OR WAIT 15 SECS
A review of taints and odors in the pharmaceutical and consumer healthcare industries.
Odorous taints, although relatively common in the food and beverage industries, are extremely rare in the pharmaceutical and consumer healthcare industries. Since 2009, multiple pharmaceutical and consumer healthcare companies have recalled products for a musty moldy odorous taint, imparted by 2,4,6-tribromoanisole (TBA), and detected through customer complaints. These taints arise from an external source as opposed to off-odors or off-flavors, which can arise from internal changes to a product from microbial spoilage. The haloanisole taints, TBA and 2,4,6-tricholoroanisole (TCA), generated by the fungal biomethylation of 2,4,6-tribromophenol (TBP) and 2,4,6-trichlorophenol (TCP) respectively, have very low thresholds for detection by the human nose, at parts per trillion (ppt), and high volatility relative to other organohalogen taints.
In the fall of 2010, the Parenteral Drug Association (PDA) established a task force of industry representatives with expertise in quality assurance, packaging, supply chain, microbiology, analytical chemistry, and toxicology. The Technical Report written by this task force benchmarks the current situation regarding TBA, TCA, TBP, and TCP taints and odors in the industry, addresses supply chain and analytical risk-mitigation strategy, and provides toxicology and safety data as well as highlights concerns from regulatory authorities. The purpose of this article is to summarize the findings of the PDA Technical Report No. 55 on this topic, published in April 2012 (1).
NICHOLAS RIGG/GETTY IMAGES
Wood pallets—an unexpected source. Almost all of the recalls cited in this report have been linked to TBA taints entering into Puerto Rico facilities from wood products that have been treated with TBP. Although TBP is registered as a wood preservative in South America, it is not registered in the United States. Thus, the risk of TBP entering the supply chain via wood pallets is higher for manufacturing sites based in Puerto Rico than in the continental United States. In addition, the warm, humid environment in Puerto Rico increases the probability that xerophilic fungi will grow on wood pallets converting TBP to TBA through biomethylation.
Prior to the series of TBA incidents in 2009, pallets were considered to have a very low risk within the supply chain, and as such, had limited quality oversight. A typical Quality Assurance (QA) manager, therefore, would have little or no knowledge about the standards or lifecycles of pallets. Some information on the pallet industry and types of pallets is included in the PDA Technical Report.
Multiple companies affected. Starting in December 2009, multiple companies identified issues related to TBA taints. As of August 2012, there have been 20 recalls from eight pharmaceutical and consumer healthcare companies that share many common features (see Table I). Details from these recalls are provided in the PDA Technical Report. Some commonalities among the cases include: TBP-treated lumber was used for wood pallets; the geographical location of plastic packaging components manufactured was Puerto Rico; climatic conditions where plastic packaging components were manufactured and stored were conducive to fungal growth; tainting of plastic packaging components occurred; and the implicated products were largely compressed tablets.
Table I: TBA/TCA taint cases from the pharmaceutical and consumer healthcare industries.*
The initial investigations of musty, moldy odors by Johnson and Johnson McNeil Consumer Healthcare inconclusively focused on fungal growth on the tablets because of the complaints of a moldy odor (1). Subsequent analysis by gas chromatography (GC)/mass spectrometry (MS)/olfactometry indicated that the odor was due to the taint TBA. It was concluded that the most probable root cause of the odor from TBA was exposure of the product to packaging components that had been shipped to and stored at the pharmaceutical manufacturing site on wood pallets constructed from TBP-treated lumber that had entered the supply chain. It was observed that the odor persisted in the open containers after tablets had been removed, thereby suggesting that TBA resided in high density polyethylene (HDPE) containers and/or cap liners (1). In March 2010, FDA, in response to these recalls, developed a Level 2 GMP Building and Facility Q&A guidance to address the use of pallets to prevent risks from TBA taints (4).
These cases reinforce that pharmaceutical and consumer products are susceptible to contamination from haloanisole taints, such as TBA and TCA. Thus, the intent of the PDA Technical Report is to provide more guidance for the bio/pharmaceutical industry on how to identify and mitigate risks from TBA or TCA tainting.
Lessons from the food and beverage industries
TBA and TCA taints are long and well recognized in the food, wine, and beverage industries (2). The technical literature from these industries is useful to the pharmaceutical industry regarding the origin of these odors and taints, analytical methods developed, and risk-management strategies that may be employed.
