Abstract
Background:
Butterbur supplements are available in the USA and Canada and are commonly used for treating migraines. Petadolex, a special butterbur extract from Petasites hybridus, is a natural herbal product and the only butterbur extract with proven clinical efficacy in migraine prevention. The Complimentary Migraine Guidelines of the AAN mention butterbur as level A recommendation for the prevention of chronic episodic migraine. However, these guidelines have been retired.
Methods:
We review suspected serious liver cases, pyrrolizidine alkaloids, regulatory issues, preclinical and clinical data of the special butterbur extract Petadolex.
Results:
The RUCAM (Roussel Uclaf Causality Assessment Method) test found no probable relationship between the butterbur root extract Petadolex® and cases of serious liver injury. Two cases of non-serious reversible liver enzyme elevations were rated as probably related to Petadolex®. The safety is supported by preclinical data in animals as well as in-vitro toxicology experiments. In addition, Petadolexis free of detectable levels of pyrrolizidine alkaloids.
Conclusion:
There is no evidence that the special butterbur root extract Petadolex poses a substantial risk of liver injury for patients.
Introduction
Many butterbur extracts or even drug powders are marketed in various countries, especially in the United States. However, those extracts are not comparable among each other. Similar to biopharmaceutical products, the manufacturing process critically determines the properties of the butterbur extract. The following variables impact the properties and differ considerably between extracts: variety of species (petasin chemotype or furan chemotype), plant part (whole plant, leaves, and roots), cultivation versus wild collection, extraction solvent, extraction procedure, purification process, and content and amount of pyrrolizidine alkaloids (PAs). Among the many commercially available butterbur products for migraine prevention, only Petadolex®, a proprietary CO2-butterbur extract made from the root of Petasites hybridus, is well characterized based on its unique and distinct manufacturing processes. Therefore, published results of Petadolex are not transferable to any other butterbur extract. This special extract is mainly available in the United States and Canada and is the only butterbur extract in the market with a proven published clinical efficacy in placebo-controlled randomized clinical trials of migraine prevention. 1 –3 There is only one more well-characterized butterbur extract with published preclinical and clinical data available among the plethora of commercial butterbur products. However, this extract (ZE399) is used for seasonal rhinitis and is made from leaves of P. hybridus. 4 –6
The main pharmacologically active components of the P. hybridus plant belong to the group of sesquiterpenes and include petasin and isopetasin, which are present at a minimum of 15% in the special butterbur root extract Petadolex. These molecules are spasmolytics acting on smooth muscle and tracheal rings in vitro through effects involving calcium channels and calcium mobilization. 7,8 Petasin and isopetasin have also been demonstrated to exert anti-inflammatory effects by inhibiting leukotriene synthesis and cyclooxygenase-2. 9,10 Isopetasin has also been reported to desensitize peptidergic nociceptors by acting on TRPA1 cation channel. 11 The mode of action of butterbur in migraine prevention is not known. However, sterile inflammation and ion channels play a role in the genesis of migraine attacks. Apart from the sesquiterpenes, butterbur as a plant contains toxic PAs. These toxic substances are removed from the extract by a proprietary treatment but may be present in other butterbur extracts.
The data for this review are based on literature research of the PubMed library and the latest Periodic Safety Update Report (PSUR) of the butterbur extract Petadolex, which has been provided by the manufacturer. Marketing authorization holders are required to submit PSURs, which are regularly updated pharmacovigilance documents intended to provide an evaluation of the risk–benefit balance of a medicinal product at defined time points post-authorization.
Material and methods
Suspected adverse drug reactions in clinical trials and from spontaneous reporting were provided by the manufacturers in the form of the latest PSURs. A recent publication is summarized that reevaluated the serious adverse reactions that were suspected to be related to Petadolex administration using the updated 2016 version of the hepatotoxicity-specific diagnostic Roussel Uclaf Causality Assessment Method (RUCAM) test. 12 Also reviewed are published analyses of PAs in various US butterbur products. 13 Internal data regarding regulatory issues, in vitro studies, and preclinical and clinical data of Petadolex were provided by the manufacturer. 14
Clinical trials in migraine prevention with Petadolex
The guidelines of the American Academy of Neurology, 15,16 the German Headache and Migraine Society, 17 and the Canadian guidelines 18 recommend β-blockers (propranolol, metoprolol, and timolol), antiepileptic drugs (topiramate, divalproex sodium, and sodium valproate), and the tricyclic antidepressant amitriptyline for migraine prevention. However, many patients have a desire for natural or herbal treatments that are believed to be a “safe” form of treatment with little or no risk of side effects.
