Introduction

Botulinum toxin is a naturally occurring protein produced by the anaerobic bacterium Clostridium botulinum. The protein acts as a neurotoxin by blocking the release of acetylcholine at the neuromuscular junction of striated muscle, thereby blocking neuromuscular motor transmission. This unique property has been found to have great clinical utility, treating such diverse medical conditions as blepharospasm, hyperhidrosis, and dynamic rhytides.

Seven serologically distinct types of botulinum toxin have been identified – designated as types A, B, C1, D, E, F, and G. Only botulinum toxin A and B are used clinically, with serotype A being the most potent. At this time, rimabotulinumtoxinB or Myobloc^® (Solstice Neurosciences, Inc; South San Francisco, CA) is the only commercially available botulinum toxin B. Initially approved by the FDA in 2000 for the treatment of cervical dystonia, its use in cosmetic applications has been limited by systemic anticholinergic adverse effects and high antigenicity. Therefore the following discussion will focus solely on botulinum toxin A (BoNT-A).

BoNT-A, like all the botulinum toxin serotypes, causes flaccid muscle paralysis by blocking acetylcholine release at neuromuscular junctions. The compound, first purified in 1946 by Dr Edward Shantz, was first found to have a cosmetic application in 1988 when Drs Jean and Alastair Carruthers observed improvement in the appearance of periorbital rhytides in patients treated for blepharospasm. In 1992, the Carruthers published their first study describing its cosmetic use in the treatment of glabellar frown lines. Subsequent randomized controlled studies provided evidence for its safety and efficacy leading to FDA approval of BoNT-A for cosmetic procedures in 2003. Today BoNT-A is considered a vital component of non-invasive facial rejuvenation and is the most common non-surgical cosmetic procedure performed.

Types of botulinum toxin A

There are multiple formulations of BoNT-A available in the US and internationally; however, the practitioner must be aware that these products are not interchangeable and important differences exist. As such, in 2009 the US FDA established unique drug names to ‘help ensure patient safety and reduce confusion’ (Table 9.1). What follows is a review of available products and their known and theoretical differences.

There are currently three formulations of BoNT-A available in the US today for cosmetic applications: onabotulinumtoxinA or Botox^® (Allergan, Inc, Irvine, California, USA), abobotulinumtoxinA or Dysport^® (Ipsen Biofarm Ltd, Berkshire, UK) and incobotulinumtoxinA or Xeomin^®® (Merz Pharmaceuticals, Frankfurt am Main, Germany). For ease of discussion, the products will be referenced by their trade names moving forward. Botox^® (Fig. 9.1) was the first to become available and continues to dominate the worldwide cosmetic market. In the USA it is approved for the treatment of glabellar lines and hyperhidrosis. It has also been shown to be highly effective for such ‘off-label’ uses as the treatment of dynamic rhytides of the forehead, periorbital area, midface, perioral area, and neck.

Figure 9.1 Vial of Botox, botulinum toxin A with the unreconstituted product seen as the small white ring of powder visible around the base of the vial. Each vial contains 100 units of botulinum exotoxin A in lyophilized form, as well as a small amount of sodium chloride (0.9 mg) and human albumin (0.5 mg).

Dysport^® (Fig. 9.2), another preparation of BoNT-A, is approved for the treatment of glabellar lines. It has become increasingly popular worldwide since its introduction and has been shown to have similar cosmetic applications to Botox^®.

Figure 9.2 Vial of Dysport^® supplied as sterile, single-use, 3 mL glass vial. Each vial contains either 300 or 500 units of lyophilized form of the botulinum toxin type.

A third product, Xeomin^® (Merz Pharmaceuticals, Frankfurt am Main, Germany) was previously indicated only for treatment of cervical dystonia and blepharospasm. However, as of July 2011, it also gained FDA approval for the treatment of glabellar rhytides. Xeomin^® has become available in the United States in the spring of 2012.

Other BoNT-A products currently in development include Purtox^® (Mentor Corporation, Santa Barbara, California, USA), Neuronox^® (Medy-Tox, Chung-cheongbuk-do, Korea), and Prosigne^® (Lanzhou, China). Purtox^® recently completed the first of three Phase III clinical trials with promising results. Clinical trial data in the United States are still pending on the other BoNT-A formulations, some of which are currently available outside the United States. In the future it will be interesting to see how the addition of other BoNT-A products to the market will affect product pricing and patient access to them.

Mechanism of action

All botulinum toxin serotypes produce their clinical effect by blocking neuromuscular transmission. Specifically, BoNT-A acts through the cleavage of SNAP-25 (synaptosomal-associated protein of 25 kDa), a protein required for the release of acetylcholine at the neuromuscular junction. The toxin anchors to motoneuronal synapses and becomes internalized via its heavy chain. The light chain of the toxin then catalyzes the cleavage of SNAP-25 with the resultant inhibition of acetylcholine release. All BoNT-A products act through this mechanism.

