Journal of Chromatography B, 962 (2014) 147152

Contents lists available at ScienceDirect

Journal of Chromatography B
journal homepage: www.elsevier.com/locate/chromb

Determination of irinotecan and its metabolite SN-38 in rabbit plasma

and tumors using a validated method of tandem mass spectrometry
coupled with liquid chromatography
Dae J. Park a, , Jun H. Won a , A.R. Cho a , Hye J. Yun a , Jeong H. Heo b , Tae H. Hwhang c ,
Dae H. Lee a , Woo M. Kim a
a
Department of Clinical Pharmacology, College of Medicine, Kosin University, Busan 602-703, Republic of Korea
b
Department of Molecular Biology and Immunology, College of Medicine, Kosin University, Busan 602-703, Republic of Korea
c
Department of Clinical Pharmacology, Pusan National University, Busan, 603-729, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: New tandem mass spectrometric method coupled with liquid chromatography (LCMS/MS) has been
Received 11 January 2014 developed to determine the total concentration of camptothecin derivatives (irinotecan and SN-38)
Accepted 17 May 2014 regardless of inter-conversion phenomenon between carboxylate and lactone forms. At rst, all sam-
Available online 27 May 2014
ple solutions were acidied for 1 h in order to completely convert CPT derivatives into their lactone
forms and then CPT derivatives were extracted with organic solution containing diethyl ether and ethyl
Keywords:
acetate (2:1, v/v) just after alkalization in the range pH 8.08.5 in acid-treated solutions. Analytes were
Irinotecan
separated on a reverse phase C18 column (150 2.1 mm) and eluted isocratically with a mobile phase
SN-38
Camptothecin
which consisted of acetonitrile-methanol-buffer (0.1% formic acid, 5 mM ammonium formate) (3:4:3,
Rabbit v/v). CPT derivatives were monitored by tandem mass spectrometry in electrospay-positive ionization
VX2 tumor and multiple reaction mode programmed to the following transitions (m/z): 587.6 167.2 of CPT-11,
LCMS/MS 393.6 349.3 of SN-38 and 349.4 305.2 of CPT. The method was validated to have the proper lin-
earity (r2 > 0.99) over the range of 51000 ng/ml of CPT-11 and 1250 ng/ml of SN-38 with good accuracy
(89.8114.3%) and precision (less than 10%). In all stability tests, concentration of CPT-11 and SN-38 had
been left in the acceptable range of 88.8110.7% when sample solutions were acidied before determi-
nation of CPT derivatives. Newly developed LCMS/MS method was suitable for the determination of CPT
derivatives of both rabbit plasma and tumor tissues in the pharmacokinetic study.
2014 Elsevier B.V. All rights reserved.

1. Introduction DNA topoisomerase I inhibitor that binds to the covalent complex

Irinotecan (CPT-11) and SN-38 are pentacyclic derivatives of
camptothecin isolated from the Chinese tree, Camptotheca acumi-
nata, in 1960s [1]. They have shown strong antitumor activity
against a wide variety of tumors including colorectal and gastric
cancer [2,3]. CPT-11 is converted to its active metabolite, SN-
38 (7-ethyl-10-hydroxycamptothecin) mainly by carboxyl esterase
(Fig. 1), which has 1001000 times greater cytotoxic activity than
the parent, CPT-11 [4,5]. SN-38 carry out its anticancer effect as
Abbreviations: CPT, camptothecin; CPT-11, irinotecan; CV, coefcient of vari-
ation; IS, internal standard; LLE, liquid-liquid extraction; MRM, multiple reaction
monitoring; SD, standard deviation.
Corresponding author. Tel.: +82 51 990 6488; fax: +82 51 990 3081.
E-mail address: paaranvit@kosin.ac.kr (D.J. Park).
between topoisomerase I and the ligated DNA strand. A jammed
complex with SN-38 can prevent DNA replication of tumor cells,
thereby causing death of cancer cells [6]. Intact lactone ring of CPT
derivative is required to exert cytotoxicity. However, the lactone
ring undergoes pH-dependent, reversible spontaneous hydrolysis
to produce a predominant carboxylate form at physiologic pH,
which is an important point to be considered for determining
CPT derivatives in a biological matrix [7]. For that reason, HPLC-
uorescence detection together with simple protein precipitation
was the most frequent method because most of CPT derivatives
could be still remained in the simply pretreated samples and both
lactone and carboxylate forms were able to be detected by u-
orescence due to their similar excitation/emission spectrum [8].
However, such methods are still insufcient to assay the dirty
samples like as tissues with delicate assay equipment of a tan-
dem mass spectrometry coupled with HPLC [9]. In the present