The clinical benefit of the butterbur root extract in migraine prevention has been demonstrated in two controlled trials with adults and in one open trial in children with a total of 397 patients. The two placebo-controlled clinical studies have been performed according to the guidelines of the International Headache Society (IHS) for controlled trials of drugs in migraine. 19
In the first single-center randomized placebo-controlled trial with 60 migraine patients, the mean attack frequency per month decreased from 3.4 at baseline to 1.8 after 3 months in the verum group and from 2.9 to 2.6 attacks per month in the placebo group (p = 0.0024). The responder rate, which is defined as an improvement of migraine frequency by greater than 50%, was 45% in the verum group and 15% in the placebo group. 1
The multicenter randomized placebo-controlled trial included 229 migraine patients. Migraine attack frequency was reduced by 45% for the special butterbur root extract at a dose of 75 mg bid (p = 0.005). The proportion of patients with a ≥50% reduction in attack frequency was 68% (p<0.05). 2
In the open study with 108 children and adolescents, 77% of all patients reported a reduction in the frequency of migraine attacks of at least 50%. Attack frequency reduced by 63% from baseline. 3
Pyrrolizidine alkaloids
Pyrrolizidine alkaloids are hepatotoxic substances. European Food Safety Authority considers a threshold of 0.007 μg PA/kg body weight per day as acceptable exposure via foodstuffs based on lifelong exposure. 20 –22 This threshold corresponds to 0.35 μg PA daily intake for a person weighing 50 kg. There are no publications issued by the Food and Drug Administration (FDA) limiting PA intake. The National Center for Natural Products Research analyzed 21 commercial butterbur products available in the United States employing the latest chemical analysis techniques for the detection of petasins and PAs. The analysis included the proprietary butterbur root extract Petadolex. 13 Several different companies market the proprietary butterbur root extract from the original manufacturer under different brand names. Products containing butterbur powder or an extract that is not identical or comparable with Petadolex are also sold. In 7 of the 21 butterbur dietary supplements analyzed, the amount of petasin was found to be within limits of label claim, and with no detectable levels of PAs. Three of those seven products (labeled #9807, #9808, and #9819 in the publication by Avula et al.) 13 were provided by the manufacturer and contain the butterbur extract Petadolex. Six products tested contained no detectable amounts of petasins at all. In seven products, PA levels between 0.1 and 4.48 μg per tablet, capsule, or gel content were detected, and one product was a liquid with 8.43 μg of PAs per milliliter.
Toxicity studies in vitro
In a publication by Anderson et al., 23 three butterbur root extracts from P. hybridus with different amounts of petasins were analyzed in vitro in primary cultures of rat and human hepatocytes. One of the extracts was the commercially available butterbur root extract Petadolex, which contains approximately 15% petasin. The other two butterbur root extracts had 37% petasin and only trace amounts of petasin, respectively. The aim of these experiments was to study whether petasins could contribute to liver toxicity. The three extracts had no effect on cell viability, cytotoxic cell damage, or changes in aspartat-aminotransferase (AST) and alanin-aminotransferase (ALT) levels in primary rat hepatocytes. In primary human hepatocytes, only the extract containing 37% petasins, which corresponds to a >90-fold higher concentration than the therapeutic plasma level, showed signs of decreased cell viability and metabolic capacity and elevation of ALT and AST.
