Petition to the FDA requesting regulatory action concerning the possible spread of botulinum toxin (Botox, Myobloc) from the site of injection to other parts of the body (HRG Publication #1834)
January 23, 2008
Andrew von Eschenbach, MD
Dear Dr. von Eschenbach:
Public Citizen, representing 100,000 consumers nationwide, hereby petitions the U.S. Food and Drug Administration (FDA), pursuant to the Federal Food, Drug, and Cosmetic Act 21 U.S.C. Section 355(e)(3) and 21 C.F.R. 10.30, to immediately require Allergan and Solstice Neurosciences to issue a warning letter to physicians regarding all formulations of botulinum toxin (Botox and Myobloc, respectively). This letter would alert physicians to serious problems, including hospitalizations and deaths, resulting from the spread of the toxin from the site of injection to other parts of the body. In order to alert both physicians and patients, we also request additional warnings in the form of a black box in the product label and a MedGuide for patients, to be dispensed by doctors at the time the drug is injected. These significantly improved warnings to doctors and patients would increase the likelihood of earlier medical intervention when symptoms of adverse reactions to botulinum toxin first appear and could prevent more serious complications, including death.
The European Union (EU) has posted a series of warnings concerning botulinum toxin on its web site, the latest in March 2007, alerting physicians in its 27 member states about the need to monitor for signs of botulinum toxin adverse events. The U.K. and Germany amplified the EU warning with “Dear Doctor Letters,” but no similar official warnings have been forthcoming from the FDA. The spread of toxin has been implicated in serious adverse events including muscle weakness, dysphagia (difficulty swallowing), and aspiration pneumonia, the latter sometimes resulting in death. Public Citizen has done its own analysis of the FDA Adverse Event database (AERS) for Botox and Myobloc (excluding foreign reports) and found 180 adverse event cases submitted by drug manufacturers relating to these conditions, including 16 deaths. Four of these deaths occurred in children less than 18 years of age.
Description, indications, and mechanism of action
Botulinum toxins are proteins produced by a bacterium, Clostridium botulinum. Botulinum toxin acts by blocking the transmission of nerve impulses to muscles, causing those muscles to relax and resulting in a loss of muscle control. In the case of food poisoning from botulinum toxin, in which the toxin spreads widely around the body, early symptoms include dry mouth, difficulty swallowing, slurred speech, drooping eyelids, and muscle weakness. Subsequent paralysis of respiratory muscles can lead to death.
In the case of injected therapeutic or cosmetic use of botulinum toxin, if the product spreads from the injection site to another area of the body, this loss of muscle control can be similarly harmful. For example, when muscle control to the esophagus is lost, one loses the ability to control swallowing; food and drink can then accidentally reflux and be aspirated into the respiratory tract and lungs, causing a serious complication, aspiration pneumonia, and occasionally lead to death.
Botulinum toxins are classified as distinct serotypes, A-G, with different potencies but the same mechanism of action. Types A and B have been developed commercially for use as drugs. In the U.S., there are two approved products: Botox (Allergan), which is Type A, and Myobloc (Solstice Neurosciences), which is Type B.
The two broad indications for the use of botulinum toxins are therapeutic and cosmetic. FDA-approved therapeutic uses include cervical dystonia (contractions of the neck and/or shoulder muscles that cannot be controlled), strabismus (crossed eyes), blepharospasm (spasmodic blinking of the eyes), and primary axillary hyperhydrosis (excessive underarm sweating; Botox only). Botox has one approved cosmetic use (Myobloc has none) and that is for temporary improvement of glabellar lines (wrinkles between the eyebrows). Most cosmetic use of botulinum toxin is unapproved by the FDA and is therefore considered off-label.
