the suggested dosing of the toxins. The potency of the toxins is expressed in units, as explained in paragraph [52] above. The dosages used in human therapeutic applications are roughly proportional to the mass of muscle being injected. Typically, the dose administered to the patient may be from about 0.01 to about 1000 units and preferably in the range of from about 80 to about 460 units per patient per treatment. These are typical doses used for DYSPORT and BOTOX.

Paragraph [0027] is also important. It reads:

“As the physicians become more familiar with the use of this product, the dose may be changed. In the Botulinum toxin type A, available from Porton, DYSPORT, 1 nanogram (ng) contains 40 units. 1 ng of the Botulinum toxin type A, available form Allergan, Inc., i.e., BOTOX^®, contains 4 units. The potency of Botulinum toxin and its long duration of action mean that doses will tend to be administered on an infrequent basis. Ultimately, however, both the quantity of toxin administered and the frequency of its administration will be at the discretion of the physician responsible for the treatment and will be commensurate with questions of safety and the effects produced by the toxin. ”

To my mind the teaching of this paragraph is clear. It is explaining that as physicians become more familiar with the use of the invention the dose may be changed. It also explains that the botulinum toxin type A, presumably also suitable for use in the invention, is available from Porton International as DYSPORT and from Allergan as BOTOX.

The specification then proceeds to describe the invention by reference to the examples. The reader is told in paragraph [0030] that appropriate areas of each patient were injected with a sterile solution containing the composition of botulinum toxin. Four examples are detailed.

Example 1 describes the use of botulinum toxin types A to G in the treatment of muscle spasms and control of pain associated with muscle spasms in temporal mandibular joint disorders. A patient was treated by the administration of 0.5 to 50 units of botulinum toxin and relief from pain occurred in about 1 to 3 days.

Examples 2 to 4 are in similar form. In each case the example describes the use of botulinum toxin in the treatment of muscle spasms and that pain was relieved or controlled.

That brings me to the claims. Two are asserted to have independent validity. Claim 1 reads:

“Use of the neurotoxic component of Botulinum toxin for the manufacture of a medicament for the treatment of pain associated with muscle activity or contracture.”

Claim 5 reads:

“Use according to any one of the preceding claims, wherein the neurotoxin component is the neurotoxin component of Botulinum type A, B, C, D, E, F or G.”

It is to be noted at the outset that the claims are drafted as “Swiss” or “medical use” claims. They are process claims to the method of manufacture of a medicament for the treatment of pain using the claimed materials. Two issues of construction arise on claim 1:

One further issue arises in relation to claim 5: Is the claim limited to the production of a medicament containing the neurotoxic component of only one serotype?

The general principles to be applied in construing a claim were explained by the House of Lords in Kirin-Amgen Inc v Hoechst Marion Roussel [2005] RPC 169. The key question is what the person skilled in the art would have understood the patentee to be using the language of the claim to mean; and that meaning must be determined from the claim in the context of the specification as a whole.

There can be no doubt that claim 1 covers the use of the isolated neurotoxic component to make a medicament for the treatment of pain. The issue between the parties is whether it is so limited or whether it also includes the use of the neurotoxic component when in a complex with its associated NAPs. As to this the claim itself is ambiguous. The wording is apt to describe the use of the neurotoxic component alone or in association with the NAPs.

Allergan submits that that, seen in context, the claim is plainly limited to the use of the neurotoxic component alone. It says that the broader construction would mean that the patentee has claimed what he has expressly described in the specification as being old, namely the use of BOTOX and DYSPORT to treat torticollis, and that can hardly have been his purpose. As Jacob J said in Beloit v Valmet [1995] RPC 705 at 720:

“There is normally no reason to suppose the patentee when he set the limits of his monopoly knew of a particular piece of prior art which is therefore irrelevant in deciding what those limits are. Of course the position is different if the prior art is specifically acknowledged in the patent. The purposive construction would lead to a construction of a claim which did not cover that acknowledged prior art: it can hardly have been the inventor's purpose to cover that which he expressly recognises was old.”

This is a powerful argument but it requires careful analysis. It is important to have careful regard to what the patentee has claimed his invention to be. The specification makes it clear that the invention provides novel uses ofbotulinum toxinsfor treating various disorders and, in particular, the provision of medicaments useful in relieving pain. This is apparent from the description of the “Field of the Invention” in paragraph [0001]. Thereafter the specification describes the background to the invention in paragraphs [0002] to [0013]. As I have explained, here the patentee acknowledges the availability of BOTOX and DYSPORT and their use to treat muscular disorders and conditions involving spasm. But there is no mention of the use of botulinum toxins to make a medicament for the relief of pain. In considering this disclosure it is significant that the claim is cast in “Swiss” form. The purpose of a such a claim is to confer novelty even if a use of the medicament is already known. I recognise, as Allergan submitted, that torticollis patients frequently suffer from pain and that the administration of botulinum toxin to such patients will inevitably relieve pain. Indeed this forms part of one of the novelty attacks on the Patent and is a matter I consider later in this judgment. However, this is not acknowledged in the specification. In the circumstances I believe the force of Allergan’s primary argument that the patentee cannot have intended to cover what he has expressly recognised to be old is much diminished.

The invention is summarised in paragraphs [0014] to [0017]. It certainly makes clear that it involves the use of the neurotoxic component. But it does not assist on the question whether it is limited to the use of the neurotoxic component alone and isolated from the NAPs.

In my judgment the rest of the description does, however, provide important guidance.

The “Field of the Invention” in paragraph [0001] refers in terms to novel uses for treating various disorders with botulinum toxins – that is to say the whole complex.

The objects of the invention are set out in paragraph [0013]. As I have already indicated, these too describe the use of novel treatments with botulinum toxin types – again the whole complex.