Compositional differences

The botulinum neurotoxin molecule is a 150 kDa single-chain polypeptide consisting of a 100 kDa heavy chain and a 50 kDa light chain. These chains are held together by a disulfide bond and are associated with a zinc atom (Fig. 9.3). In its native state the 150 kDa toxin is found complexed with large protective proteins. These complexing proteins, which are primarily hemagluttinins, act to protect ingested toxin from degradation within the gastrointestinal tract. In neutral to basic pH environments the BoNT-A complex dissociates into the free 150 kDa neurotoxin and high-molecular-weight hemagglutinin components.

Figure 9.3 The structure of the 150 kDa botulinum toxin A: it is composed of a 100 kDa heavy chain (responsible for the high-affinity, irreversible, docking on the presynaptic nerve membrane) on the left and a 50 kDa light chain (responsible for the cleavage of proteins in the synaptic fusion complex) on the right. The heavy and light chains are held together by a disulfide bond.

The complexing proteins of the native molecule seem to have evolved as a clever mechanism to protect the toxin when ingested by the host. However, the role that these proteins play, if any, in cutaneous injection of BoNT-A is unclear and remains a subject of debate. It has been argued that the proteins help to stabilize the core toxin or that they may help to enhance activity; however, these hypotheses have not been substantiated.

The three available formulations of BoNT-A vary in the presence, absence, and amount of complexing proteins. Botox^® is the largest of the available preparations at 900 kDA (Fig. 9.4) and is associated with the most complexing proteins. Due to the method employed in its production, Dysport^® may be associated with a variable number of proteins with a molecular weight ranging from 500 to 900 kDa. Uniquely in the group, Xeomin^® is composed only of the 150 kDa neurotoxin, free of any complexing proteins. Purtox^® will also be a non-complexed formulation of botulinum toxin A.

Figure 9.4 The structure of the large 900 kDa botulinum toxin A: note that the actual botulinum toxin itself makes up only 150 kDa of the structure; the rest is made up of large protective hemagglutinin (HA) and non-toxic non-hemagglutinin (NTNH) proteins. The complex is held together by non-covalent bonds that are stable in the acidic environment of the gastrointestinal tract, but dissociate as the pH increases in the blood stream and releases the free toxin. (LC = light chain; HC = heavy chain; SS = disulfide bond; Zn = zinc.)

There is good evidence to support the concept that the complexing proteins in Botox^® and Dysport^® may dissociate when the vial is reconstituted with saline. A recent study by Eisele et al (funded by Merz GmbH) showed that, after reconstitution by manufacturer standards, over 85% of the Botox^® was present in the free state without associated hemagluttinins. The remainder of the neurotoxin was associated with complexes in the 500 kDa range as opposed to its original 900 kDa complex. In this study, Dysport^® was found to be only in the free neurotoxin state. Furthermore, the complexing proteins dissociate readily after being injected into the higher pH of the skin. In this study, the proteins are estimated to dissociate in less than 60 seconds. Therefore, the small number of complexing proteins remaining after reconstitution seem to dissociate rapidly upon injection into the skin. This argues against the complexing proteins having any effect on the spread or diffusion of the toxin. The presence of complexing proteins may, however, contribute to BoNT-A resistance, as discussed later.

Dosing

It is imperative that the practitioner be knowledgeable regarding the differences in dosing for the available BoNT-A formulations. Unfortunately, there is currently no accepted ‘conversion factor’ between products, and this remains an area of controversy and debate. To appreciate product-dosing differences, an understanding of the definition of unit potency is helpful.

Most drugs are measured and administered by their weight, typically in the milligram range. However, the amount of BoNT-A used is so minute that it must be measured in terms of its biologic activity rather than by its mass. The dose of BoNT-A, usually in the nanogram range (10⁻⁹), is measured in units of biologic activity in a standardized mouse model. One unit of biological activity, commonly referred to as 1 unit, is the amount of toxin required to kill 50% (lethal dose or LD₅₀) of 18–20 g female Swiss-Webster mice when injected intraperitoneally. For comparison, the estimated lethal dose (LD₅₀) in a 100 kg human is approximately 3500 times this, or 3500 units.