Toxicity studies in animals
A 6-month chronic toxicity study of the butterbur root extract Petadolex was performed in a study with 40 rats per group. 24 Dose ranges were from 45 mg/kg to 270 mg/kg, corresponding to a 10- to 60-fold greater amount of the maximal clinical human dose. Increases in alkaline phosphatase (AP), gamma-glutamyl-transferase (GGT), and bilirubin were noticed early in the treatment, but also in the vehicle group at nearly the same frequency in all dose groups. All levels in all dose and vehicle groups returned to normal levels after 13 weeks. This normalization of GGT indicates an adaptive metabolic process rather than a toxic effect. AST and ALT did not change during the 6-month treatment. Histological findings showed bile duct dilatations and biliary duct hyperplasia without liver cell damage in the two highest dose groups, which would be an expected response to the metabolic effects of large amounts of the very dense and lipophilic butterbur root extract Petadolex.
The special butterbur root extracts was recently tested in mini-pigs for 28 days. 14 A butterbur extract, which contained only trace amounts of petasins, was used as control to study if petasins had a toxic effect on the animals. The extracts were applied at a dose of 720 mg/kg body weight, which corresponds to a human equivalent dose of 654 mg/kg. Since the maximal clinical dose is 3 mg/kg per day, the test dose represents a 218-fold increase beyond the clinical dose. No detrimental effect of petasins could be detected. All laboratory parameters, including hematology and clinical chemistry (including liver parameters bilirubin, ALAT, ASAT, AP, LDH, GGT, GLDH, and 5-nucleotidase), showed normal values in all test groups at the end of the study at day 28. Histopathology of the liver did not reveal any drug-related changes suggesting liver toxicity, such as the formation of vacuoles.
Suspected nonserious adverse drug reaction of the hepatobiliary system in humans
Twelve reports of suspected adverse drug reactions involving the hepatobiliary system were reported in clinical trials of the special butterbur root extract Petadolex with a total of 397 patients. 1 –3 All of these cases concerned increased liver enzymes.
Three cases from the single-center double-blind randomized placebo-controlled trial with 60 migraine patients were categorized as not clinically relevant since liver enzyme elevations were well below a two-fold increase over the reference values. 1 In the multicenter, double-blind, randomized placebo-controlled trial study of 229 patients, nine cases of increased liver enzymes were documented. Seven of them were assessed as possibly and two as unlikely related to the special butterbur root extract. 2 No liver-related adverse events occurred in the open trial, with 108 children suffering from migraine. Liver enzymes were not analyzed in this noninterventional study. 3
Between 1972 and November 2015, a total number of 233 suspected adverse drug reactions have been reported outside of clinical trials 14 ; 198 of these spontaneous reports came from German patients/doctors, 20 reports from Switzerland, five reports from the United Kingdom, one report from Austria, and nine from the United States.
Suspected adverse drug reactions reported with the highest frequency were mild gastrointestinal discomforts, such as nausea, burping or belching, and stomach pain. There were 50 nonserious suspected adverse drug reactions involving the hepatobiliary system reported. Of these cases, two were assessed as probably related to the butterbur root extract, 15 as possibly related, seven as unlikely related, and four as not being causally related to the butterbur root extract; 22 suspected adverse drug reactions could not be classified because requested data were not obtained by the physicians after repeated requests.
Suspected serious adverse drug reactions of the hepatobiliary system in humans
Of the 233 suspected adverse drug reactions reported outside of clinical trials, 12 were serious. Ten of them concerned the hepatobiliary system. The other two serious cases were believed to be unrelated to the proprietary butterbur extract: one was a case of pancreatitis and the other case was a status asthmaticus. Reports of suspected serious adverse hepatic drug reactions came from Germany (eight cases between 1998 and 2006), UK (one case in 2004) and Austria (one case in 2011).
Nine of the 10 reported serious cases involving the hepatobiliary system have been previously evaluated by an independent expert group (Prof. Juergen Borlak, Prof. Hans-Christoph Diener, Prof. Stefan Evers, and Dr. Reinhard Spanel) on 28 July 2008. The evaluation performed according to the WHO-UMC criteria, and results have been published by Evers. 24 In summary, of the nine serious hepatobiliary suspected adverse drug reactions, only one reversible case of hepatitis with cholangitis was assessed according to WHO-UMC as likely related to the intake of the butterbur root extract Petadolex (Table 1).