Turning to the “Detailed Description”, first, it explains in paragraph [0018] that the botulinum toxins used according to the invention are botulinum toxins type A to G. This can only be a reference to the toxin complexes. Second, in paragraphs [0021] to [0023] the description of the administration of the medicament of the invention is cast in terms of the toxin, by which I believe the reader would again understand the patentee to be referring to the toxin complex. Third, the section dealing with formulation in paragraphs [0022] to [0024] refers in terms both to the toxin - that is to say the complex - and, in one embodiment,the neurotoxic component. Fourth, the description of dosing in paragraphs [0024] to [0028] is clearly cast in terms of the toxin complex. I have set out paragraph [0027] in full. It explains that as the physician becomes more familiar with the use of the product - which must mean the product of the invention - the dose may be changed. It explains that the BOTOX and DYSPORT toxins are available in particular concentrations, that their potency means that doses will tend to be administered on an infrequent basis but that the dose and frequency of administration are matters of discretion for the physician. Once again, the description clearly suggests that BOTOX and DYSPORT may be used in the method of the invention. Fifth, the specification says in terms in paragraph [0029] that the invention is illustrated by reference to the examples. Each of those examples involves the use of the toxin complex.

In summary, the skilled addressee would read the specification as a whole. He would certainly understand that it describes the use of the neurotoxic component in the method of the invention. He would appreciate that DYSPORT and BOTOX have been used to treat muscle afflicted by spasm. But he would also understand that the invention is concerned with a method of making a medicament for the treatment of pain. The teaching of how that invention is to be implemented is cast almost entirely in terms of the toxin complex. Further, the examples which illustrate the invention use the toxin complex.

Finally, and importantly, there is no teaching in the specification that there is any benefit to be obtained from using the isolated neurotoxic component. On the contrary, the purpose of the invention is to provide a medicament for the treatment of pain. The specification teaches that purpose is met by the use of a neurotoxic component whether or not it is free of its associated proteins – the NAPs.

I have to say that I have not found the specification an easy document to construe. I rather think this is a reflection of its ancestry – a matter to which I return in considering the allegation of added matter. However, for all the reasons I have given in paragraphs [112] to [117] I have reached the conclusion that the skilled addressee would understand claim 1 to cover the use of the neurotoxic component whether or not it forms part of the toxin complex.

The second issue which arises on claim 1 is whether the claim excludes the presence in the final medicament of a botulinum toxin of, for example, another serotype.

Allergan accepts that claim 1 covers the use of more than one neurotoxic component to make the medicament and, moreover, that the second neurotoxic component could be derived from a different botulinum toxin serotype. However, it contends that the claim excludes the presence in the medicament of botulinum toxin of any serotype.

I am unable to accept this submission. For the reasons I have given I do not believe that claim 1 is limited to the use of a neurotoxic component free from its surrounding proteins. But whether that is so or not, it seems to me that there is nothing in the claim which excludes the use of another active component – a conclusion which I understand Allergan accepts. The claim requires the use of a neurotoxic component for the manufacture of a medicament for the treatment of pain. Provided that a neurotoxic component is used for that purpose then it matters not that another active component is used as well. There is nothing in the body of the specification which suggests that the patentee intended the claim to be limited to the use of a single active component. Once that is accepted, as Allergan appears to, then I can see no basis for suggesting that the other active component cannot be a botulinum toxin.

Turning to claim 5, Allergan submits that, in contrast to claim 1, this claim refers to “the neurotoxic component of botulinum type A, B, C, D, E, F, or G” (emphasis added) and that therefore claim 5 is limited to a medicament containing the neurotoxic component of only one serotype. If it were otherwise, claims 1 and 5 would be indistinguishable.

Once again I am unable to accept this submission. It appears to me to place more weight on the word “or” than it can bear in context. In my judgment the claim reflects the caution of the patentee in recognising that claim 1 includes the use of the neurotoxic component of any serotype whereas, at the priority date, only seven had been identified – as noted in paragraph [0006] of the specification. Claim 1 therefore covers the use of the neurotoxic component of an entirely new serotype. Claim 5, on the other hand, requires the use of a neurotoxic component of one of the seven serotypes known at the priority date. Provided that one such neurotoxic component is used in the method of the claim (whether free of surrounding proteins or not) then it is irrelevant whether or not another one is used as well.

Section 72 (1)(b) of the Act provides that a patent can be revoked if the matter disclosed in the specification extends beyond that disclosed in the application for the patent as filed. Where, as here, the patent is a divisional patent, s.72(1)(d) provides that the patent must not add matter over the earlier application. In this case the ultimate parent application, from which the Patent takes priority, is the Original Application. So much was common ground between the parties.

In Bonzel v Intervention [1991] RPC 553 Aldous J. explained that in considering whether or not there has been an extension of disclosure the task of the court is threefold:

It is the court which must carry out the comparison, but it must do so through the eyes of the skilled person. Matter is not added if the patent includes matters which the skilled addressee would take for granted on reading the earlier application.

Merz says that there are two clear examples of subject matter being added in this case:

The Original Application is concerned with an invention entitled “Botulinum toxins for treating various disorders and associated pain”. In the section entitled “Field of the Invention” it explains that the invention provides novel methods of treating various disorders and conditions with botulinum toxins.

There then follows a section entitled “Background of the Invention” which, for practical purposes, is the same as that set out in the Patent at paragraphs [0002] to [0013].

The “Summary of the Invention” begins on page 4 of the Original Application. It summarises the methods and compositions of the invention and explains that the methods comprise administering to a patient a therapeutically effective amount of botulinum toxin. It proceeds to describe the various serotypes of botulinum toxin which have been identified, their mechanism of action and that it is expected to introduce the toxins directly by local injection.

