Purity Analysis Series, Part 1

Identifying and Addressing Unexpected Peaks in Chromatographic Testing

The one area of analytical testing that cause the most anxiety is purity analysis or the determination of impurities/related compounds in the material of interest, either drug substance or drug product. The methods are often chromatographic methods, such as HPLC (high-performance liquid chromatography), UPLC (ultra-performance liquid chromatography) or GC (gas chromatography). A new peak observed during the analysis of purity/related compounds is sure to raise questions, but it does not need to be a show-stopping event. Take a breath and let’s review the next steps. In this article, we will consider the initial steps to determine if the peak is real and not an immediate artifact of the test procedure.

Step 1 is to understand what samples contain the new peak. Is the new peak observed in the blank or diluent injection? If yes, then the source of the peak is related to the analytical system (instrument or mobile phase preparation).  Ensure that your system suitability requirements contain a check for the absence of unknown peaks.  Make sure your instrument is performing as intent.  Use clean glassware and fresh solvents to prepare new mobile phase. Repeat the analysis. If the peak remains, the source is identified, and an investigation is required to determine the actual root cause.  The investigation should evaluate different instruments, lots of mobile phase reagents, and different grade or supplier of the mobile reagents and potentially a different HPLC column.

If the blank/diluent injection are clean but the new peak is observed in the standard injections as well as the sample injections, ask the question- what changed related to the standard preparation. 

Some questions to consider during analytical testing are:

• Is the solution used to prepare the standard equivalent to the previous time when the new peak was not observed?  Consider pH, organic concentration, buffer concentration, age of solution used.

• What data do you have on the stability standard solution?  Is this observation consistent with the available data?

• How old are the standard at the time of the analysis? 

• How are the standards stored? 

• Are the standards protected from light?

Review the injections of the standard throughout the analysis. Are there any trends in the level of the new impurity? Conduct a forced degradation study on the prepared standard to see if the new peak is formed and under what conditions.

Step 2 is to determine if the peak is truly present in the sample by conducting confirmatory testing.  Confirmatory testing consists of re-injecting the original sample vial and preparing a second sample from the original sample stock preparation (referred to as re-vialed sample).  If the sample was injected neat (without dilution), take a second sample from the sample container as the original sample. The re-injection of the original sample and the re-vialed sample should be conducted within the stability of the sample solution and use with the same standards, mobile phase and instrument. Ideally, the analysis should be conducted on the same day as the new peak was seen. No need to wait for the data to be checked. Most companies build confirmation testing into their SOP on investigation of out-of-trend/out-of- specification laboratory results. This simple exercise of performing confirmatory testing can provide valuable information.

If the peak of interest is absent in both the reinjected original sample vial and re-vialed sample, the presence of the peak in the original sample was the result of an instrumental artifact and not related to the sample.  Ensure that the instrument is functioning correctly and there are no leaks and check to see that all HPLC vials are properly capped.  Remember that instruments are not infallible and sometimes bad injections occur and strange things happen; the true source of the artifact may never be known. 

If the peak is present in the re-injected original sample but absent in the re-vialed sample, the source of the peak is related to the preparation of the sample for analysis.  Something changed during the preparation of the second sample from the original stock solution. The most likely explanation is contamination.  Contaminated HPLC vials, dirty glassware and/or sloppy laboratory practices are likely contributors. When considering sloppy laboratory practices, be open-minded. Inadvertent contact of the sample solution which is highly organic with the analyst’s gloves may result in the extraction of a leachable from the glove material. Also carefully review the disposable components used to prepare the original and re-vialed sample noting any difference in material suppliers or material lot numbers. Depending on the sensitivity and nature of the assay, these seemly insignificant changes could be the related to source of the unknown peak. Once encountered a situation where the preferred syringe used to filter the sample was not readily available and the analyst grabbed a different syringe to perform the filtration. The method only specified the use of 10‑mL syringe. The alternative syringe was not equivalent to the one commonly used and contained a black rubber gasket on the syringe barrel. The black rubber gasket was not compatible with diluting solution and a new leachable peak was observed in the sample chromatogram. As a result of this finding, the method is now updated to include the exact syringe to use for filtration of the sample.

If the peak is present in both the re-injected original sample and the re-vialed sample, the present of the peak is confirmed, and the peak is considered real.  At this point, it is not possible to the determine that peak is truly in the samples analyzed.  The source of the peak may still be related to the sample preparation. One has only confirmed that the sample preparation steps are performed consistently.  A more comprehensive investigation is required to determine if the root cause is related to the laboratory or the actual test sample itself.  Another aspect to consider at this point is do you want to determine the actual chemical identity of the new peak.  The chemical identity can be determined by specialized analytical laboratories and will require additional planning as the comprehensive investigation plan is developed and outlined for execution.  Determination of the chemical composition often requires larger quantities of test solution than the standard analytical analysis as the new peak will need to be separated and concentrated for analysis by orthogonal methods.

The critical considerations and design of the comprehensive investigation will be the subject of a subsequential article.

Written By: Bette Monnot-Chase