Comparison of causality evaluations of the 10 suspected serious adverse drug reactions of Petadolex® concerning the liver using RUCAM versus WHO-UMC.
RUCAM: Roussel Uclaf Causality Assessment Method; ALT: alanin-aminotransferase; DILI: drug-induced liver injury.
All 10 serious hepatobiliary suspected adverse drug reactions were reassessed in a recent publication using the RUCAM as updated in 2016 (Table 1). 12 In contrast to the WHO-UMC method, the RUCAM test (previously called CIOMS) was developed and validated for drug-induced liver injury (DILI) or herb-induced liver injury (HILI) and does consider hepatotoxicity-related characteristics. 25 Based on individual narratives, causality assessment by RUCAM including a search for alternative causes, the clinical assessment, confounding variables, and comparative evaluations, there is no evidence of liver injury by the butterbur root extract. Only two cases were found to be possibly related to Petadolex, five cases were found to be unlikely related, and three cases were not linked to the butterbur root extract at all. None of the cases was probably related to the butterbur root extract. Two cases of liver transplantations occurred in the 10 serious adverse events. One case was rated as unlikely by RUCAM as well as WHO-UMC and the other case was rated as unlikely by RUCAM and possible by WHO-UMC.
The authors of the article state that idiosyncratic HILI characteristics were not applicable to the 10 liver cases, as RUCAM-based causality for the butterbur root extract does not exist. There was also no evidence of an intrinsic HILI by the butterbur root extract because dose dependency was not evident; the liver diseases occurred with short- and long-term use. 12
The reports of the 10 suspected cases of hepatotoxicity were mainly restricted to Germany. Not a single case of liver damage was reported in the United States, Canada, or Japan.
Regulatory issues of the butterbur root extract Petadolex
Germany
Petadolex was available in Germany from 1972 to 2009 as a herbal medicine. Until 1988, methylene chloride was used in as an extraction solvent in the preparation of the extract. In 1988, the solvent was changed to CO2, which is superior for removing PAs present in the plant. In 2002, the German Drug Regulatory Authority BfArM reviewed the dossier of the proprietary butterbur root extract and determined that the change in the extraction method that took place in 1988 was a change in the active ingredient, and therefore, the BfArM considered the product marketed since 1988 not to be the same as the one for which a marketing authorization was granted in 1978. As a consequence, the renewal of the marketing authorization in Germany was rejected in late 2008, even though chemical analysis suggested that the first extract with methylene chloride was comparable with the extract with CO2 as solvent. It needs to be noted that all clinical trials in migraine prevention have been performed with the CO2-butterbur extract Petadolex that is available in the United States and Canada.
The United Kingdom
The butterbur root extract Petadolex was sold under a different brand name in the United Kingdom as a section 12 product (traditional herbal product with no indication or claim). Effective April 2011, all products of this class could not be marketed unless a new license was obtained. The UK Medicines and Healthcare products Regulatory Agency (MHRA) has rejected the application for a new traditional license for the butterbur root extract on safety concerns related to the distribution of the extract as a migraine preventative through mass market retail stores, rather than under supervision by doctors. As a result, Petadolex is not available for sale in the United Kingdom.
Canada
Health Canada approved the butterbur root extract Petadolex as a natural health product for the prophylactic treatment of migraine in 2012. The extract has already been sold beginning in 2005 as a herbal supplement.
Switzerland
Bioforce AG Roggwil was a distributor of the butterbur root extract Petadolex under the brand names Petadolor and Dolomed in Switzerland. The Swiss health authority Swissmedic revoked the previously granted marketing authorization for Switzerland in January 2004. The revocation was based on six suspected serious adverse reactions of the hepatobiliary system reported by that time in Germany; however, there were no cases in Switzerland. The withdrawal by Swissmedic was also influenced by a recall of two batches of the butterbur root extract from the market by the German manufacturer as a precautionary measure not requested by the authorities. The reason for this recall was the detection in 2003 of a substance suspected to be a PA by a worst case assumption. However, the substance was never confirmed to be a PA.
The United States
The special butterbur root extract Petadolex from P. hybridus has been available in the United States as a dietary supplement since 1998.