Chronology of European cautions
The European Medicines Agency (EMEA) is the European counterpart of the U.S. FDA. Under its jurisdiction, the Committee for Medicinal Products for Human Use monitors adverse event reports for all forms of botulinum toxin and has been posting information about adverse effects on its web site for over two years. Highlights include the following:
On August 9, 2007, the Danish Medicines Agency, in cooperation with the European Pharmacovigilance Working Party, published an analysis of adverse reactions from the four products marketed in the EU stating that, “Since the entry of the product [botulinum toxin] on the market, more than 600 reports of this type [adverse events related to botulinum toxin] have been registered [in the EU]. Approx. half of all serious adverse reactions were caused by spread of toxin.” (Italics added) This warning mentioned the dangers of “muscular weakness and difficulties in swallowing and breathing.” The Danish agency also pointed out that these adverse events are “substantially under-reported” and have been seen in connection with cosmetic as well as therapeutic uses.
The Danish Agency stated further that muscle weakness may be long-lasting, especially for those at increased risk of this type of adverse event: children, weakened elderly patients, and those suffering from serious neurological disorders. In addition, it noted that the risk of adverse events is increased if the dose is too high or the recommended injection technique is not followed. In the cases where use is off-label, there is no approved dose or technique.
Information for physicians
The Summary of Product Characteristics (SPC) is the EU counterpart to the U.S. label. The EU label for the therapeutic uses of Neurobloc has a “Special warnings and precautions for use” section that succinctly addresses all the major issues related to migration of injected drug. The latest SPC, in its “Special warnings and precautions for use” section located near the beginning of the document, warns physicians:
Unlike the EU warnings, the U.S. information is scattered throughout the labels (typically under “Warnings” and the still less-prominent “Adverse Reactions” sections). Most of the information in the EU label is present, but the Myobloc label makes no mention that the mechanism of toxicity is through distant spread. Some labels do not make clear enough that the phenomenon of distant spread is not restricted to patients treated for cervical dystonia. Neither the EU nor the U.S. has the information in a black box.
Information for patients
“Package Leaflet: Information for the User” is a document, included in the package of each drug that a patient receives, which provides comprehensive information on the prescribed drug. This package leaflet has two bolded sections, “Possible Side Effects” and “How You Will Be Given Neurobloc,” that alert patients to possible serious adverse events and what actions they should take. The following quotations appear in these sections in the Neurobloc (therapeutic) label:
Unlike in Europe, FDA-approved information is not consistently provided to patients. Some labels do have sections of the physician labeling devoted to information that is supposed to be imparted to patients. For botulinum toxin-containing products, the “Information for Patients” section is very brief, and does not approach the five-page EU patient information in comprehensiveness. In the Botox cosmetic label, literally the only information for patients is “Patients or caregivers should be advised to seek immediate medical attention if swallowing, speech or respiratory disorders arise.” In the Botox therapeutic label, there is, in addition to the above sentence, a warning for cervical dystonia patients about the risk of dysphagia leading to aspiration, shortness of breath, and pneumonia. There is no warning about distant spread of toxin, or that treatment of conditions other than cervical dystonia can also lead to these adverse events. Myobloc has no “Information for Patients” section.
Even if this section of the labeling were more complete, there is no evidence that physicians actually discuss this information with their patients consistently. The FDA does have the authority to require that detailed written information in the form of FDA-approved Medication Guides be routinely given to patients. Unfortunately, this has been required for no more than approximately 100 of the thousands of drugs approved by the FDA. Medication Guides are usually dispensed in the pharmacy when patients get their prescription filled, but there are at least four examples in which the Medication Guide is provided in the doctor’s office when the drug is injected.
Pre-approval evidence of dysphagia
Pre-approval clinical studies provide ample proof that botulinum toxin can cause dysphagia. In the analysis below, we have generally restricted ourselves to randomized, controlled trials (RCTs) with placebo controls.
Significant rates of dysphagia were documented in the Medical Officer’s clinical reviews of both Myobloc (October 1999) and Botox therapeutic (November 1999). In most studies, patients had previously been exposed to botulinum toxin. Since those with previous adverse events would presumably be less likely to enroll in subsequent trials, the incidence of dysphagia may be underestimated.
Botox: There was only one RCT for Botox therapeutic that was placebo-controlled (Study 140). The rates of dysphagia were 7% vs. 4% (treated vs. placebo). Severity grades were only provided for the six cases in the treated group: three moderate and three mild cases of dysphagia.