Pages 5 to 10 contain the “Detailed Description” of the invention. This refers only to botulinum toxin and not to the neurotoxic component. From pages 10 to 20 there follow 11 examples, all of which involve the use of botulinum toxin. Finally, the Original Application contains 20 claims which, once again, are cast entirely in terms of botulinum toxin.

I have dealt with this in detail earlier in this judgment. I would simply note that large parts of the body of the Original Application are repeated, but the expression “neurotoxic component” has been added into the description at various places – and it is this which has made the issue of construction a difficult one. In particular, it has been added in paragraphs [0014], [0017] and [0024]. Of course, it has also been used in the claims.

In these circumstances Merz says that there is no clear and unambiguous disclosure in the Original Application of the use of the neurotoxic component alone in the manufacture of any medicament. Allergan, on the other hand, says that although much of the specification of the Original Application is concerned with toxin complexes, there is a substantial passage on page 3, lines 5-24 dealing specifically with the neurotoxic component in the same terms as paragraphs [0007] and [0008] of the Patent, which I have set out in paragraph [89] above. Allergan says that this refers to the functions of the 50 kDa “short” and 100 kDa “long” chains and goes on to explain that the two are linked together by means of a disulphide bridge. In certain serotypes (for example type E) the two chains exist in the form of a single chain which can be converted to the more active dichain by nicking with a protease such as trypsin. Allergan places particular reliance upon the closing sentence:

“both the single and the dichain are useful in the method of the present invention”

Allergan submits that this disclosure could not be clearer and that the invention described in the Original Application is not confined to the use of the toxin complex.

I am quite unable to accept this submission. I have reached the clear conclusion that the disclosure of the Original Application is concerned with the use of the botulinum toxin and not the neurotoxic component stripped of the NAPs.

First, the title and “Field of the invention” are cast entirely in terms of the use of botulinum toxin.

Second, the “Summary of the Invention” section is limited to a description of the use of botulinum toxins, the various serotypes that have been identified, their mechanism of action and how it is expected that they should be introduced in accordance with the invention. In contrast to the Patent, there is no mention of the use of the neurotoxic component.

Third, the “Detailed Description” section makes it quite clear that the invention is concerned with the use of botulinum toxins and, in particular, toxins type A-G. It explains how the toxin types may be produced, administered and formulated and how the appropriate dose may be determined. In contrast to the Patent there is no mention of the use of the neurotoxic component.

Fourth, the examples of the invention all involve the use of botulinum toxin, as do all of the claims.

Fifth, the particular passage upon which Allergan relies is contained in the “Background of the Invention” section of the Original Application which describes what was known about botulinum toxins, their use in treatment of various disorders and conditions and their commercial availability. It is in this context that the Original Application teaches that the neurotoxic component is considered to be responsible for the toxic properties of the toxin, that it comprises the long and short polypeptide chains and how those chains are believed to operate. It is also in this context that it explains the long and short chains are linked together by means of the disulphide bridge. Further, and importantly, the particular sentence upon which Allergan relies stands in a section of the description which is placed in parentheses and which relates specifically to a description of certain types of the toxin, such as type E, which may exist in the form of a single chain un-nicked protein, as opposed to a dichain, and hence are less active, but which may be converted to the corresponding dichain by nicking with a protease. It explains that both the single and the dichain are useful in the invention. In the context of the whole description I do not believe that this constitutes an express or implicit disclosure that the neurotoxic component isolated from the NAPs can be used in the invention. On the contrary, it is simply explaining that all the serotypes of botulinum toxin can be used, irrespective of whether they contain the neurotoxic component in the more or less active form. Further, it explains that the less active form may be converted to the more active form by nicking with a protease.

In summary, I have no doubt that there is no implicit or explicit disclosure in the Original Application of the use of the neurotoxic component of botulinum toxin stripped of the NAPs.

I can take the second aspect of the added matter objection shortly. Paragraph [0024] of the Patent discloses that the neurotoxic component of botulinum toxin can be formulated in a particular solution containing saline and pasteurized human serum albumin, which stabilises the toxin and minimises loss through non-specific adsorption. It then proceeds to describe how the solution is sterile filtered, filled into individual vials and then vacuum-dried to give a sterile lyophilised powder. Further it explains that in years to come the powder can be reconstituted by the addition of sterile unpreserved normal saline.

There is no description of these matters in the Original Application. On the contrary, essentially the same passage appears but it is limited to a description of the formulation of the botulinum toxin. I do not begin to see how it could be said to be implicit or explicit in the Original Application that precisely the same method of formulation could be used for the neurotoxic component. In my judgment this also constitutes added matter.

923 (WO 94/28923)

923 is another Allergan application. It has a priority date of 10 June 1993, that is to say before that of the Patent, but was not published until December 1994. Hence it is available as a novelty citation only under s.2(3) of the Act.

Much of the specification of 923 is in identical form to that of the Original Application. It discloses, inter alia, methods for the manufacture of a medicament for the treatment of pain associated with muscle activity or contracture. In particular, it discloses the use for this purpose of therapeutically effective amounts of a combination of at least two neurotoxins selected from the group consisting of botulinum toxin types A - G. In contrast to the Patent, it is therefore specifically concerned with medicaments containing a combination of toxins.

923 contains a section entitled “Background of the Invention” which includes exactly the same description of the neurotoxic component as appears in the Original Application. Accordingly, if, contrary to my conclusion, the Original Application contains a clear and unambiguous disclosure of the use of the neurotoxic component of botulinum toxin for the manufacture of a medicament for the treatment of pain then so also does 923, subject to the qualification that it discloses the use of a combination of such components.

As I have construed claim 1 of the Patent, it extends to the use of botulinum toxins (containing as they do the neurotoxic component) for the manufacture of a medicament for the treatment of pain. Claim 1 is therefore anticipated by 923.

