Importance of Metabolite Testing in Regulated Bioanalysis

The use of bioanalytical methods to determine drug concentration levels in biological matrices has become more regulated with each passing year. Although the process of updating regulatory guidance and documents is often slow, there is no shortage of conferences where their representatives present their latest positions and expectations regarding method validation and use of the methods in routine analysis. It is also often after the discussions held at these conferences that the guidances are developed and accepted. An example is the US FDA Bioanalytical Method Validation Guidance [1], which was created after several meetings in Crystal City to discuss various tests required to determine accurate and rugged bioanalytical methods.

With each new conference report and guidance document, the industry is forced to adapt their outlook and procedures on various ‘hot topics’ in order to meet the regulatory and industry expectations. In February 2008, the FDA’s Guidance for Industry – Safety Testing of Drug Metabolites (more popularly known as the MIST Guidance) was finalized [2]. This document discusses the expectations of the FDA for a tiered approach to testing metabolites during various stages of drug development. This guidance does not address how to test for the metabolites; however, it succeeded in drawing the industry’s attention toward the presence of metabolites in bioanalytical methods. A contract research organization is now questioned much more frequently regarding the impact of nonquantified metabolites on the quantitation of the analyte(s) of interest.

It is interesting to note, however, that, despite the newfound emphasis placed on metabolite testing, the requirement is by no means new. All the guidance documents addressing method validation (e.g., from the FDA [1], Health Canada-Therapeutic Products Directorate [3] and Brazil-ANVISA [4]) discuss the need for a method to be specific and selective for the analytes of interest. The most explicit discussion is perhaps in the current FDA guidance, which states “selectivity is the ability of an analytical method to differentiate and quantify the analyte in the presence of other components in the sample... Potential interfering substances in a biological matrix include endogenous matrix components, metabolites, decomposition products, and in the actual study, concomitant medication and other exogenous xenobiotics” [1].

In bioanalytical methods for the analysis of small molecules, where MS/MS is the most common mode of detection, metabolic interference on quantifiable analytes can impact at three different stages: ex vivo, in vitro and insource/interface (in the mass spectrometer). The ex vivo impact of certain metabolites is due to their varying stabilities when in blood. Metabolites present in vitro, or after blood processing into plasma, can also have an impact depending on their stability in plasma and during the extraction of the analyte(s) from study samples. Finally, the insource/interface impact of metabolites takes place during the injection of study samples. The potential impact at each stage needs to be investigated and different solutions need to be applied to solve the various problems that can occur.

The first logical step in method development is a thorough literature search to identify all the possible metabolites that could be present and to what degree they are expected. However, it should be noted that not all existing potential metabolites may be found in the literature, nor are their reported quantities always accurate to what you may experience during sample analysis, mostly due to the increased sensitivity of newer technologies. In the following paragraphs, some case studies are described on how metabolites can impact at the three stages outlined previously, and how the literature, while a necessary first step (when available), should not be used to blindly discount potential future issues with bioanalytical methods.

One common challenge involves acyl-glucuronide metabolites. Acyl-glucuronide metabolites are well known for their stability issues and can convert to their parent drugs ex vivo, in the MS/MS source/interface and/or in vitro. Therefore, additional tests must be performed during method development to determine extraction and injection conditions that are adequate to prevent or minimize the impact of this metabolite on the quantitation of the analyte of interest [5–7].

One example of this is a method developed for the determination of indomethacin in human plasma using MS/MS detection. A review of the literature indicated that the indomethacin acyl-glucuronide metabolite is not detectable in human plasma. However, tests run using incurred samples demonstrated its presence in significant quantities. In this case, although it was originally assumed that it would not be necessary to purchase the acyl-glucuronide metabolite and perform additional tests, verifying the validity of the literature data proved fortuitous; stability of indomethacin acyl-glucuronide was determined at various steps of the extraction and injection in order to prove that it did not convert into indomethacin. In order to achieve the desired stability, samples had to be extracted at a temperature of 4°C. Under these conditions, method validation was successfully performed with no impact from this metabolite.

