Manufacturers Should Generate Own Elemental Impurity Data To Comply With ICH Q3D

Manufacturers should create internal databases on elemental impurities for product risk assessments and not depend on suppliers to provide this information, which is called for under ICH Q3D. So far risk assessments have shown that most products are significantly below elemental impurity thresholds. Pharmaceutical manufacturers must complete assessments of legacy drugs by Dec. 31.

Pharmaceutical manufacturers should not count on getting information from excipient or API suppliers to complete the risk assessments called for under the International Council for Harmonization’s Q3D guideline on elemental impurities. Instead they should rely on their own internal testing programs or the published literature, company officials told a recent meeting.

Manufacturers are also finding that most products are well below control thresholds. These observations were made at a Nov. 2-3 meeting on ICH Q3D implementation sponsored by the Product Quality Research Institute and USP in Rockville, Md. At the meeting, officials also described the magnitude of the work involved in conducting, in some cases, thousands of risk assessments for legacy products.

The ICH Q3D guideline for elemental impurities, published in December 2014, sets permitted dally exposure (PDE) limits for 24 elements. The guideline takes effect in January 2018 for legacy products in the US, the EU and Japan; it has been in effect since June 2016 for new drug products in the US and Europe. It calls on manufacturers to assess the impurity risks posed by excipients, manufacturing equipment, active pharmaceutical ingredients and container closure systems in the final drug product.

There are two approaches that can be used to conduct an evaluation to ensure that an elemental impurity does not exceed the threshold: a component approach and a drug product approach. Under the first approach, manufacturers would assess potential elemental impurities in each component of the drug product. Under the second approach manufacturers would assess the potential impurities in the final drug product.

ICH Q3D specifies that elements to be considered in the product risk assessment include published literature, supplier information and empirical data.

Manufacturers must complete risk assessments for the thousands of legacy drugs in their portfolios by the January 2018 deadline. At the meeting, officials from Novartis AG, Hoffmann-La Roche Inc., Teva Pharmaceutical Industries Ltd. and Apotex Inc. described the progress they have made in conducting risk assessments.

Difficulties Getting Information From Suppliers

Industry officials said one factor complicating implementation of ICH Q3D is getting good information from suppliers. Elisabeth Kovacs of Apotex Inc. said that the company sent out a questionnaire to API suppliers to see whether the 24 elemental impurities were present in their products. The company sent out over 400 letters, and received no feedback from 40% of suppliers, good feedback from 30%, and fair feedback from the other 30%.

Kovacs said that “as drug product manufacturers we find ourselves between a rock and a hard place. We have to comply with the regulatory expectations yet the API vendors are not providing us with information. ... Right now based on their feedback it appears that they are not aware or do not understand what is required.”

Industry representatives said that, in the absence of reliable supplier information, manufacturers must generate elemental impurity data internally.

Mark Schweitzer, the global head of analytical science and technology for Novartis and former rapporteur for the Q3D guideline, said that manufacturers have to generate their own data to supplement the information they are not getting from suppliers.

Novartis, for example, now has a rich body of internal data on drug components, something it did not have in 2009. “Back then we had limited knowledge. In 2009 there was not a whole lot of data on elemental impurities. We had little data but much fear.”

The Sands Have Shifted

Yet the sands have now shifted to the point where some sites are collecting more information than necessary. Striking a balance is key. “I work with some sites that do not have a limit to the data that they like to collect.”

Novartis has been conducting drug substance elemental impurity screening studies for 10 years, which Schweitzer said “is benefiting for our risk assessments. We have a rich history of what is in our drug substance.” To date, the company has screened over 100 drug substances for elemental impurities.

Novartis then expanded the testing to cover manufacturing equipment. The in-house assessment used and examined a wide range of reaction chemistries and pH profiles, solvent profiles and temperatures extremes and the company tested most product with over five lots. Novartis did the screening from development through commercialization.

Schweitzer said that “we can categorically say we don’t have any elemental impurities in the manufacturing equipment because of the drug substance testing we have done.”

2,300 Risks Assessments Available For Inspection

Schweitzer said that Novartis has been able to leverage this internal data to prepare over 2,300 individual risk assessments for products.

Yet the magnitude of the task of conducting these risk assessments is still daunting, he said.

“We have 67 manufacturing facilities in 58 countries and organized into one division. We have more than 10,000 excipients and drug substances and they come from multiple suppliers and multiple manufacturing sites, from small to large molecules and a wide range of products such as oral, parenteral, inhalation, topical and ophthalmic and combination products.”

Schweitzer said that “as the amount of data increases, our comfort increases as well. There is still a challenge in what constitutes an acceptable elemental impurity risk assessment."

Yet the jury is still out as FDA has not yet approved these risk assessments. He said that a significant challenge remains on what constitutes an acceptable elemental impurity product risk assessment.

“This is a work in progress. We will have more assessments to do. Hopefully we are participating in our first feedback from an inspection. … I really want to see if we are going down the right path. If not, we can course correct. It is only through an inspection that we are going to find what is acceptable.”

Hoffman Takes Platform Approach

Helmut Rockstroh of Hoffmann La Roche concurred that it is important for manufacturers to collect data internally and not rely on excipient or API suppliers to provide the data.

He said that “we realized we would be getting difficulties from our suppliers so we realized that we had to do our own excipient testing.” Rockstroh said that the company, while smaller than Novartis, does have a few very high-volume products.

The company has developed a platform approach which is a “worst-case combination of products per platform.” These platforms constitute different products, dosage strengths, compounding approaches, filling lines and equipment combinations, such as low pH, and a high elemental impurity burden for excipients. If these “worst-case” combinations do not exceed elemental impurity thresholds, there is no reason to believe that lower-risk combinations would exceed impurity thresholds.

Rockstroh said that firms should also leverage the information from published literature in their risk assessments.

For example, to ensure that the materials used in container closures are not leaching into products, manufacturers should consult the published literature instead of doing their own testing. For example, a recent paper published by container closure expert Dennis Jenke shows that the probability of elemental impurity leaching into solid dosage forms is minimal and does not require any further consideration and no further testing.

Rockstroh said that the worst thing to do is to test for the sake of testing. For example, it does not make sense to test steel containers for cobalt and palladium since the risk of leaching is low. “My QC people would say ‘oh no let’s do it. We have to do it.’ Yet everything that you do extra will hurt you down the road if you have a million other tests to do.”

Novartis, Teva See Little Risk Of Elemental Impurities

Yet after all this work in conducing the assessments the results are showing that a negligible number of products are exceeding elemental impurity thresholds.

Schweitzer said that of the 2,300 risk assessments, there is no significant elemental impurity contribution to drug products. “From our database there is nothing there 99.9% of the time so we are doing a whole lot of testing for nothing but this was better than where we were first in it in 2009. This is a good thing.”

Orit Schwartz of Teva Pharmaceuticals said Teva also is finding a negligible risk of elemental impurities in drug products as well. Teva has conducted over 2,500 risk assessments. It has submitted over 90 US NDA and ANDA submissions and 40 EU new drug submissions since June 2016.

Teva's analysis showed that 1.4% of the products had elemental impurities that exceeded the 30% threshold under either finished product testing or component testing.

From the editors of The Gold Sheet.