For the unit defined by the mouse model, 1 unit of Botox^® will have similar potency to 1 unit of Dysport^® and to 1 unit of Xeomin^®. However, there is significant interspecies variation and it has been shown that this same one-to-one relationship does not hold true for Dysport^® when compared with Botox^® and Xeomin^® in other species. Further muddying the waters is the way in which early studies measuring the LD₅₀ of the different formulations were designed. The diluents used in assays measuring the LD₅₀ of the products were not identical. Dysport^® toxin was diluted by a phosphate buffer containing gelatin that stabilizes low-concentration toxins, whereas saline was used in the Botox^® assays leading to apparently reduced potency. Though experience with Xeomin^® remains limited, clinical studies have demonstrated a dose ratio of roughly 1 : 1 when compared with Botox^®.

Clinical studies have failed to yield a clear consensus concerning the precise dose ratio between the BoNT-A preparations as standardization of clinical studies has been lacking. One review by Karsai & Raulin comparing Botox^® and Dysport^® showed that a dose ratio of less than 1 : 3 is probably most suitable for the two products. More specifically, others have proposed a dose ratio of 1 : 2.5, whereas the largest retrospective review on the topic by Hevia established a rather cumbersome dose ratio of 1 : 2.67. In our practice, we believe a dosing ratio of 1 : 2.5 between Botox^® and Dysport^® is accurate.

The conversion ratio between Botox^® and Xeomin^® is also not well established. However, in two large comparative trials by Benecke and colleagues using a conversion factor of 1 : 1, non-inferiority of Xeomin^® to Botox^® was shown, establishing the likelihood of similar dosing between the two products. Although the precise dosing of the BoNT-A preparations is by no means agreed upon, in practice the dosing ratio between Botox^® : Dysport^® : Xeomin^® is likely to be around 1 : 2.5 : 1.

Efficacy

Trials on efficacy are somewhat difficult to interpret given the lack of standardized dosing between the products as discussed above. Assuming appropriate and equivalent dosing ratios, there is no evidence to suggest greater efficacy of any of the BoNT-A preparations. All have been shown to be efficacious in treating glabellar rhytides, garnering their FDA indication for this treatment. They are also used ‘off-label’ for other indications with comparable results.

Duration of action

The precise duration of action of BoNT-A is difficult to assess and is dependent on a host of factors including the area treated, patient characteristics, and dosing. The currently available preparations appear to have a lasting effect of 3–4 months when employed for aesthetic use. In general, there is a trend towards a longer duration of action when higher dosages are used, up to a plateau point. Furthermore, when skeletal muscle is repeatedly treated with BoNT-A, some degree of atrophy may be demonstrated. This may result in future treatments that require less units of toxin as well as treatments that may occur at increased time intervals. It is these authors’ observation that patients retain some improvement of their dynamic rhytides for more than a year even without repeat treatment. Patients returning 1–2 years following a single BoNT-A injection often have less apparent rhytides by photographic evaluation when compared with before their initial treatment. One explanation for this is that, because there is less collagen at the depth of an active rhytide, allowing the area to rest for a period of time may give the body the opportunity to deposit more collagen in the depth of the rhytide.

The existing comparative trials on duration of action between Botox^® and Dysport^® vary across a wide range of doses making interpretation difficult; however, the preponderance of data suggest that the two have similar durations of action. A recent study by Jost and colleagues comparing the effects of treatment with Xeomin^® and with Botox^® on muscle activity of the hand showed both products to have a similar duration of effect. Until further studies are performed it is reasonable to expect that the different BoNT-A products will not differ significantly in their duration of action.

Diffusion / field of effect

Initially it was hypothesized that variations in the number of complexing proteins found in the BoNT-A complex would alter the degree of diffusion of a given product. Formulations containing a greater amount of complexing proteins with resultant higher molecular weights were speculated to have decreased areas of diffusion due to particle size. Additionally, a pilot study by Trindade and co-workers comparing the degree of diffusion of Dysport^® and Botox^® on forehead hyperhidrosis showed that Dysport^®, with a lower molecular weight, had a greater area of diffusion as evidenced by the size of the anhidrotic halo produced. The clinical significance of this finding was less clear, however, as there was no relationship between anhidrosis and inhibition of frontalis muscle contraction.

Newer studies have shown the diffusion between the products to be similar and have cast doubt on the role of complexing proteins in the diffusion of toxin. In a mouse model using a highly sensitive assay to measure diffusion, Botox^®, Dysport^®, and Xeomin^® all had similar degrees of diffusion. Since most toxin exists in its unbound form after reconstitution, it is logical that complexing proteins would have little effect on area of diffusion. Nevertheless, until further definitive studies are performed one should be mindful of the possible diffusion differences between formulations.