It is widely recognized that TBA and TCA taints can migrate into packaging materials, ingredients, and products by leaching or diffusion. Literature examples from the food and beverage industries indicate that chemically porous polymeric materials, such as wood, corrugate, and plastic, are susceptible to contamination with these taints (see Table II). Likewise, food and beverage products are quite porous and can readily absorb these taints from contaminated materials. Consumers having sensory expectations are more likely to notice contamination of food, beer, and wine because these taints can also produce unpleasant tastes that would more likely be noted during consumption.
Table II: Representative literature references to Organohalogen taints in the food, beverage, consumer healthcare, and pharmaceutical industries (1966 to 2010).
PDA TBA/TCA benchmarking survey
The PDA task force prepared a survey to benchmark knowledge of TBA/TCA odors and taints and actions taken within the bio/pharmaceutical industry to mitigate the risks of these musty/moldy odors and taints. PDA distributed the survey in May 2011 to 27 pharmaceutical, consumer healthcare, and biotechnology manufacturers, as well as packaging suppliers represented on the Task Force, to collect definitive feedback (3). The survey was sent to specific experts within these companies, and it was requested that the responses reflect the current position of the organization regarding how issues with these taints are handled.
The responses from 19 companies (70% of those polled) were used to help benchmark industry practice and conduct a gap analysis. The 32-question survey covered the following areas: complaint-handling system, analytical methods, supply-chain controls, and regulatory issues. The full results from this survey are available through PDA (3).
Vulnerabilities in the supply chain
One outcome of the PDA Technical Report is that each company should review its supply chain to ensure that proper good distribution practices (GDPs) and controls are implemented to mitigate the risk associated with TBP, TCP, TBA, and TCA contamination. To date, the primary focus in industry has been on the use of heat-treated (HT) wood pallets that are certified to be TBP/TCP-free to prevent risks of TBA/TCA odors and taints from entering into the supply chain. Other potential sources through which TBP/TCP can enter the supply chain should also be considered and are addressed in the PDA TBA Technical Report. For example, TBP/TCP can come from recycled materials and disinfectants. In addition, because fungi perform the biomethylation of TBP (or TCP) to TBA (or TCA), ventilation and moisture controls should be implemented to minimize fungal growth on pallets.
Figure 1: Halophenol to haloanisole biomethylation conversions (adapted from Ref. 2). (FIGURE COURTESY OF AUTHORS)
The primary root cause for TBA-tainted product recalls in the bio/pharmaceutical industry was that wood treated with TBP (or TCP) was used to construct pallets that were then used to ship and or store plastic packaging components. In this supply chain, the TBP (or TCP) from the pallets was biomethylated to TBA (or TCA) which then contaminated plastic packaging components before product filling and, ultimately, the drug products stored in these components. The ability of TBP/TCP and the more volatile TBA/TCA, formed by biomethylation, to readily migrate from one material to another also makes the taint distribution uneven. It also makes determining the ultimate root cause of the contamination challenging.
According to the International Plant Protection Convention (IPPC), most pallets shipped across national borders must be made of materials that are free of invasive insects and plant diseases. The standards for these pallets are specified in the International Standards for Phytosanitary Measures (ISPM) No. 15. Regulation of Wood Packaging Material in International Trade (2009 Revision). In accordance with ISPM 15, wood pallets intended for international trade should be constructed of either heat-dried or methyl bromide-treated lumber. Therefore, tainting could be associated with imported TBP/TCP-treated wood or treated wood pallets from outside the US.
There are approximately 1–2 billion wood pallets in circulation in the US, and an estimated 500 million pallets are replaced annually. Total control of TBP-impregnated wood pallets entering into US commerce, therefore, may be difficult, if not impossible. A unique challenge may be the control of wood pallets constructed from TBP-treated lumber entering Puerto Rico from neighboring South America, where TBP is registered for use as a wood preservative. It is known that TBP is used as a wood preservative in other regions of the world, including Northern Asia and Eastern Europe. In addition, multiple examples in the literature from the food and beverage industries highlight that TBA tainting has been an issue in other regions of the world, including Australia. Based on the available literature, goods are more susceptible to TBA tainting if exposed to TBP-treated wood when moving from higher to lower temperatures under high humidity because the risk of condensation increases. Higher moisture levels increase the risk for fungal growth and, ultimately, for biomethylation of TBP (TCP) to TBA (TCA). The PDA Technical Report provides more details about these specific risks.