If, however, I am wrong as to the scope of claim 1 of the Patent, and it is limited to the use of the neurotoxic component of botulinum toxin free of any associated proteins then, in my judgment, claim 1 is not anticipated. 923 does not give clear and unmistakable directions to use such isolated neurotoxic components. It is for this reason that I have found that subject matter has been added to the specification of the Patent. If, however, I had reached the conclusion that the Original Application does disclose the use of the isolated neurotoxic component then it would have been inescapable that so too does 923. In that event claim 1 would have been anticipated by 923, even on the narrow interpretation urged on me by Allergan.

As to claim 5, this requires the use of a neurotoxin component of botulinum type A, B, C, D, E, F or G. But in my judgment it is not limited to the use of only one component. Therefore, claim 5 lacks novelty over 923 for like reasons to those I have given in relation to claim 1.

Before leaving 923 I should add that Allergan accepted for the purposes of this hearing that in the light of the decision of the Court of Appeal in Ranbaxy v Warner Lambert [2006] EWCA Civ 876 a teaching which embraces different serotype combinations of (i) neurotoxic components, (ii) complexes and (iii) components and complexes does constitute clear and unmistakable directions to use combinations of neurotoxic components. However, it reserved its position in case the matter should go further.

The use of BOTOX and DYSPORT

This issue turns on the proper construction of the claims of the Patent. For the reasons I have given I have reached the conclusion that they are not limited to the use of the neurotoxic component isolated from the toxin complex.

By the priority date, both BOTOX and DYSPORT were being used to treat torticollis. Further, it was a matter of common general knowledge that both BOTOX and DYSPORT comprised the complex of botulinum toxin type A and the active part of this complex is the neurotoxic component. Allergan accepted before me that the treatment of torticollis inevitably involved the treatment of pain associated with muscle activity or contracture. Accordingly I conclude that the manufacture, sale and use of BOTOX and DYSPORT for the treatment of torticollis and hence the treatment of pain made available the inventions of claim 1 and claim 5 of the Patent. The inventions of these claims were therefore not new.

Boghem

This adds nothing to the case of anticipation by BOTOX and DYSPORT.

The SCRIP article

This adds nothing to the case based upon BOTOX and DYSPORT unless the claims are limited to the use of the isolated neurotoxic component. The parties so approached it and I shall do the same. SCRIP is a bi-weekly newsletter providing news about recent events in the global biopharma industry. The particular article in issue was published on 9 November 1993. It describes recent developments at Porton International and the announcement by Porton International that its pharmaceutical R&D, manufacturing and commercial activities were being restructured as a single operating company, Speywood. It reports that Speywood has four pharmaceuticals on the market, one being DYSPORT. It says that the company believes that DYSPORT has potential in a wide range of indications and that dossiers have been filed for torticollis, in which substantially higher amounts of the toxin are used for each patient.

In a section headed “R&D” the article explains that Speywood will obtain future products from two sources, CAMR at Porton Down and from Speywood Medical Laboratory, based at St George’s Medical School in London. It continues:

“In collaboration with CAMR scientists, work is being carried out on a second-generation product derived from botulinum toxin type A – a highly purified form of botulinum neurotoxin type A, which may be less immunogenic than the current product, and thus more appropriate for torticollis patients. A highly purified form of botulinum neurotoxin type F has also been developed and is in preclinical studies, with a view to its potential in patients with dystonias refractory to current therapies. A pilot clinical study carried out under the auspices of the US NIH has been encouraging, the company says.”

Merz contended that the skilled person reading the SCRIP article would read it as clearly and unambiguously disclosing the use of the isolated 150 kDa component of the botulinum toxin to make a medicament for the treatment of torticollis. This submission was based on the evidence of Dr Schiavo who had that opinion for the following reasons: first, the product is said to be a second-generation product; second, the reference to the highly purifiedform would be understood to be the 150 kDa neurotoxic component; third, the article refers to the second generation product being less immunogenic than the current DYSPORT product; and finally, the word neurotoxin seems to be used to distinguish between botulinum toxin type A complex and the second generation neurotoxin which must be the neurotoxic component.

Allergan contended that the teaching was not so clear and that the skilled person would be left pondering whether Speywood might be working on the neurotoxic component alone or on a purified form of the DYSPORT complex. Professor Dolly explained that when he read SCRIP for the first time, his initial reaction was that it was referring to the development of two neurotoxic component products (type A and type F). However, on further consideration he could not be sure that his initial reaction was correct.

In the light of all the evidence I have reached the conclusion that the SCRIP article does not contain a clear and unambiguous disclosure of the use of the neurotoxic component to make a medicament for all the following reasons. First, the reference to the highly purified form is ambiguous. This could be referring to the purified 150 kDa neurotoxic component. But it is equally apt to describe an improved form of DYSPORT which has being subjected to high resolution separation techniques to separate away extraneous protein and any molecules of the complex which were in an inactive state.

Second, it seems to me that it would be entirely appropriate to describe such a purified preparation of DYSPORT as a second generation product.

Third, I accept that the skilled person would understand that it is best in any therapeutic application to have as pure a formulation as possible. This is a matter I consider further below in addressing the question of obviousness. Nevertheless, it is not clear from the article whether Speywood had simply eliminated all foreign impurities and inactive complexes or taken the further step of isolating the neurotoxin component from the toxin complex.

Fourth, the use of the word neurotoxin seems to me be ambiguous. It could be referring to the neurotoxic component. However, as Professor Dolly explained, the terminology used in connection with botulinum toxins is not consistent. It is also apparent that Dr Schiavo himself used the expression neurotoxin in relation to the botulinum toxin complex.