Myobloc: The three RCTs for Myobloc consisted of one Phase II range-finding study (Study 009) and two Phase III trials (Studies 301 and 302). The only difference between Studies 301 and 302 was that subjects in the former had cervical dystonia symptoms that were still responsive to Botox (Type A toxin), while those in Study 302 had ceased to respond to Botox. (Myobloc is Type B.) The data below summarize the incidence of dysphagia in all placebo-controlled trials of this product, based on only one dose. (Doses of Botox and Myobloc cannot be compared directly as they are different botulinum toxins.)
A summary of the severity of dysphagia for all Myobloc-exposed patients, both those in placebo-controlled and open-label studies, is shown below.
Although these trials were small and consisted only of a single dose, the results have a consistency that lends them added weight. The Medical Officer noted that dysphagia was consistently the second most common adverse event after dry mouth and, “...the most common important adverse effect of BOTOX treatments for cervical dystonia reported in the medical literature.” While not having as high an incidence of serious cases as dry mouth, serious dysphagia cases “can be more medically risky” and “even moderate grade events in these two categories have been responsible for patients discontinuing repetitive injections...”
The Botox cosmetic reviews found the incidence of muscle weakness in treated patients in two placebo-controlled clinical trials to be 1% and 3%. There were no cases of muscle weakness in the placebo groups and no cases of dysphagia in either the treated or placebo groups.
Thus, dysphagia was a common adverse event in the clinical trials conducted prior to approval and occurred in a dose-related fashion. Although most cases were mild, some severe cases did occur.
FDA analysis of adverse events
FDA subsequently highlighted the dangers of botulinum toxin in a published 2005 analysis of adverse events covering the period from 1989 to May 2003. These adverse event data add to the clinical trial information because severe cases of dysphagia were not common in the clinical trials and the adverse event data would tend to include more severe cases of dysphagia. The adverse event data also emphasize that, although dysphagia was more common in therapeutic than cosmetic clinical trials (presumably due to higher doses and, for many indications, greater proximity to the esophagus), cosmetic cases have been reported and some have been serious.
Therapeutic use: The FDA analysis found 406 adverse event reports related to therapeutic use, 217 of which met the FDA’s definition of serious. There were 26 reports of serious adverse events involving dysphagia (including one death due to aspiration pneumonia) and 13 reports of non-serious dysphagia.
Cosmetic use: There were 36 serious reports related to cosmetic use, including two reports of dysphagia, but no deaths. There were, in addition, 995 non-serious reports for cosmetic use, the most frequent of which was “lack of intended effect” (63%). Four of the non-serious reports were for dysphagia. The agency also noted that, “Numerous departures from FDA-approved recommendations for drug use, dilution, handling, site of injection, and storage were noted in these AE [adverse event] reports.”
Public Citizen analysis of adverse events
We have done our own analysis using the FDA adverse event database for cases submitted to the agency by drug manufacturers between November 1, 1997 and December 31, 2006. We searched the database using the terms “*botox*”, “*botulinum*” and “*myobloc*” as Primary Suspect drugs and did not restrict the search by Indication for drug use. We excluded cases marked “foreign.” (The EU analysis included all international reports.) The initial search included all Preferred Terms (the medical description of the adverse event), but in the principle analyses we restricted the Preferred Terms to “*dysphagia*”, “*pneumon*” and “*aspir*”. The latter categories were combined in the data presentation below. The search included Outcomes, but we provide specific information only about deaths and hospitalizations. Cases that were associated with both death and hospitalization were coded as deaths. Those that included the Preferred Term “*suicide*” were excluded from the list of deaths. We looked separately at the adverse events associated with Botox cosmetic use and the three Preferred Terms by searching on the Indication *skin* (which yielded “skin wrinkling” and “skin cosmetic procedure”).
We found 658 cases of adverse events for all Preferred Terms, of which 180 (27%) were associated with aspiration, dysphagia, or pneumonia. Of these 180, 106 had an indication listed: 18 cosmetic only and 87 non-cosmetic only. Eighty-seven of the cases associated with these three Preferred Terms were hospitalized and 16 died (including 4 children less than 18 years of age). Table 1 presents the cases broken down by Preferred Term and Outcome, while Table 2 contains a description of the 16 deaths, one of which was associated with cosmetic use.