The report addresses controls that can be employed to mitigate the possibility of TBA/TCA formation and tainting for pallets, shipping containers, corrugate, and plastics. The main approaches include:
Due to the complex nature of haloanisole tainting, there are varied analytical targets and methodologies that may be employed to detect haloanisole taints. Regulatory guidelines are available for testing and qualification of impurities, extractables and leachables, and genotoxic compounds. For haloanisoles, these levels are several times higher than the levels at which odorous taints can be detected. Further, because human sensory detection is highly variable, detection thresholds range from low ppt to not being able to detect the odor at all. Taint concentrations typically found in pharmaceutical and consumer healthcare samples range from 1 to 2000 ppt and includes both haloanisoles and halophenols. These concentrations require matrix-specific sampling, preconcentration, and the most sensitive detectors to obtain an analytical response. In addition to the complexity introduced by low concentrations, there is a formulary complexity in that materials have different rates of adsorbing, absorbing, and desorbing halocompounds. The solubility of TBA in solids is significantly related to the polarity of the material and, therefore, the impact on TBA recovery and methodology is also discussed in the technical report.
The PDA Technical Report reviews various analytical methods employed for the detection of haloanisoles and halophenols. Several of these methods include: olfactory detection, closed-loop stripping analysis, solid-phase microextraction, liquid–liquid microextraction, GC, and GC/MS. There is no single analytical method or control strategy recommended but rather a useful "toolkit" of methods is presented in the PDA Technical Report to aid in understanding and supporting a control strategy for mitigation of haloanisoles in pharmaceutical and consumer healthcare products. In general, these analytical methods are more suitable for investigatory testing than for routine monitoring. Attention is drawn to the report of TBA Analytical Methods published in Pharmaceutical Technology in November 2011 by scientists from Microanalytics (a contract analytical services laboratory specializing in aroma and odor issues) (6).
Although not widely used in prescription and over-the-counter products, sensory panels have historically been used for consumer products, including those in the food and wine industries. They have also been used for the monitoring and control of raw material shipping containers. The benefits of applying sensory assessment techniques or odor panels in quality control include timeliness and flexibility (i.e., materials can be assessed before they even leave the truck), the ability to correlate results to the potential for customer perception as part of quality risk management, and the fact that sensory perception can be both targeted and non-specific at the same time. The series of ISO standards and guidance documents related to the food industry, specifically ISO 13299, offer good insight into how sensory assessment and odor panels may be adapted for use in the context of monitoring components and finished goods for targeted compounds (9).
Monitoring moisture content of wood
Reduction of wood moisture content may be used to control fungal growth on pallets. Fungi require free moisture termed water activity greater than 0.7 to grow. Controlling moisture to levels below those necessary to support fungal growth (i.e., 20% or lower in water content) reduced the possibility of biomethylation resulting in the formation of taints.
Simple techniques to verify moisture content include conductivity and loss on drying (LOD). Inexpensive handheld conductivity meters common in the construction industry are available. One such device is the Extech MO210, available through various outlets that sell lumber and other building materials. This device measures the conductivity between two metal prongs inserted into the surface of the wood. The conductance measurement is converted to percent moisture reading through an algorithm to a digital readout.
Marketing surveillance and regulatory considerations
Pharmaceutical companies marketing products in the US are required by regulation to establish a postmarketing safety surveillance/pharmacovigilance program, and to communicate to FDA in accordance with 21 CFR 314.80 for reports of adverse events associated with the use of a drug. Further, GMP regulation 21 CFR 211.198 requires pharmaceutical companies to establish a complaint vigilance program and have written procedures for the investigation of any product quality complaint. Such procedures are required to include provisions for review by the quality control unit, of any complaint involving the possible failure of a drug product to meet any of its specifications and, for such drug products, a determination as to the need for an investigation in accordance with 21 CFR 211.192.
Such procedures shall include provisions for review to determine whether the complaint represents a serious and unexpected adverse drug experience, which is required to be reported to the FDA in accordance with 21 CFR 310.305 and 314.80. Two key components of vigilance programs are safety signaling and adverse product quality complaint trending. Adverse events are evaluated by healthcare professionals, typically physicians running the pharmacovigilance program, and product quality complaints are typically investigated by QA professionals.
Toxicology and safety considerations
In a recent state-of-the-art battery of toxicology studies, TBA demonstrated no signs of mutagenic toxicity or systemic toxicity in rodents when doses for up to 28 d at levels reaching 109 times higher than any potential exposure from product (7). TBA levels that have been observed in bio/pharmaceutical and consumer healthcare products are within the range observed in beverages (ppb-ppt) including chlorinated drinking water, wine, and beer. Based on an in-depth review of the literature by the PDA Task Force, these levels of TBA in beverages were not associated with any adverse events.