Fifth, both Professor Dolly and Dr Schiavo maintained their respective opinions under cross examination. However, it became apparent that their reasoning was based in part upon their own personal knowledge. In the case of Professor Dolly, his understanding was affected by his own personal knowledge of the thinking at Porton Down. Dr Schiavo, on the other hand, drew on his private knowledge that Porton Down had made and supplied to him purified type F neurotoxin component which he used for his own published work.

Finally, I should mention a dispute over the reference to the purified type F neurotoxin. Professor Dolly suggested that a simple search of the literature at that time would have revealed a letter by Ludlow et al to the New England Journal of Medicine, published in January 1992 and entitled Therapeutic Uses of Type F Botulinum toxin which shows that the type F toxin was a complex. It was not suggested that the Ludlow letter was common general knowledge and there is no cross reference to it in the SCRIP article. In my judgment it is not permissible to seek to invoke the results of a literature search when considering a novelty objection. The disclosure of SCRIP must be considered on its merits when read in the light of the common general knowledge. In any event, a literature search would also have revealed that scientists at Porton Down had isolated the neurotoxic component of the type F toxin. So the point goes nowhere.

In summary, I reject the submission that the SCRIP article clearly discloses the use of the neurotoxic component to make a medicament to treat torticollis. It does, however, quite clearly suggest the use of a purified form of botulinum toxin to treat torticollis patients and, in particular, torticollis patients who have become refractory to current therapies.

Prior disclosure and use by Dr Frevert and Professor Rohkamm

Once again the parties approached these disclosures on the basis that the claims are limited to the use of the neurotoxic component. There are two separate allegations that I have to consider. The first is that the invention was made available to the public by a letter dated 28 May 1990 from Professor Rohkamm to Dr Frevert reporting on the treatment of a patient suffering from torticollis. The second is that the invention was made available to the public through the prior use by Professor Rohkamm of the neurotoxic component supplied to him by Dr Frevert to treat that same patient.

Before dealing with each of the allegations it is convenient first to set out the background facts. Dr Frevert worked at the Battelle Institute in Frankfurt from 1983 until 1993. In the 1980s Dr Frevert purified serotype A to its neurotoxic component. It was apparent to him that the neurotoxic component could be used to treat neuromuscular conditions and he discussed this possibility with clinicians at the University Hospital in Mainz in 1986. In particular, he proposed the use of an ultra pure preparation of botulinum toxin to treat blepharospasm in a press release from the Battelle Institute dated 3 December 1986. Through his work, discussions with clinicians and press reports, clinicians were aware that he was a potential source of toxin.

Professor Rohkamm wrote to Dr Frevert on 3 January 1990 requesting botulinum A toxin to treat a patient with blepharospasm. The first supply was made on 19 February and further supplies were made on 5 March, 25 April, 23 July, 25 September and 8 or 9 October. The dispatch letters indicate that a total of 66 vials were shipped. Professor Dolly has calculated that each vial contained 0.1ng of toxin, making a total shipment of 6.6ng toxin. 0.1ng of toxin represents about 50 units of activity.

Each of the supplies was accompanied by a letter from Dr Frevert to Professor Rohkamm informing him that the samples were to be used only for research. Nevertheless Dr Frevert was fully aware when he made the supplies that Professor Rohkamm was a clinical neurologist and that he wanted to use the samples as a medicament. Dr Frevert did not prohibit him from doing so.

Dr Frevert gave evidence, which I accept, that the material he supplied was the purified neurotoxic component. For his part, Professor Rohkamm understood that he had ordered and been supplied with botulinum toxin. He did not analyse any of the material supplied, nor did he ask anyone else to analyse the material for him. He took it on trust Dr Frevert had supplied samples of Botulinum toxin which he considered suitable for treating a patient. He had no idea that Dr Frevert had in fact supplied him with samples of the neurotoxic component alone.

To the best of Professor Rohkamm’s recollection, the first patient he treated with the material supplied to him by Dr Frevert was a young female patient with very severe torticollis. Given the severity of the condition he offered to try to treat her with botulinum toxin. Professor Rohkamm had considerable concerns about administering the treatment and as to the appropriate dose that he should use. He first administered 100 units. It worked well and the patient’s symptoms were partially alleviated. Professor Rohkamm thereupon wrote to Dr Frevert by letter dated 28 May 1990 in the following terms:

“Concerning the efficacy of botulinum A toxin in another female patient having torticollis, I would like to report to you that the treatment has shown a good success. Now, the patient can turn the head to the opposite side without difficulties, since she could not be sufficiently treated for years - also with various medicaments.”

It seems that at about the same time, if not shortly before, Professor Rohkamm used some of Dr Frevert’s material to treat a patient with hemi-facial spasm and another patient with Meige syndrome (an oral facial dystonia). Relatively low doses of the toxin were used.

Given the success of the first treatment of the patient with torticollis, Professor Rohkamm administered a second treatment and applied a much larger dose of about 750 units in the hope of alleviating her symptoms further and for a longer period of time. Immediately after the treatment Professor Rohkamm went on holiday. He returned to find that his patient had been taken into intensive care suffering from botulism. She was treated for about nine days before recovering sufficiently to be transferred to a general ward.

Despite her serious reaction to the treatment, the patient requested further treatments with botulinum toxin. Professor Rohkamm was concerned about further use of the product supplied by Dr Frevert and had by that time become aware that the toxin was available from a commercial source under the name BOTOX. If he had known at the outset that the complex was commercially available then he would have used that product instead.

Finally, I would note that some 36 vials of the neurotoxic component were supplied by Dr Frevert to Professor Rohkamm after he had written the letter to Dr Frevert on 28 May 1990.