It should be noted that these data come from voluntary reports submitted to the FDA which have been estimated to represent approximately 10% of the actual occurrences. Additional limits to our data include: causality cannot be proved, other Preferred Terms may exist that might increase our counts, some fields such as Outcome are not consistently filled in, and reports from individual consumers were not included.
Failure to immediately take these actions is likely to result in more preventable serious adverse reactions and deaths from these products.
Environmental impact statement
Nothing requested in this petition will have an impact on the environment.
We certify that, to the best of our knowledge and belief, this petition includes all information and views on which this petition relies, and that it includes representative data and information known to the petitioners that are unfavorable to the petition.
Elizabeth Barbehenn, PhD
Peter Lurie, MD, MPH
Shiloh Stark, B.A.
Sidney M. Wolfe, MD
Public Citizen’s Health Research Group
 http://www.emea.europa.eu/humandocs/PDFs/EPAR/neurobloc/411400en8b.pdf. Accessed September 4, 2007.
 http://www.emea.europa.eu/humandocs/PDFs/EPAR/neurobloc/411400en8b.pdf. Accessed September 6, 2007.
 http://www.dkma.dk/visUKLSArtikel.asp?artikelID=11694. Accessed September 6, 2007.
 http://www.emea.europa.eu/humandocs/PDFs/EPAR/neurobloc/H-301-PI-en.pdf. Accessed September 21, 2007.
 http://www.emea.europa.eu/humandocs/PDFs/EPAR/neurobloc/H-301-PI-en.pdf. Accessed December 4, 2007.
 http://www.botoxmedical.com/images/pdfs/BOTOX_prescribing_information.pdf. Accessed September 21, 2007.
 Medical Officer’s Review of botulinum toxin type B. http://www.fda.gov/cder/biologics/products/botelan120800.htm. Accessed December 6, 2007.
 Medical Officer’s Review of Botulinum toxin type A. http://www.fda.gov/cder/biologics/products/botaller122100.htm. Accessed December 6, 2007.
 Medical Officer’s Review of Botulinum toxin type A. pp.42-43. http://www.fda.gov/cder/biologics/review/botaller122100r2p4.pdf.
 Medical Officer’s Review of botulinum toxin type B. p.74. http://www.fda.gov/cder/biologics/products/botelan120800.htm. Accessed December 6, 2007.
 Medical Officer’s Review of botulinum toxin type B. p. 78. http://www.fda.gov/cder/biologics/products/botelan120800.htm. Accessed December 6, 2007.
 Medical Officer’s Review of botulinum toxin type B. p. 44. http://www.fda.gov/cder/biologics/products/botelan120800.htm. Accessed December 6, 2007.
 Medical Officer’s Review of botulinum toxin type B. p.77.
 Medical Officer’s Review of botulinum toxin type A. pp.43 & 54. http://www.fda.gov/cder/biologics/review/botuall041202r2-2.pdf. Accessed January 17, 2008.
 Cote TR, Mohan AK, Polder JA, et al. Botulinum toxin type A injections: Adverse events reported to the U.S. Food And Drug Administration in therapeutic and cosmetic cases. J Am Acad Dermatol. 2005;53:407-15.
 There could be multiple sources of reports for one patient (case), but if any one was listed as “foreign,” we classified the case as foreign.
 Although these search terms also occasionally produced records with other preferred terms such as pneumonitis, we refer to the results of the restricted preferred term search as dysphagia, pneumonia and aspiration as these were by far the most common.
 Smalley W, Shatin D, Wysowski DK, et al. Contraindicated use of cisapride. Journal of the American Medical Association 2000;284:3036-9.
 Graham DJ, Drinkard CR, Shatin D, Tsong Y, Burgess MJ. Liver enzyme monitoring in patients treated with troglitazone. Journal of the American Medical Association 2001;286:831-3.
 Willy ME, Manda B, Shatin D, Drinkard CR, Graham DJ. A study of compliance with FDA recommendations for pemoline (Cylert). Journal of the American Academy of Child and Adolescent Psychiatry 2002;41:785-90.