FDA Releases List of 'Safe' Valsartan and ARBs While Experts Explore and Debate Root Causes of Contamination

On April 5, 2019, FDA’s Commissioner Scott Gottlieb and Janet Woodcock, director of FDA’s Center for Drug Evaluation and Research (CDER) issued an update on the agencies investigation of genotoxic nitrosamine impurities in angiotensin receptor blockers (ARBs) that are used to treat high blood pressure and heart failure.  The agency released a list of 40 ARB medicines that have been found not to contain any known nitrosamine impurities; Gottlieb and Woodcock expect this number to increase. The list also includes other products on the market are still being evaluated, for which impurities either have not been detected or are below interim acceptable contamination levels.

Acknowledging that nitrosamines are present in the environment, and found in water and food, Gottlieb and Woodcock maintained that their presence in ARB medicines, even in small amounts, is unacceptable. A multidisciplinary working group within FDA, as well as global regulators, has been collaborating to understand the root cause of the contamination and to ensure patient safety. Included in the analysis has been testing for impurities, evaluation of the API manufacturing process used to manufacture valsartan, and other information submitted by manufacturers and regulatory agencies. The agency will continue to update this list as more is learned.

Gottlieb and Woodcock say they are working with manufacturers to ensure that sartans containing impurities at higher than interim acceptable levels (e.g., N-Nitroso-N-methyl-4-aminobutyric acid [NMBA] levels of 9.82 ppm), are removed from the supply chain as soon as possible. 

Investigations have led to shortages of sartans, and Gottlieb and Woodcock acknowledged that supplies of related products could also be threatened.

As an interim solution, FDA leadership has decided that it poses less risk for patients to take sartans with levels of contaminants that are below the acceptable interim ceiling, than it would for them to stop taking the drugs, which could result in stroke or other serious health problems.

Gottlieb and Woodcock expect many manufacturers to resume production of impurity-free losartan within roughly six months. FDA has also contacted manufacturers and distributors about conditions and chemicals used in the API manufacturing process that can result in impurities. Included are the use of recycled solvents, catalysts, and raw materials. FDA has urged manufacturers not to rely on supplier claims that materials are new but to ensure it themselves, and to provide regulators with documentation showing that tests were performed.

“Patients should continue taking their medicine until their pharmacist provides a replacement or their doctor provides an alternative treatment option, even if they learn that their ARB medicine is recalled. The risk associated with abruptly discontinuing the use of these important medicines far outweighs the low risk that our scientists estimate to be associated with continuing the medicine until the patient’s doctor or pharmacist provides a safe replacement or a different treatment option,” wrote Gottlieb and Woodcock, both of whom are physicians. “Today’s news, of the certainty and broad number of nitrosamine-free ARB medicines, is another positive step. Health care practitioners should familiarize themselves with alternative medicines that can be used to treat hypertension, heart failure or renal disease in case of shortages,” the regulators wrote.

Experts in toxicology and chemical synthesis continue to study possible sources of contamination. Although FDA pointed to several GMP violations at the API manufacturing facility, some observers have linked possible contamination to improvements that were made in Zhejiang Huahai’s API manufacturing process, and noted that regulators may not have addressed contamination risk adequately in existing guidance.

Philippe André, GMP auditor with Qualandréin China, who had audited Zhejiang Huahai’s valsartan API facility in 2018, noted that there may be gaps in manufacturer’s scrutiny of processes that could generate genotoxic impurities, singling out levocarnitine, synthesized from a probable carcinogen, epichlorohydrin, as an example. Depending on how it is synthesized, he said, the compound may not only contain epichlorohydrin, but also traces of cyanide.

Although he noted that it would be difficult to predict whether a residue of unreacted epichlorohydrin might be carried over in the final product, he notes that compendial testing only considers chlorides, sulfates, sodium, and potassium. Even the Chinese manufacturing process for acetaminophen (a.k.a. paracetamol) could be a point of concern, André said, since one of the early intermediates is the probable carcinogen, 1-chloro-4-nitrobenzene. 

David Snodin, a consultant with Xiphora Pharma Consulting in Bristol UK and David Elder, a consultant in Hertford UK, dismiss the idea of carryover contamination in a March commentary in Regulatory Toxicology and Pharmacology, while acknowledging that “the knowledge of parts-per-million (ppm) chemistry is still very much in its infancy” (1).

However, they note that risk assessment efforts might have focused more closely on the potential presence of aliphatic secondary amines as solvent impurities, which could permit the formation of N-nitrosamine byproducts. They also noted that EU regulators have viewed the potential presence of benzene as a contaminant in other common solvents such as toluene, acetone and short chain alkyl alcohols, such as methanol and ethanol. 

Overall, they believe that global regulators have handled recalls and patient advisory appropriately. 

Currently ICH M7, published in 2014 and updated three years later, sets guidance for controlling mutagenic impurities in pharmaceutical syntheses. 

“ICH M7 is quite a sophisticated guideline which has over the last few years been built on in various ways, for example in relation to theoretical and/or experimental purge-factor assessment, compound-specific limits based on carcinogenicity potency data (e.g., TD₅₀or T25) or on data from in-vivo assays in transgenic rodents with genotoxicity endpoints (2), ”wrote Snodin in an email response to Pharmaceutical Technology.

 “Reactive/potentially toxic reagents and intermediates are employed in API synthesis to facilitate building the molecule,” Snodin continued. “But in many cases, these building blocks are readily purged and so are not present in toxicologically significant concentrations in the API. This concept is supported by the European Directory for the Quality of Medicine’s statement that 'alerting structures are not a cause for concern unless there is additional information'.”

 “Companies (API manufacturers and their customers) can address the risk by looking at the synthesis processes critically, looking up the toxicology databases to see if any of the materials and intermediates have any probable carcinogenic properties. It's a simple thing to do. Some do it, but most don't,” wrote Andréby email to Pharmaceutical Technology. “When they identify a potential impurity with carcinogenic properties, some customers require the addition of a suitable test in their specifications. Sometimes the API manufacturer takes the initiative. But again, most of the time, there is no such awareness.”

In addition, Andrénotes, the risk of formation of carry-over impurities is real, and should neither be minimized nor exaggerated. For example, he mentions the synthesis of levocarnitine starting from epichlorohydrin. “There is no test of residual epichlorohydrin in the pharmacopoeia monographs for this substance.” The European monograph was based on an old process using chloroethanol, he wrote. “Now, manufacturers use ethylene oxide. “I didn’t see a purge factor analysis of ethylene oxide in metronidazole,” he wrote.

“Often, when we audit generic API manufacturers in China, we ask them ‘Do you have an ICH M7 compliance program in place?’ Many have never heard of it,” wrote André.

Clearly, the industry can expect more discussion and debate on this topic in the coming months.

References

1. D. Snodin and D. Elder, Regulatory Toxicology and Pharmacology, 103(2019), p 325-329.
2. Müller, L. et al,  Toxicol Lett.  190(3), pp 317-29 (2009).