Discussion
Data from several sources were analyzed to evaluate the hepatotoxic potential of the special butterbur root extract Petadolex from P. hybridus. A total exposure of about 2.3 million patient months is estimated from 1992, when records are available, until 2016. The exposure in the United States, Canada, and Japan is estimated to be about 1.029 million patient months. These numbers are based on the total packs sold at the recommended daily dose of 150 mg extract. 14 The actual number of exposed patients is believed to be higher. Many patients use a daily dose of 100 mg, which would increase the patient exposures. It is of interest to note that the CO2 extract had already been marketed since 1988, following the methylene chloride extract marketed from 1972 until 1988. No sales records are available from 1972, the launch date of the first methylene chloride extract until 1992.
Only 12 liver-related adverse events were reported in clinical trials, and all of them concerned elevated levels of liver enzymes, which were reversible and mostly moderate. A total of 38 cases involving the hepatobiliary system were spontaneously reported by doctors or patients outside of clinical trials. Among them were 10 cases of hepatitis that were considered serious. Initially evaluated by the WHO-UMC system, the hepatitis cases were recently reevaluated using the RUCAM test since the WHO-UMC method was developed for general adverse events and is not liver-specific. The RUCAM test found only two possible cases of liver injury that may have been caused by the butterbur root extract Petadolex in relation to a total exposure of about 2.3 million patient-months as estimated from 1992. 12
Not a single case of liver damage was reported in the United States and Canada with this extract. The difference in reporting compared with Europe could be explained by the fact that in contrast to the United States, the butterbur root extract was regulated in Europe as a medicine. Because the first suspected serious adverse liver case was rated by the treating clinician as probable, the manufacturer included hepatitis as a side effect in the patient leaflet. Due to the wording in the package insert, physicians may have more often considered the butterbur root extract as a cause in hepatitis cases.
In the United States, the proprietary butterbur root extract is a food and not a medicine, the terminology of patient leaflets is therefore different: The US package insert contains a warning that addresses symptoms at which the butterbur root extract has to be discontinued.
No correlation of the occurrence of hepatitis with liver enzyme increases in ALT and AST or dosage or treatment duration existed. Suspected hepatitis cases occurred at the lowest daily intake dose of 50 mg, as well as at a daily dose of 150 mg irrespective of treatment duration or levels of liver enzymes. This lack of correlation is an unusual finding in DILI or HILI.
Several surveillance databases exist in the United States for post-marketing assessment of herbal hepatotoxicity. The National Electronic Injury Surveillance System supports the Consumer Products Safety Commission and is a national probability sample of hospitals in the United States and its territories. A search of the database revealed no entries for butterbur. The same was true for a search of the FDA site and the FDA Adverse Event Reporting System. 26
Preclinical data in animals as well as in vitro toxicology experiments do not support the reports of suspected adverse drug reactions involving the hepatobiliary system. In addition, there is no evidence that petasins at therapeutic levels are toxic substances based on in vitro toxicology experiments.
Even though the manufacturer of the butterbur root extract Petadolex states that PAs have been removed during the extraction and purification steps of the extracts, concern still exists that the suspected cases of hepatitis were caused by PAs. The National Center for Natural Products Research, Research Institute of Pharmaceutical Sciences, The University of Mississippi could not detect PAs in the butterbur root extract Petadolex. 13 The limits of detection using HPLC-TOF-MS for PAs was 1 ng/ml. The analysis was able to detect the PAs typically present in P. hybridus, like seneciphylline, integerrimine, senecionine, senkirkin, 7-angeloylretronecine, 9-angeloylretronecine, and their corresponding N-oxides. However, the institute detected PAs in seven products of butterbur root extracts from other manufacturers. The authors are not aware whether liver toxic adverse events have been reported with these extracts. It is obviously difficult for consumers to trust the label whether a product sold in the United States is really PA-free. It would be helpful if products obtained a FDA confirmation of its PA-free status. Alternatively, proven PA-free butterbur extracts could be listed along with the brand names in migraine guidelines.
