Differences in botulinum toxins: Time to end the confusion

Sirs,

For over two decades, botulinum toxins (BoNTs) have proven to be a highly effective intervention, bringing significant benefit to patients with a wide range of medical and cosmetic conditions. Globally available, the most prevalent products in Europe are serotype A (BoNTA), as BOTOX^® (onabotulinumtoxinA, Allergan, Irvine, CA, USA) and Dysport^® (abobotulinumtoxinA,Ipsen, Wrexham, UK). Inevitably, comparisons between products are explored and recent letters to this journal (1,2) have unfortunately generated confusion over the clinical differences between these products. We seek to clarify that BoNT products are not interchangeable, have different physicochemical characteristics and consequently different benefit–risk product profiles, the latter being central to patient care.

The study conducted by de Boulle (3) (reviewed by Pickett and Rzany (2)) was specifically designed to replicate a clinical situation in which a physician chooses to switch satisfied patients from one BoNTA product to another. It was not designed as a randomized controlled clinical trial and the design and methodological limitations are clearly articulated in the paper's ‘Discussion’ section. However, in this sentinel study, patients effectively concluded that the products are not interchangeable.

BoNTA is a natural product of the Gram‐positive bacillus Clostridium botulinum. The process of manufacturing a biological involves many steps including fermentation, purification, stabilization, the addition of excipients and potency determination based on individual assay conditions. As a consequence there are fundamental differences between each product, which determine the differences reported in clinical practice (i.e. ‘the process is the product’) (4). These differences are fundamental, recognized by regulatory agencies worldwide and are legally required in labelling, as all BoNT SmPCs explicitly state that the units of a given BoNT formulation (including within BoNTA serotypes) are not interchangeable with the units of any other (5,6). Consistent with this regulatory position, we agree with Pickett and Rzany (2) that there is no dose‐ratio for conversion between different BoNTAs. It is therefore disappointing that the authors proceed to contradict themselves by stating ‘When a suitable dose conversion factor is employed with a highly standardized injection technique, the results for both products are equal at equal doses’.

The label warnings regarding non‐interchangeability are reflected in clinical and human experimental pharmacology experience. Clinical studies conducted in movement disorder indications such as torticollis and blepharospasm have established that Dysport is associated with a higher rate of local side effects than BOTOX (7,8). This typically manifests as undesired muscle relaxation in muscles adjacent to the site of injection resulting in adverse events such as dysphagia and ptosis. As an example, Nussgens and Roggenkamper (7) reported an incidence of ptosis of 1.4% in the BOTOX group compared with 6.8% in the Dysport group in blepharospasm patients at doses with similar duration of efficacy. More recently, Chapman et al. (8) performed a systematic review of published literature assessing the rates of dysphagia and dry mouth in BoNTA products. BOTOX was associated with a significantly lower rate of dysphagia than Dysport (mean dysphagia rate 10.5% for original BOTOX, 8.9% for current BOTOX and 26.8% for Dysport (both, p<0.05)). In cosmetic indications, differences are also seen in efficacy outcomes when comparing products at approved doses. In the only double‐blind, randomized, comparative study, Lowe et al. demonstrated statistically significant improvement in glabellar line severity in patients (moderate severity at baseline) treated with BOTOX (20 U) compared to Dysport (50 U) at 16 weeks (9).

Investigators have attempted to explain clinical differences by investigating the local diffusion of the products from the target muscles following injection. Pickett et al. (1,2) claim that the differences seen in diffusion are solely related to the use of too high a comparative dose of Dysport and criticize a study by de Almeida et al. (10) reporting differences in anhidrotic action halos. In this randomized, double‐blind comparison, a range of dose comparisons of BOTOX:Dysport (1:2.5, 1:3 and 1:4) was used since no fixed dose ratio exists. Given that the approved dose for glabellar lines is 20 U BOTOX and 50 U Dysport, it would seem reasonable that the investigators would chose 1:2.5 as the lowest comparison tested. However, at each dose of Dysport investigated, the area of anhidrosis/diffusion was greater for Dysport than for BOTOX, even with an identical injection volume.

In contrast, in an unblinded study, Hexsel et al. reported no difference in diffusion halos (11) between BOTOX and Dysport. The diameter of the anhidrotic areas in healthy volunteers were assessed with a ruler whereas de Almeida et al. calculated the area of anhidrosis using standardized imaging photography (Canfield Scientific, Inc). In addition, Hexsel injected a tiny volume (0.02ml) that is neither clinically practical nor meaningful and would tend to minimize any potential differences in diffusion characteristics.

Human clinical models of neurotoxin diffusion (12,13) illustrate that when other parameters such as product concentration (12) or injection volume (13) are controlled, Dysport demonstrates a greater radius of diffusion than BOTOX across a range of comparator doses, confirming the above clinical observations.

Pickett et al. (1,2), in their response to de Almeida's and de Boulle's papers, raise the notion of the dissociation kinetics of the botulinum toxin complex after injection. Dissociation of the neurotoxin complex must occur to allow the neurotoxin to bind to nerve terminals; however, there is no published, peer‐reviewed data indicating that this process is instantaneous upon injection. To the contrary, several authors have shown, directly and indirectly, that rapid dissociation of the complex does not occur at physiologic pH (14–16).

There should be no confusion about diffusion: these are distinct products with unique preclinical and clinical characteristics and cannot be expected to perform with the same benefit‐risk profiles if used interchangeably. The use of BoNTA, as an important therapeutic tool, should be based on a thorough understanding of the science, clinical experience, and regulatory considerations for individual products.