By s.2(1) of the Act an invention is to be taken as new if it does not form part of the state of the art. By s.2(2) the state of the art comprises all matter (or, as Article 54(2) of the EPC says, “everything”) made available to the public by means of a written or oral description, by use or in any other way, before the priority date.

In the present case there are two allegations: prior use and prior disclosure. They were both considered by the House of Lords in Merrell Dow v Norton [1996] RPC 76. The approach to be adopted in relation to each of them is, in substance, the same. It is important at the outset to have in mind that an invention is a piece of information and making matter available to the public within the meaning of s.2(2) therefore requires the communication of information. In contrast to the law under the Patents Act 1949, a prior secret or uninformative use by an earlier inventor does not deprive a later patent of novelty. As Lord Hoffmann explained at p.86:

“This provision [Art 54] makes it clear that to be part of the state of the art, the invention must have been made available to the public. An invention is a piece of information. Making matter available to the public within the meaning of section 2(2) therefore requires the communication of information. The use of a product makes the invention part of the state of the art only so far as that use makes available the necessary information.

The 1977 Act therefore introduced a substantial qualification into the old principle that a patent cannot be used to stop someone doing what he has done before. If the previous use was secret or uninformative, then subject to section 64, it can. Likewise, a gap has opened between the tests for infringement and anticipation. Acts done secretly or without knowledge of the relevant facts, which would amount to infringements after the grant of the patent, will not count as anticipations before.”

The key question is therefore whether the prior act or disclosure relied upon conveyed sufficient information which would have enabled anyone to work the invention.

The House of Lords gave further consideration to this question in Synthon v SmithKline Beecham [2005] UKHL 59; [2006] 1 All ER 685. Synthon contended that the invention of the patent in issue was disclosed by its own earlier patent application. The House emphasised that to make good its case Synthon had to make good two points. The first was that its application disclosed the invention the subject of the later claim – the requirement of disclosure. The second was that that an ordinary skilled man would be able to perform the disclosed invention if he attempted to do so using the disclosed matter and the common general knowledge – the requirement of enablement. These are distinct concepts, each of which has to be satisfied and each of which has its own rules.

As to disclosure, Lord Hoffmann explained at [22] and [23] that the prior art must disclose subject matter which, if performed, would necessarily result in an infringement of the later patent. In the case of a prior use, this is unlikely to be an issue – either it was or was not within the scope of the later monopoly. In the case of a disclosure, however, it is not enough that infringement is a possible or even likely consequence or performing the matter the subject of the prior disclosure. It must be necessarily entailed.

As to enablement, Lord Hoffmann explained at [26] to [33] that this means the skilled person would have been able to perform the invention which has satisfied the requirement of disclosure. An earlier use of a product will be enabling if a person skilled in the art would have been able to discover its composition or internal structure and reproduce it without undue burden.

There is one further point I would mention at this stage. A product forms part of the state of the art if it was available to the public and a person skilled in the art would have been able to reproduce it, irrespective of whether or not he would have been motivated to do so: see the decision of the Enlarged Board of Appeal of the EPO in G 01/92 Availability to the public [1993] EPOR 241. A good illustration of the principle is provided by Milliken v Walk Off Mats [1996] FSR 292. Perforated mats the subject of the patent in issue were placed in various customer locations before the priority date as part of a field test. The customers were not told that the mats were different from previous mats with which they had been supplied which were unperforated. Jacob J held that the prior use invalidated the patent because the customers could have examined or conducted tests on the mats or called in an expert for that purpose. It mattered not that there was no reason why the customers should have conducted any such tests or have been alerted to any point in conducting such tests. The rule, he said, provides a “bright line” test – avoiding subjectivity and most questions of degree (save, of course, for undue burden).

Merz submitted that the letter written to Dr Frevert by Professor Rohkamm on 28 May 1990 constituted a prior enabling disclosure of the use of the toxic component to treat torticollis, and its associated pain. It submitted that what matters is not how the invention was described but what information the description conveyed. Dr Frevert knew that the material was not the botulinum toxin complex but the isolated toxic component and so, to him, the description conveyed the information that the toxic component could be used to make a medicament for the treatment of pain.

I am unable to accept this submission. As the House of Lords made clear in Synthon the prior disclosure itself must disclose subject matter which, if performed, would necessarily result in infringement. The letter does not disclose the use of the isolated toxic component to make a medicament for the treatment of torticollis. This is hardly surprising because it was information that Professor Rohkamm did not know himself. On the contrary, the letter refers to and discloses the use of the botulinum toxin complex. The ordinary skilled person reading the letter would have assumed, as did Professor Rohkamm, that botulinum toxin could be used to make a medicament to treat torticollis, and its associated pain.

I turn then to the contention that the invention was made available to the public through the prior use by Professor Rohkamm of the neurotoxic component of botulinum toxin to treat the patient suffering from torticollis. Merz submitted that the use was plainly of the invention, that it was non secret and that the only question is whether the use was enabling.

In considering this submission it is important to have in mind the following matters. No reliance is placed upon any disclosure or supply by Dr Frevert to Professor Rohkamm. In particular, it is not suggested that Dr Frevert disclosed to Professor Rohkamm that he was supplying the isolated toxic component, nor that the medicament supplied was suitable to treat torticollis or any other condition involving pain associated with muscle activity or contracture. On the contrary, it is accepted that Dr Frevert simply indicated that the material he was supplying was the botulinum toxin complex. Further, there is no dispute that it was Professor Rohkamm’s decision to use it to treat the patient suffering from torticollis. Accordingly, it can only have been Professor Rohkamm who decided to take the final step necessary to make the invention.

In these circumstances did the use of the medicament by Professor Rohkamm make the invention of using the isolated toxic component available to the public? I believe the answer to this question is that it clearly did not.

