An Alternative Solution to High Blank Problems Anne Schmedes (Presenter) Department of Clinical Biochemistry >> POSTER (PDF)

Authors: Anne Schmedes, Christoffer Kleve Røndbjerg, Hashmatullah Nasimi
Department of Clinical Biochemistry, Lillebaelt Hospital, Vejle, Denmark

When setting up new analysis for drugs or endogenous compounds a gradient separation on a UPLC column is often used before tandem mass spectrometry. However on two occasions we have experienced that it can be impossible to have clean blank samples when running a gradient. Two examples will be shown: 1) analysis of Caffeine and 2) analysis of Erlotinib. For both these compounds we experienced that mobile phases inevitably contain the compounds or similar compounds. The only way to obtain acceptable blank samples was to use isocratic elution.

Problem

Problem:

When setting up new analysis for drugs or endogenous compounds a gradient separation on a UPLC column is often used before tandem mass spectrometry. In two occasions we experienced problems with very high blank samples: 1) analysis of Caffeine and 2) analysis of Erlotinib.

Method Information

Method information:

Caffeine:

Sample preparation: Protein precipitation 100 µL serum plus 200 µL methanol with 13C3 Caffeine (1250 µg/L) as internal standard. Mixed, centrifuged, 10 µL supernatant diluted with 800 µL water

LC-MS/MS system: Waters Acquity UPLC with Waters Xevo TQ-S tandem mass spectrometer, ESI

Mobile phase A: 2 mM ammonium acetate and 0.1% formic acid in MilliQ water

Mobile phase B: 2 mM ammonium acetate and 0.1% formic acid in Methanol LC-MS grade

5 min gradient: 0-1.0 min 5% B, 1.0-3.0 min increase to 95% B, 3.0-4.0 min 95% B, 4.0-5.0 min 5% B, flow 0,5 mL/min

Column: Waters Acquity UPLC HSS C18 1.8 µm, 100 x 2 mm

Column oven 35°C

Injection volume 10 µL

MRM transitions: Caffeine 195.03>138.03, 13C3 Caffeine 198.03>140.03

Erlotinib:

Sample preparation: Protein precipitation 100 µL serum plus 300 µL acetonitrile with 13C6 Erlotinib (2.5 µmol/L) as internal standard. Mixed, centrifuged, 20 µL supernatant diluted with 980 µL water

LC-MS/MS system: Waters Acquity UPLC with Waters Xevo TQ-S tandem mass spectrometer, ESI

Mobile phase A: 0.1% formic acid in MilliQ water

Mobile phase B: Methanol LC-MS grade

5 min gradient: 0-0.5 min 20% B, 0.5-3.5 min increase to 100% B, 3.5-4.5 min 100% B, 4.5-6 min 20% B, flow 0,5 mL/min

Column: Phenomenex Kinetex 2.6 µm, 100 x 2 mm

Column oven 40°C

Injection volume 2 µL

MRM transitions: Erlotinib 394.09>278.21, 13C6 Erlotinib 400.09>284.21

Troubleshooting Steps

Steps taken:

Before realizing that high peaks in blank samples were present even without any injection we performed the usual trouble shooting steps:

Increase the needle wash

Check if the high peak in blank sample was caused by impurities in the internal standard

Cycle the inject valve several times between the runs

Replace all solvents including testing a new lot number of methanol

Changing the gradient in different ways

Try another reverse phase column and a Hilic column

Try using another fragment

Check if the high peak in blank sample was dependent on injection volume

Check if the high peak in blank samples decreased in height when running several blanks after each other

Finally, since the unwanted peak was present even though no injection was performed and on every column tested, we chose to use isocratic separation. In this way the high peak in blank disappeared since there was no equilibration phase where the compound could build up on the column.

Outcome

Outcome:

Analysing caffeine or erlotinib in serum is quite simple to perform since the concentrations found in serum samples are always far above the limit of quantitation (LOQ). Therefore isocratic elution could be used although this affects the sharpness of the peaks and hence the LOQ. A further advantage of isocratic elution is very short run times of only 2 minutes.

When experiencing problems with high blank samples in gradient methods the trouble shooting process should always start by investigating if a gradient run with no injection shows a high blank. In that case the investigation can focus on the solvents, column and fragments and skip all autosampler issues.