Several adverse-event analyses have been conducted by companies represented in the PDA TBA Task Force. The results of these analyses do not indicate any causal relationship between TBA exposure and GI events. The "musty/moldy" odor associated with products contaminated with TBA has been reportedly associated with acute, transient, gastro-intestinal adverse events (e.g., nausea, vomiting, and diarrhea). The rodent is incapable of vomiting, but is a good model for diarrhea. In the battery of rodent toxicology studies noted above, there were no signs of diarrhea or any macroscopic or microscopic pathological effects observed along the GI tract.
A potential exposure to TCP, TBP, or TCA in the 100 to 1000 ppt range in product would lead to an acceptable, de minimis (minimal effect) exposure based on the acceptable chronic exposure levels of TCP. An analysis of all of the above, toxicology data, environmental exposure data, and adverse events reporting suggest that any human health risk associated with clinical exposure to these ppt levels of TBA is inconsequential and presents a de minimus safety risk. TBA occurs naturally in our environment and human exposures to these ppt levels are not uncommon.
A risk-mitigation plan for TBA/TCA tainting can be established by a pharmaceutical company at its manufacturing sites that follows the International Conference on Harmonization (ICH) Q9 guideline principles (8). Details from this plan are available in the PDA Technical Report for the following:
Useful risk-mitigation steps identified by the PDA Task Force include not constructing pallets from TBP treated lumber, controlling the moisture content of wood to levels not conducive to fungal growth, improved supply chain awareness of haloanisole taints, elimination of other sources of halophenols, and adequate environmental control and ventilation in warehouses and during transportation.
Toxicological and safety studies conducted on TBA demonstrated no mutagenicity or systemic toxicology in rodents when dosed for up to 28 days at levels a billion-fold higher than potential human exposure from the recalled product. TBA dosing produced no diarrhea or any macroscopic or microscopic pathological effects along the GI tract in rat toxicity studies. Although nausea was reported by consumers sensing the musty, moldy odor, adverse event analysis by multiple recalling companies have not established a causal relationship between TBA and gastrointestinal events. Therefore, reactions of disgust to TBA taints appear to be a sensory and/or behavioral response and not toxicological and, therefore, is a patient compliance risk rather than a patient safety risk.
Based on the high margin of safety demonstrated in toxicity studies, there is no meaningful analytical threshold that can be based on toxicity. It is therefore necessary for individual companies to consider how the odor is being perceived by their customers and the likelihood that perception to the odor could impact patient therapy (i.e., the concern is that the musty, moldy odor from these taints could increase the likelihood that patients will not take their medication). Finally, it is up to the pharmaceutical and consumer product manufacturer to understand their pallet supplier(s), classify them appropriately, and work with the supplier's controls on pallet manufacturing. Companies should implement internal pallet controls to minimize risks from TBA (TCA) tainting accordingly as noted in the PDA Technical Report.
This paper was contributed by the Parenteral Drug Association (PDA) TBA Task Force Members.
The content of this article, including the figures and tables, was adapted from PDA Technical Report No. 55, available at https://store.pda.org/ProductCatalog/Product.aspx?ID=1549.
1. PDA, "Technical Report No. 55: Detection and Mitigation of 2,4,6-Tribromoanisole and 2,4,6-Trichloroanisole Taints and Odors in the Pharmaceutical and Consumer Healthcare Industries," (April 2012), available at https://store.pda.org/ProductCatalog/Product.aspx?ID=1549.
2. Anon, "Organohalogen Taints in Foods," Australian Food and Grocery Council Supplement to Food Australia, 59 (3), (2007), www.afgc.org.au/cmsDocuments/Organohalogen.pdf
3. PDA, "Risk Mitigation of Tribromoanisole (TBA)/Trichloroanisole (TCA) Taints and Odors: A Pharmaceutical Industry Benchmarking Survey" (2011), available at https://store.pda.org/bookstore/ProductDetails.aspx?productabbreviation=45000.
4. FDA, "Questions and Answers on Current Good Manufacturing Practices, Good Guidance Practices, Level 2 Guidance–Buildings and Facilities" (FDA, Rockville, MD), www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm192869.htm.
5. T. Ramstad and J.S.Walker, Analyst 117 (8), 1361–1366 (1992).
6. R.J. Bleiler, F. Kuhrt, and D. Wright, Analytical Technology and Laboratory Testing, supp. to Pharm. Technol. 35 (11) s10-14 (2011).
7. F. Koshier et al., Food Chem. Toxicol. 49 (9), 2074–2080 (2011).
8. ICH, Q9 Quality Risk Management (2005).
9. ISO 13299:2003, Sensory Analysis-General Guidance for the Design of Test Rooms.