Another acyl-glucuronide example involves the determination of ketoprofen in human plasma using MS/MS detection, using ketoprofen-D₃ as the internal standard. Not only was it necessary to determine stability of ketoprofen in the presence of ketoprofen acyl-glucuronide under various extraction conditions, but it was not possible to optimize the MS/MS insource/interface conversion, such that ketoprofen acyl-glucuronide would not convert into ketoprofen. It instead became necessary to chromatographically separate the analyte and metabolite to ensure selectivity of the method, since the observed conversion was significant when compared with the lower limit of quantitation. A further challenge was that the original literature search did not indicate the levels of the metabolite 2-(3-hydroxy[phenyl]methyl)propanoic acid present in blood. When incurred samples were tested, it was discovered that this metabolite significantly contributed isotopically to the internal standard, ketoprofen-D₃, causing a merging peak in the chromatogram. It was necessary to change the internal standard to ketoprofen-¹³CD₃ to correct this problem. This example illustrates that an exhaustive literature search does not always yield all the information one needs for the proper development of a method. The testing of incurred samples is neccessary in order to prove that the method is successful.

A final glucuronide example is the analysis of ebastine and carebastine. Ebastine was documented in the literature as being excreted as conjugated metabolites. There was no mention of which conjugated metabolites were present and, therefore, during method development, samples were scanned for the most plausible and potentially problematic metabolites; the ebastine and carebastine glucuronides. The carebastine acyl-glucuronide was present in subject samples; however, levels were too low to significantly affect the selectivity of the method. Therefore, no further testing for this metabolite was required.

Another type of metabolite that can challenge method developers is the lactone. Lactones have highly reactive chemical structures and are well known to convert to their hydroxyl acid forms in plasma and in solution.

During the development of the analysis of atorvastatin, 2-hyroxy-atorvastatin and 4-hydroxy-atorvastatin, the lactone forms of each of these compounds were found to be present in plasma samples in significant amounts. Therefore, in order to ensure the stability of the lactones during extraction, the pH had to be correctly adjusted. Furthermore, since the lactones are so sensitive, they were deemed unstable in the MS/MS source/interface. Thus, it was necessary to ensure their chromatographic separation from the analytes of interest. This was a particular challenge since it is necessary to make the methods efficient for processing thousands of study samples and it is therefore desirous to keep run times short. This is possible but challenging when trying to chromatographically separate three analytes, three lactones and the required internal standards.

One other way to stabilize a lactone is by adding a preservative to the plasma during sample processing at the clinical site. This was necessary, for example, for a method for rosuvastatin, whose lactone was so unstable in plasma that it back-converted significantly in only a few hours. The additional precaution of extracting samples at 4°C was also necessary.

Sulfate metabolites can also convert to the parent drug in the MS/MS source/interface, making the presence of this type of metabolite during method development an additional difficulty. For example, for an oxcarbazepine method, literature indicates that oxcarbazepine sulfate is supposed to be present in study samples at very low concentrations that would not typically affect the selectivity of the method. However, during analysis, the sulfate metabolite was present in significant quantities, which in turn potentially compromised selectivity of the parent drug due to its insource/interface conversion [8]. Therefore, it was necessary to adjust the chromatographic conditions to separate this metabolite and perform a partial revalidation.

Finally, despite all the literature searches and discussions with sponsors, occasionally surprises may arise when a bioanalytical method is first used for routine sample analysis. A method for the determination of fluvastatin in human plasma was developed and successfully validated. All known metabolites had been verified and accounted for and did not impact the precision, accuracy and ruggedness of the method. However, during the first use of this method for sample analysis, a peak appeared that merged with the fluvastatin peak. This peak behaved as a metabolite, in that it increased and then decreased over time, like a typical drug profile. However, the literature and sponsor had no previous knowledge of this metabolite. Since the specificity of the method was impacted, the chromatography was adjusted and the method was partially revalidated. Therefore, allowances should also be made for potential scientific discoveries.

This editorial article attempts to illustrate the importance that must be placed on metabolite testing in regulated bioanalysis. Metabolites can impact the selectivity of a bioanalytical method in many ways and at many points in the method. Stability and chromatographic interference of these metabolites must be investigated, even though the literature may indicate that a certain metabolite should have no impact. However, method detection is ever more sensitive and previously published methods may not have been able to sufficiently detect or properly quantify metabolites. Therefore, although these articles are a very important starting point, they should not be used to irrevocably prove the assumption that a potentially interfering metabolite will not interfere in the analysis of the analytes of interest. A good knowledge of the chemistry of the compounds is required to be able to determine if they can impact analysis based on their mass, the mass of potential product ions, and their stability in various biological matrices, solvents and in the MS/MS source/interface. It may be necessary to adjust extraction conditions, as illustrated previously, by adding a preservative to the samples or by changing the extraction temperature or pH. However, if there is still a potential that the nonquantified metabolite could impact the quantitation of the analytes of interest, then the next step would be to chromatographically separate them. The most important thing is to ensure that the method used for sample analysis is selective, because you can be sure that the regulatory agencies will be selective when deciding if your submission is acceptable.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.