One factor that has been shown to affect the diffusion of BoNT-A is the dilution used. BoNT-A is dispensed in a crystallized form that must be reconstituted with saline before use. The amount of saline used to reconstitute the product, which may vary by physician preference, will affect its diffusion. A study by Hsu et al comparing Botox^® mixtures reconstituted with 1 mL and 5 mL of saline found a 50% greater distance of diffusion in the more dilute mixture. This property may be used to the physician’s advantage to create larger field effects. Conversely, for the uneducated practitioner, more extensive diffusion may potentially result in unintended consequences such as lid ptosis.

Safety

The safety of BoNT-A in cosmetic use has been well established in the literature and the increasing popularity of such procedures is also a testament to this. Both short- and long-term studies exist supporting the favorable safety profiles of Botox^® and Dysport^® (reviewed by De Baulle et al). Although Xeomin^® is a newer BoNT-A with fewer clinical studies than the others, it was shown by Wabbels and colleagues to have a similar safety profile as Botox^®.

When adverse events related to injection do occur they are usually transient and mild in severity and most often arise when the injector is inexperienced or has poor technique. The two most common adverse events encountered with BoNT-A injection are headache and eyelid ptosis, both being more common in those being injected for the first time. The long-term safety of complexing proteins has not been studied specifically, but is unlikely to have any clinical significance given the overall safety of BoNT-A. Data on comparative immunogenicity of the products are also not available. To help maintain patient safety and optimize outcomes it is vital for the practitioner to possess a comprehensive understanding of the physiology and the properties of each BoNT-A formulation.

Resistance

Therapeutic failure of BoNT-A in cosmetic use is very uncommon and has been reported in only a few instances. It is believed that the development of resistance to treatment is due to the production of antibodies directed against both the heavy and light chain components of the neurotoxin. The precise cause of antibody development is rather complex and remains to be elucidated. However, as with other immunologic events, the production of antibodies depends on the immune system of the host as well as the immunogenicity of the stimulus. The external factors that appear to be most associated with the development of blocking antibodies are the dose per session and the interval between doses. The cumulative dose does not seem to be as important. Patients treated with high doses (300 units of Botox^® or higher) at frequent intervals, as in the neurological setting, seem the most likely to develop antibodies. Cosmetic patients treated with Botox^® typically utilize 25–75 units at intervals of several months and are extremely unlikely candidates for developing resistance.

The older, functional Botox^® literature reports blocking antibodies in approximately 5% of patients treated for cervical dystonia. However, these patients were treated with the original lot of Botox^®, which had a much higher protein content (25 ng/vial) than does the present lot (5 ng/vial). Whereas reports of antibody formation with BoNT-A are more common in neurological patients, the significantly lower dosages and subsequently smaller protein loads used in cosmetics have made this scenario exceedingly rare.

Although few may develop these blocking antibodies, patient non-response to treatment with BoNT-A occurs even less frequently. This is because patients who develop antibodies may still have a clinical response to BoNT-A. According to Allergan, only 1–2% of patients will develop antibodies to Botox^® and the correlation of these antibodies to clinical response is not clear (see the 2010 analysis by Naumann and colleagues).

Therapeutic failure due to autoantibodies has been reported for both Dysport^® and Botox^® in cosmetic use. There are currently no reports of antibody-induced failure due to Xeomin^®. Whether this is secondary to the lack of experience with the product or the lack of complexing proteins remains to be seen.

For patients who are resistant to chemodenervation with one serotype of BoNT, a trial with a different serotype is indicated. For example, a patient that has developed a resistance to one of the BoNT-A preparations should have a trial with the BoNT-B preparation. In addition, an attempt should be made to measure antibodies to determine whether the resistance is antibody mediated. The newer BoNT-A products should be further evaluated for antibody formation rates as their protein load is different.

Storage

Vials of Botox^® had traditionally been kept frozen (−5^oC) until used. However, new evidence suggests that there is no significant change in the shelf life or efficacy of Botox^® if it is stored in a refrigerator rather than a freezer. The package insert now recommends that the vials be stored at 2–8^oC. Similarly, vials of Dysport^® may be kept refrigerated as opposed to frozen prior to use. One potential advantage of Xeomin^® is that it can be stored at room temperature for up to 36 months. It is noteworthy that all of the commercially available neurotoxins bear labels indicating that they should be discarded if the contents are not used within 4 hours of reconstitution / broaching the vial. This is in accordance with FDA regulations; however, studies by Trindade et al have shown there to be no decrease in product efficacy when stored for weeks to months in refrigeration.

Conclusion

As BoNT-A injections have become the most common non-invasive cosmetic procedure performed in this country, it is vital to appreciate the differences between the available BoNT-A products used today. Understanding the compositional differences of the products allows for more detailed and precise injection techniques with limited side effects. Fortunately, the currently available products appear to have similar efficacy, safety profiles and duration of action, but the role of complexing proteins still merits further investigation. Moving forward, it will be important to stay familiar with evolving toxin products.