Page 89 - CRFTApplication Compendium_090613

3 Thermo Scienti c Poster Note • PN ASMS13_M011_KHassel_E 07/13S
iety of chemotherapeutic drugs
aration-liquid chromatography
ass spectrometer. The
uires less maintenance, and is
phy, LC/MS/MS,
ulfan, Docetaxel, Methotrexate
tem and TSQ mass
mplex sample preparation and
The use of liquid
S) to quantify
imatinib, and docetaxel) is
relude SPLC system in
nt consuming methods for
plasma . Sample preparation
that contained internal standard,
as removed and injected into
Thermo Scientific™
erformed with Thermo
olumns. For iminatib, busulfan,
n and Accucore PFP analytical
-C8 XL column and Accucore
.05% formic acid in water;
ethanol; and (C) 45/45/10
and methotrexate. Docetaxol
4 minutes or less and
onsumed less than 3 mL of
The Prelude SPLC system uses a two column
method for on-line clean-up, as shown in Figure
1. The first step (Figure 1A – loading step) is to
load the sample with aqueous mobile phase
onto the TurboFlow column under turbulent flow
conditions. Under turbulent flow conditions,
large molecule (>15 kDa) cannot interact with
the stationary phase and are washed to waste
while the analytes of interest are retained on
the column. Since the majority of matrix
interferences are from the matrix proteins, the
analyte of interest is removed from the matrix
during step 1. Once the sample is removed
from the matrix, the valves switch and the
TurboFlow column is back-flushed with organic
solvent stored in the loop of the first valve (filled
from the previous injection), which elutes the
analyte of interest from the TurboFlow column
to the analytical column (Figure 1B – transfer
step). In order to focus the analyte onto the
analytical column, the flow from the TurboFlow
column is teed to a valve with an aqueous flow
from a second pump. The separation step on
the analytical column provides gaussian
chromatographic peaks and further separates
any interferences. The elution step (Figure 1C)
switches the second valve so that the sample is
now eluted to the mass spectrometer and the
loop can be filled with the correct percent of
organic solvent for the next sample. Once these
steps are complete, the valves are returned to
the loading position where the columns can be
cleaned and equilibrated for the next sample
injection.
Compound name Method Range
Busulfan 20-200
Docetaxel 5-100
Imatinib 10-200
Methotrexate 10-75
TABLE 1. Method Range, Linearit
TABLE 2. Intraday Accuracy and
TABLE 3. Interday Accuracy and
Low QC
Busulfan 0.56-16.5
Docetaxel
Imatinib 1.0-9.5
Methotrexate 0.13-18.5
Intra
Compound name (%Differenc
Low QC
Busulfan 4.76
Docetaxel
Imatinib 11.00
Methotrexate 2.33
Inte
Compound name (%Differe
FIGURE 2. Standard Curves for E
System with TSQ MS
Compound name Bench Top St
Busulfan 104.2-12
Docetaxel
Imatinib 98.0-105
Methotrexate 102.4-10
TABLE 4. Bench Top Stability, Aut
0 500 1000 1500 0
5
10
15
20
25
30 Busulfan
Concentration (ng/mL)
Peak Area Ratio (Sample/IS)
0 200 400 600 0
5
10
15
20
25
Imatinib
Concentration (ng/mL)
Peak Area Ratio (Sample/IS)
Drugs
Methotrexate
usulfan
HPLC
pump 1
MS
waste
HPLC
pump 2
TFC column
Analyticalcolumn
plug
200 µ L loop
plug
HPLC
pump 1
MS
waste
HPLC
pump 2
TFC column
Analyticalcolumn
plug
200 µ L loop
plug
HPLC
pump 1
MS
waste
HPLC
pump 2
TFC column
Analyticalcolumn
plug
200 µ L loop
plug
(A)
(B)
(C)
Results
Accuracy and precision experiments were performed for system verification from three
separate preparations of calibrators and controls on three different days. Interday and intra-
day accuracy and precision results were obtained at concentration ranges of 20–2000 ng/mL
for busulfan, 10–2000 ng/mL for imatinib, 5–1000 ng/mL for docetaxel, and 1–750 ng/mL for
methotrexate. The method precision had RSD values less than 15% for all compounds
tested. Additionally, accuracy was 15% of the theoretical value for all the methods. The
correlation coefficient values for all the compounds ranged from 0.991 to 0.998, showing
linearity throughout all concentrations and analytes. All the analytes passed carryover, bench
top stability, autosampler stability, and specificity criterion. Recoveries, including matrix
effects, were all ~>90%. Data are summarized in Tables 1–4. Figure 2 depicts representative
standard curves for each compound tested. Representative chromatograms at the lower limit
of quantitation (LLOQ) for each compound are shown in Figure 3.
The improvement in run times resulting from the lower void volumes of the Prelude SPLC
system verses a conventional HPLC is illustrated in Figure 4 for Docetaxel. The same mobile
phases and columns were used for the comparison. When using on-line clean-up, the
duration of certain steps cannot be changed because they are dependent on the
chromatographic separation needed. The duration of other steps in the process are related
to how long it takes for solvent changes to reach the column. The sample clean-up and
sample elution steps are dependent on the chromatography, and therefore, the time for those
steps remain the same. However, the transfer, column cleaning and re-equilibration steps
can be reduced. On a conventional HPLC the transfer step was 75 seconds vs. 60 seconds
on the Prelude SPLC system. The column clean-up and equilibration steps were reduced
from 150 to 60 seconds. The result is a reduction in run time of 29% (5:15 minutes to 3:45
minutes). A shorter run time also reduced solvent consumption by 33%.
FIGURE 1. Valve positions for On-line
Sample Clean up and Analytical
Separation.
Mass Spectrometry
Detection of eluting analytes was done with a Thermo Scientific™ TSQ Vantage™ triple-
stage quadrupole mass spectrometer, equipped with a heated electrospray ionization
(HESI II) probe in positive ion mode using selected reaction monitoring (SRM).
Data Analysis
Quantitation was calculated with Thermo Scientific™ LCQUAN™ software.