Inconsistent Performance of Two Similar Instruments during Method Transfer: Towards Implementation of Apolipoprotein Quantification in Clinical Routine Mervin Pieterse (Presenter) LUMC Leiden

Authors: M.M. Pieterse, L.R. Ruhaak, F.P.H.T.M. Romijn, N.P.M. Smit, Y.E.M. van der Burgt, C.M. Cobbaert

In our laboratory we have two LC-MS systems which run both protein and small molecule analysis. The aim is to develop methods on the development instrument, which are, upon analytical validation, transferred to the routine instrument, to minimize interference from method development on the routine results. The two systems are from the same manufacturer and have the same type of MS. The type of LC pump and LC autosampler are different. While the carry-over was below 1% for all peptides monitored in the multiplexed apolipoprotein method [1] on the development system, the carry-over was consistently above 1% for several peptides on the routine instrument. Using a systematic approach, we were able to determine that the problem was located in the autosampler, and caused by incompatibility of a solvent system used for one of the other measurands being implemented on the routine instrument.

Problem

In our laboratory we have two LC-MS systems. One for method development and one for routine methods. The routine LC-MS system is equipped with a multicolumn compartment to enable running of multiple applications with the same instrument setup. In practice, this allows us to run proteomics and small molecule applications on the same instrument serially. The aim is to develop methods on the development instrument, which are, upon analytical validation, transferred to the routine instrument, to minimize interference from method development on the routine results. The two systems are from the same brand. They have the same type of MS, but the type of pump and autosampler are different. We previously developed and analytically validated the multiplexed quantitation of apolipoproteins using LC-MRM-MS, and are now transferring this method to the routine instrument for clinical implementation.

To ensure the LC-MS instrument functions well, a system suitability procedure has been developed for the apolipoprotein application, and a system suitability sample, containing both endogenous and stable isotope labelled synthetic peptides is measured to asses – among others- carry-over. Carry- over was below 1% for all peptides measured on the development system, but, upon transfer of the application to the routine system, carry over was over 1% for several peptides.

Method Information

The research LC-MS system consists of the following components:

1. Agilent 1290 infinity I Binary Pump

2. Agilent 1290 autosampler infinity I

3. Agilent 1200 Temperature column compartment

4. Agilent 6495 ifunnel QQQ

The routine LC-MS system consists of the following components:

1. Agilent 1290 infinity II High Speed Pump

2. Agilent 1290 infinity II Multisampler

3. Agilent 1290 infinity II multicolumn thermostat

4. Agilent 6495 ifunnel QQQ

A system suitability sample is used to assess whether the instruments perform well, and are suitable to provide accurate analytical results. To this end, a mixture of both endogenous and stable isotope labelled synthetic peptides (both 0.15 µmol/L) is measured 5x prior to the measurement of clinical samples, and 5x after the measurement of clinical samples. A blanc sample is measured after each 5 system suitability repeats to assess carry-over.

Troubleshooting Steps

A systematic approach was used to determine the location of the carry-over by

1. Exchanging the analytical column between the discovery and routine instruments

2. Bypassing all switching valves

3. Bypassing the autosampler

Using this strategy, the autosampler could clearly be identified to cause the carryover.

Several steps were then performed to solve the carryover. First, the composition of the needle wash of the autosampler varied, but this did not resolve the carryover. Next, we bypassed all the parts of the flowpath inside the autosampler, but none of the individual parts resolved the carry-over, and even after exchange of most replaceable parts of the autosampler by the manufacturer, the carryover was consistent.

It was then brought to our attention that another method for a different analyte was being implemented on the routine system at the same time. Once the implementation of this analyte was discontinued and the instrument was flushed for 48 hours, the carry-over of the apolipoprotein peptides was below the threshold.

Outcome

The carry-over was confirmed to be located in the autosampler of the routine instrument, which is of a different type compared to the autosampler of the discovery instrument. Because two methods were being implemented on the routine instrument at the same time, we did not realize one of them was interfering with the implementation of the apolipoprotein quantitation, causing severe carry-over in the autosampler. The troubles observed could have been avoided with a better implementation scheme, where only one test is being implemented at the same time. Alternatively, method development and routine analysis could be performed on the same system. This will overcome the problem of transferring/implementing a method onto a different system but might result in down time of a routine instrument when difficulties arise during method development. Implementation of system suitability tests for each individual method allows identification of potential interferences between methods at an early stage, and, if the system suitability tests are passed, ensures that proteomics and small molecule applications can safely be run serially. Overall, the practical approach towards finding and locating problems within an LC-MS system used here illustrates the need of the implementation of a system suitability test.

[1] van den Broek, I., Romijn, F. P., Nouta, J., van der Laarse, A., Drijfhout, J. W., Smit, N. P., van der Burgt, Y. E., and Cobbaert, C. M. (2016) Automated Multiplex LC-MS/MS Assay for Quantifying Serum Apolipoproteins A-I, B, C-I, C-II, C-III, and E with Qualitative Apolipoprotein E Phenotyping, Clin Chem 62, 188-197.