http://dx.doi.org/10.1016/j.jchromb.2014.05.042
1570-0232/ 2014 Elsevier B.V. All rights reserved.
148 D.J. Park et al. / J. Chromatogr. B 962 (2014) 147152
Fig. 1. Structures and inter-conversion phenomenon of camptothecin derivatives (CPT-11, SN-38 and CPT).
study, we developed a tandem mass spectrometry coupled with
HPLC to determine total concentrations of each CPT derivatives,
regardless of the inter-conversion phenomenon between the lac-
tone and carboxylate forms and also applied to the pharmacokinetic
study in tissues by using LCMS/MS with a preparation method of
liquidliquid extraction (LLE).
2. Experimental
2.1. Materials and reagents
CPT-11, SN-38 and CPT were purchased from Sigma (St. Louis,
MO, USA). Ammonium formate, formic acid and modied Hanks
balanced salt solution (HBSS without Ca2+ and Mg2+ ) were also
purchased from Sigma. Acetonitrile, methanol, diethyl ether and
ethyl acetate were obtained from Merck (Darmstadt, Germany). All
solvents had high analytical grade for HPLC or mass spectrome-
try. Water was deionized by a Milli-Q water purication system
of Millipore (Bedford, MA, USA). Camptosar (Irinotecan hydrochlo-
ride injection, CPT-11 40 mg/ml) was purchased from Pzer (Pearl
River, NY, USA). Zoletil 50 was purchased from Virbac (Carros
Cedex, France) and rompun was purchased from Bayer (AG, Lever-
kusen, Germany). Twelve New Zealand white rabbits were from
DBL (Chungbuk, Korea). All animal procedures were approved by
the Kosin Institutional Animal Care and Use Committee.
2.2. Preparation of standards, quality controls and samples
solutions
Stock solutions of CPT-11, SN-38 and CPT were prepared with
100% methanol to the concentrations of 0.2 mg/ml and were kept at
70 C. Working solutions (10-fold standard solutions) were pre-
pared by serial dilutions of stock solutions with 80% methanol at the
following concentrations: 50, 100, 500, 1000, 5000, 10,000 ng/ml
of CPT-11; 10, 20, 100, 500, 1000, 2500 ng/ml of SN-38. Working
solution of CPT (internal standard) was prepared at one level of
1000 ng/ml. The standard and quality control solutions (QCs) were
prepared by mixing 200 l of blank plasma or tissue extracts (25%,
w/v) with each 20 l working solutions of CPT-11, SN-38 and CPT
in 2-ml mini-eppendorf tubes. QCs were prepared at the follow-
ing concentrations: 5, 100 and 1000 ng/ml for CPT-11; 1, 50 and
250 ng/ml for SN-38. All sample solutions of rabbit plasma or tissue
extracts were also added with 20 l working solutions of CPT.
2.3. Sample preparation
We used three sample types of rabbit (plasma; two tissue
extracts of muscles and VX2 tumors). When the determination
process started, all solutions (220260 l) of standard curve, QCs,
plasma and tissue extracts were acidied with 40 l of 2 M formic
acid and incubated for 1 h at room temperature, resulting in the
transformation of all CPT derivatives into lactone forms. After 1-
h incubation, all of them were alkalinized by adding 80 l of 2 M
ammonium hydroxide in order to adjust acidity in the range of
pH 8.08.5 just before mixing organic solutions for the extraction
of CPT derivatives. Alkalized solutions were mixed with 1.2 ml of
extraction solutions consisted of diethyl ether and ethyl acetate
in ratio of 2:1 (v/v). The mixture was vortex-mixed for 10 min
and centrifuged at 13,500 rpm for 5 min. A 1 ml of the upper layer
(lipophilic) was transferred to another tube and evaporated to dry-
ness in a Centrivap mobile system (Labconco, Kansas, MO, USA) for
60 min at 45 C on centrifugal negative pressure. The residue was
reconstituted with a 100 l of mobile phase and ltered through
a 0.22 m PVDF lter attached in 1-ml syringe (Norm-ject). Fil-
tered solution was collected in 300-l insert equipped in a 2-ml
polypropylene vial. A 5 l of the ltrate was injected to the column
from the autosampler of HPLC system.
2.4. Tissue homogenization
VX2 tumors and femur muscles of the opposite side were
excised from the sacriced rabbits, pieced into small sections and
stored frozen at 70 C in polypropylene tubes until homogeniza-
tion. The frozen tissues were thawed, chopped and weighed at
4 C. More or less 1 g of chopped tissue was homogenized in four
volumes of ice-cold homogenizing solutions containing HBSS and
acetonitrile (3/1, v/v) with a polytron tissue homogenizer [10].
Homogenization was paused 30 s after every 30 s of homogeniza-
tion at a medium speed. The homogenization was repeated 34
times until a uniform homogenate was obtained. The whole pro-
cess was nished by moving the supernatant to a new tube after
tissue homogenate was centrifuged at 15,000 rpm for 5 min. Final
tissue extract was stored at approximately 20 C prior to analy-
sis. The homogenizer probe was washed sequentially with water,
methanol and water after every homogenization.
2.5. LCMS/MS conditions