First, Professor Rohkamm told the patient and her parents that he was proposing to treat her condition with botulinum toxin, by which he meant the toxin complex. Second, it is true that the treatment itself involved the use of a medicament containing the isolated toxic component, but this was a matter of which the patient and her parents were unaware. Third, the use of the toxic component which was made was entirely uninformative. It conveyed no information which enabled the patient or her parents to work the invention in any way at all. The medicament was administered by Professor Rohkamm. It was not open to the patient or her parents to analyse the medicament and seek to determine its contents. In practical terms they were in the same position as the patients in Merrell Dow. In that case volunteers were given terfenadine capsules by employees of the patentee for the sole purpose of swallowing them. There, as here, the use conveyed no information about how to perform the invention.

Merz contended that it was enough that Professor Rohkamm could analyse the medicament and that he would then have been free to disclose the invention to others. In my judgment this is no answer. Let me assume that Professor Rohkamm had taken the step of analysing the material and consequently knew that the medicament contained the isolated toxic component. Let me assume also that, with that knowledge, he decided to use it to treat the torticollis patient and duly did so. At that point Professor Rohkamm would unquestionably have been possessed of all the information necessary to perform the invention. Indeed, he had devised the new use the subject of the claim. But in my judgment that would not have made the invention available to the public. As Merrell Dow makes clear, anticipation requires the communication of information about the invention – and that Professor Rohkamm did not do. He neither disclosed the invention to Dr Frevert, nor did he make a use of it which conveyed any information about how to put it into practice.

The final issue which arises is whether Professor Rohkamm could have determined the composition of the medicament supplied to him by Dr Frevert without undue effort. In the light of the foregoing it is not necessary for me to make a finding on this issue. Nevertheless, I think it right to set out my conclusions, albeit briefly.

Merz contended that a straightforward analysis would have enabled a skilled person to determine the composition of the medicament supplied by Dr Frevert.

Allergan submitted this was not the correct starting point. Dr Rohkamm asked for and was supplied with product described as type A toxin. He believed that it was the toxin complex which had been used in clinical trials, as reported in the literature. In such circumstances, he might have tested for biological activity so as to confirm the presence of botulinum toxin but it would not have been obvious to analyse the structural composition of the sample so as to determine the presence or absence of non-toxic proteins. In any event, the structural composition could only have been analysed with the utmost difficulty, if at all.

In my judgment the fact that Dr Frevert described the material he supplied as type A toxin does not assist Allergan. Professor Rohkamm was free to do with the medicament what he wished. He could use it himself or pass it on to another. Anyone was free to analyse it and seek to determine its composition. As is apparent from Milliken and the decision of the Enlarged Board in G 01/92, there is no requirement that the public should have a reason for analysing a product put upon the market in order to identify its internal composition or structure. It is the fact that access to some particular information is possible which makes the latter available, whether or not there is any reason for looking for it. In Milliken the customers were not told that the mats were different from the previous mats with which they had been supplied and the Judge accepted there was no reason why any customer should have conducted any particular test or been alerted to any point in conducting a test. Nevertheless, he found that if a technologist had wanted to investigate a particular characteristic of the mats (their permeability) he would have found the characteristic of the invention, namely their perforations.

Allergan suggested that this approach is inconsistent with the decision of Laddie J in PCME v Goyen Controls [1999] FSR 801. There the Judge said at [32]:

“It seems to me that three questions need to be answered in any case where one is trying to decide what information about the composition or structure of a product is made public by reason of the public availability of the product, namely (1) what tests or analyses would it be obvious to carry out on the product, (2) what would be the result of those tests, and (3) how would those results be interpreted by the man skilled in the art.”

So, Allergan submitted, if it was not obvious to conduct any test at all then nothing about the material was made available. In my view this submission is founded on a misconception. It is clear from his judgment that Laddie J had the decisions in Milliken and G 01/02 well in mind. The issue before Laddie J was not whether it was obvious to test the prior art device but whether the tests which had been carried out were tests which would have been obvious to a skilled worker who wanted to find out what he could about that device. Clearly, if the testing itself required invention then it could not be said that the structure of the device (and hence information about the invention) was obtainable without undue effort.

Dr Schiavo contended that the samples could have been analysed using a variety of techniques available in 1990 but in the event the dispute focused on the use of ELISA. Dr Schiavo considered that it was possible to determine the composition of a single vial, containing 0.1 ng/ml of solution. Professor Dolly disagreed. He thought that the detection limit was 0.15 ng/ml. However they both agreed that it was possible to pool the contents of different vials and concentrate the gathered solution using antibody capturing beads. Professor Dolly thought that the recovery rate would be as low as 20%. Dr Schiavo was much more optimistic and suggested a recovery rate of 80%.

Professor Rohkamm received at least 36 vials of medicament after he had carried out the first successful treatment of the patient suffering from torticollis. These vials would have contained a total of 3.6 ng of the neurotoxic component. Applying Professor Dolly’s conservative recovery rate of 20% this would have produced a concentrated solution containing about 0.7 ng of neurotoxic component. This is substantially above the detection limit favoured by Professor Dolly. The picture that emerged in cross examination was rather confused. However, taken as a whole, I consider the weight of the evidence is that it would have been possible to determine the composition of such a concentrated solution using ELISA and without undue effort at the priority date.

The case of obviousness is maintained over the common general knowledge, the SCRIP article and the Frevert and Rohkamm disclosure and prior use. Once again, this only arises for consideration if, contrary to my finding, the claims are limited to the use of the neurotoxic component alone.