Petadolex is not available in Germany and the United Kingdom anymore. The manufacturer has not obtained a new approval in these countries. A full documentation, as for a new drug application, would have been mandatory. In addition to the already published trials, a long-term clinical trial in a large number of patients as well as another toxicity study plus carcinogenicity, reprotoxicity, and clinical pharmacology studies would have been required for the new registration of the proprietary butterbur root extract in Germany. The manufacturer is a small-sized enterprise, hence the costs of this program were considered to be too high.
According to the manufacturer, in the United Kingdom, a herbal product such as butterbur can only be licensed as a traditional medicine. However, the indication “prophylactic treatment of migraine” is not accepted for the proprietary butterbur root extract as a traditional herbal medicine and a full pharmaceutical license is not possible for herbal preparations in the United Kingdom.
A butterbur extract made from leaves instead of roots is sold in Switzerland and other countries for the treatment of seasonal rhinitis under various brand names. 4 –6,27 There are no reports of adverse drug reactions with the butterbur leaf extract affecting the liver. One could argue that this observation is due to the fact that treatment for seasonal rhinitis is usually shorter than the treatment for migraine prevention. Market experience, however, shows that many migraine patients use the butterbur root extract for not more than 4 weeks. In addition, suspected liver cases with the butterbur root extract Petadolex did not show a correlation with time. Both extracts, the root extract and the leaf extract, contain similar amounts of petasins and other molecules. Both extracts do not contain PAs and both extracts have the same nonclinical in vivo and in vitro safety profile. There is no convincing evidence that the butterbur root extract Petadolex poses an increased risk for patients. This is additionally supported by the reevaluation of the suspected liver cases by RUCAM. Clinical data of placebo-controlled double-blind clinical studies show that patients with frequent migraines benefit from the butterbur root extract.
Key findings
Many different butterbur products for migraine prevention are commercially available in the United States. Only a proprietary root extract of P. hybridus (Petadolex) is well characterized and the only extract with a proven clinical efficacy in migraine prevention. Unlike some other butterbur extracts in the market, Petadolex is free of detectable levels of PAs. Reports of suspected cases of hepatitis with Petadolex in the past have raised concerns. According to the RUCAM test, there is no probable relationship between the butterbur root extract Petadolex and serious cases of liver injury. The safety of Petadolex is supported by preclinical data in animals as well as in vitro toxicology experiments.
Footnotes
Authors’ contribution
HCD and UD were involved in drafting, conception, and design of the manuscript. HCD and FGF provided analysis and interpretation of the data. All authors read and approved the manuscript.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: HCD was the principal investigator of a randomized study with Petadolex®. He did not receive any compensation for the writing of this article. He chairs the Writing Committee of the German Neurological Society Guidelines on the treatment of migraine. HCD received honoraria for participation in clinical trials, contribution to advisory boards, or oral presentations from Addex Pharma, Alder, Allergan, Almirall, Amgen, Autonomic Technology, AstraZeneca, Bayer Vital, Berlin Chemie, Böhringer Ingelheim, Bristol-Myers Squibb, Chordate, Coherex, CoLucid, Electrocore, GlaxoSmithKline, Grünenthal, Janssen-Cilag, Labrys Biologics, Lilly, La Roche, 3M Medica, Medtronic, Menerini, Minster, MSD, Neuroscore, Novartis, Johnson & Johnson, Pierre Fabre, Pfizer, Schaper and Brümmer, Sanofi, St. Jude, Teva, and Weber & Weber. HCD has no ownership interest and does not own stocks of any pharmaceutical company. FGF has served on advisory boards for Promius Pharmaceuticals, Teva Pharmaceuticals, Zogenix Pharmaceuticals, and Biohealthonomics. He has served on speaker’s bureaus for Allergan Pharmaceuticals, Teva Pharmaceuticals, and Zogenix Pharmaceuticals. He holds 50 shares of common stock in Capnia Inc. UD is employed by the manufacturer of Petadolex.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial support for research projects was provided by Allergan, Almirall, AstraZeneca, Bayer, Electrocore, GSK, Janssen-Cilag, MSD, and Pfizer. Headache research at the Department of Neurology in Essen is supported by the German Research Council (DFG), the German Ministry of Education and Research (BMBF), and the European Union.