Common general knowledge

Dr Schiavo believed that in 1993 it was obvious to use the isolated neurotoxic component to make a medicament for the treatment of torticollis. There were a number of strands to his reasoning. First, the mechanism of action of the toxin was well understood. It was known that the active element of the complex was the 150 kDa neurotoxic component, that it comprised two chains and that they cooperated to produce the toxic effect. Second, it was obvious to purify any medicament as much as possible and in this case that meant down to the neurotoxic component. Third, there was by 1993 a concern as to secondary non-responders and this was an additional reason to remove from the medicament any unnecessary proteins that might be contributing to the patients’ immune response.

Dr Schiavo maintained this position under cross examination. As he put it, by 1993 workers knew exactly what the heart of the preparation was. Applying basic principles he would have tried to eliminate the majority of foreign substances present in the formulation.

Professor Dolly was of much the same view. He accepted that in principle it was better to use a pure material for a therapeutic application. However he maintained it was not obvious to purify down to the neurotoxic component because of concerns about stability and the possibility of serious side effects caused by the diffusion of the component away from the site of injection arising from its smaller size and the removal of the HA proteins. He also maintained that it was known that BOTOX produced fewer side effects than DYSPORT and that this was thought to be attributable to the fact that DYSPORT contained fewer HA proteins than BOTOX.

I have addressed each of the concerns raised by Professor Dolly in considering the technical background and the common general knowledge. In summary, I do not accept that there was a general perception that the instability of the neurotoxic component made it unsuitable for formulation into a medicament. I do accept that it was recognised that the neurotoxic component was more labile than the toxin complex and that this was likely to make it more difficult to use. But by 1993 it was known that the neurotoxic component could be prepared and stored, and many laboratories had used it alone. There was no general perception that it was not technically possible to use it to make a medicament. It is also relevant to note that there is nothing in the Patent to suggest that there are any stability problems with the neurotoxic component or that the Patent has provided a new and non obvious way to formulate the neurotoxic component. On the contrary, the formulation given in paragraph [0024] of the Patent is entirely standard.

As to the fear of side effects, I have rejected the suggestion that in 1993 this was generally perceived to be a reason for not using the neurotoxic component. By this time it was generally known that the active element of the complex was the neurotoxic component and how it worked. In particular it was known how the heavy and light chains cooperated together to achieve binding to and penetration of the presynaptic membrane. There was no general understanding that the HA proteins were necessary to prevent diffusion. Nor was it understood that DYSPORT produced more side effects than BOTOX or that any such difference was attributable to a difference in their compositions.

Professor Bhatia was concerned about secondary non responders. He considered that there were a number of ways forward to deal with this problem, namely (1) higher doses for patients with existing antibodies, (2) a reduction in dosage to see if this avoided the formation of antibodies in the first place, (3) formulations based on other serotypes, and (4) making a purer toxin. Each of these were, to his mind, obvious things to do. It is fair to say that in considering the last option he was thinking of a formulation that did not produce antibodies, and not specifically using the neurotoxic component alone. But I understood his evidence to be that how this was to be achieved was a matter outside his expertise and for the other part of the skilled team.

I must also take into account the submission advanced by Allergan that the obviousness case flies in the face of history. People had known of the neurotoxic component for many years and yet no one had suggested its use in clinical applications. This is a powerful point. But I have reached the conclusion that there are a number of answers to it. First, understanding of the mechanism of action of botulinum toxin increased through the 1980s and early 1990s. Second, this was accompanied by increasing familiarity with the use and handling of the neurotoxic component. Third, the clinical picture was also changing. Initially, the use of the use of the toxin was limited to the use of very small doses of the toxin to treat conditions such as strabismus and blepharospasm. But by 1993 it was being used to treat torticollis on a wide scale and the issue of secondary non responders had become a serious one. There was a general understanding that it was desirable to minimise antibody formation and remove unnecessary protein from the formulation. There was a substantial dispute between the experts as to whether the other proteins in the complex could be having an adjuvant effect. But they were all agreed as to the desirability of making the formulation as pure as possible. Finally, as Dr Schiavo said, there was, prior to 1993, a degree of inertia. There was no commercial need to provide another product.

In summary, if there is invention here it lies in overcoming misconceived fears that the neurotoxic component could not be formulated into a medicament due to its instability and that its use would increase the incidence of side effects. The Patent provides no solution to any technical problem at all. It teaches that the neurotoxic component can be formulated and administered in the same manner as the toxin. There can be invention in overcoming a technical prejudice, but it must be a prejudice which is generally held by those in the art. In the light of all the evidence I have reached the conclusion that by 1993 there was no general understanding that the instability of the neurotoxic component made it unsuitable for formulation into a medicament and there was no general understanding that the risk of side effects was such that the neurotoxic component could not be used to make a medicament to treat torticollis. In my judgment it was obvious in 1993 to use the neurotoxic component to make a medicament for the treatment of pain associated with muscle activity or contracture.

The Scrip article

The case of obviousness over the Scrip article is, if anything, slightly stronger. It suggests that to the skilled person that Speywood were working on a highly purified form of botulinum toxin but is not clear as whether this is the neurotoxic component or purer formulations of active complex. The skilled person would understand that both were possibilities. For like reasons to those I have give in relation to the case based upon the common general knowledge the claims are obvious in the light of this disclosure.

Frevert/Rohkamm

These add nothing to the allegation based upon the common general knowledge.

I understood this allegation to be presented essentially as a “squeeze”. If it required a non obvious technical step to use the neurotoxic component to make a medicament for the treatment of pain then this is not disclosed in the specification. In the event the parties took the position before me that such a step was not necessary and the formulation procedures described in the Patent were conventional and would work. Dr Schiavo’s evidence was to the effect that formulation was not a problem. Professor Dolly suggested at one point that Allergan must have found a way to formulate which they had not described. But, as I have indicated, this was not something of which he had knowledge or experience. In the circumstances I conclude the allegation must fail.

Claims 1 and 5 are invalid for added matter, lack of novelty and obviousness. The Patent must be revoked.