The LCMS/MS system consisted of Quattro PremierTM XE cou-
pled with Alliance HPLC (2795XE separations module) made from
 D.J. Park et al. / J. Chromatogr. B 962 (2014) 147152
Table 1
Parameters and transitions in the determination method for CPT derivatives.
CPT derivatives Retention time (min)
CPT-11 1.8
SN-38 3.0
CPT 3.1
Waters (Milford, NJ, USA). The temperature was set at 10 C in HPLC
auto-sampler and at 35 C in separating column, a reverse-phase
Waters XTerra MS C18 column (150 2.1 mm) being packed with
the stationary phase of 3.5 m diameter particles. The mobile phase
was eluted in an isocratic manner at the ow rate of 0.2 ml/min,
which consisted of acetonitrile-methanol-buffer (0.1% formic acid
and 5 mM ammonium formate) (3:4:3, v/v). After the samples
passed through column, they was introduced into the interface of
an electrospray probe equipped in a triple quadrupole mass spec-
trometric detector, operating in electrospray-positive ionization
mode (ES+ ). CPT-11, SN-38 and CPT were simultaneously deter-
mined with the multiple reaction mode (MRM) which was set as
shown in Table 1. The running times were 7 min per each sample
and this system was controlled by Waters MassLynx v4.1.
2.6. Extraction efciencies and method validation
The condition of LLE was optimized in comparison with extrac-
tion efciencies of each CPT derivatives according to the pH
variation during the procedure of sample acidication and alka-
lization. Extraction efciencies were calculated by comparing peak
areas of each CPT derivatives extracted by LLE in biological matrix
with those of pure CPT derivatives in a mobile phase at each QC
levels.
The limit of detection (LOD) and lower limit of quantication
(LLOQ) were selected as the concentrations that have signal to noise
ratio (S/N) of 3 and 10, respectively. LLOQ were also obtained as
the lowest standard concentration of the calibration curve when
having an acceptable precision of below 20% of the true concentra-
tion and an accuracy within the range of 80120%.
All concentrations of calibration curves and QCs were calculated
by using a weighted (1/x2 ) least-squares linear regression analy-
sis with the peak area ratios of CPT-11 or SN-38 to CPT. Accuracy
and precision were validated by assessing ve copies of each QCs
levels (5, 100 and 1000 ng/ml for CPT-11; 1, 50 and 250 ng/ml for
SN-38) within 1 day (Within-day assay) and also between ve con-
secutive days (Between-day assay). Precision was reported as the
coefcient of variation (CV, %), the percentage of the standard devi-
ation to mean concentration at each levels of QCs. Accuracy was
described as the percentage of calculated mean to the QCs nominal
concentration. The values of CV should be below 15% as an accep-
tance criteria for precision, while the values of accuracy be set in
the range of 100 15% relative to the nominal concentrations.
2.7. Stabilities
All stabilities were determined by using three copies of the low
and high QCs (5 and 1000 ng/ml for CPT-11; 1 and 250 ng/ml for SN-
38). The short-term stability was assessed by determining the levels
of CPT-11 and SN-38 in plasma or tissues extracts according to the
times for which QCs had been kept at room temperature without
specic treatment. The post-preparative stability was estimated by
determining their levels after CPT derivatives had been completely
extracted and then kept in the auto-sampler of HPLC for 6 h at 10 C.
The long-term stability of QCs was tested after 31-days storage at
20 C and the calculated concentration was compared with the
nominal QC concentration.
149
Cone voltage (V) Collision energy (eV) Transition (m/z)
32 34 587.6 167.2
30 28 393.6 349.3
22 26 349.4 305.2
2.8. Assay application
VX2 tumors (Samtako, Oh San, South Korea) were grown
in femur muscles of 12 New Zealand white rabbits (weight:
2.63.2 kg). Tumors were established with oval shape having the
mean diameter of 35 cm after 20 days since implantation [11]. All
rabbits were intravenously given with CamptosarTM (CPT-11) via a
marginal ear vein at a dose of 6 mg/kg within 30 s after anesthesia
with a mixture of Zoletil and rompun. Each four rabbits were sac-
riced at the times of 4, 12 and 24 h after injection of CPT-11 and
immediately dissected to obtain tumors and normal femur muscles.
Before they were sacriced, blood samples (0.5 ml) were collected
from the marginal ear vein of available rabbits into heparinized 2-
ml mini-eppendorf tubes at 0, 5, 15, 30, 45 min and 1, 2, 4, 6, 8, 12,
24 h since CPT-11 injection.
3. Results
3.1. Assay development and sample preparation
The transitions of CPT-11, SN-38 and CPT were congured by
the optimized instrument parameters as shown in Table 1. The run
time of CPT-11 was shorter than those of SN-38 and CPT because
of its hydrophilicity (1.8 vs 3.0 min). The peak resolutions had a
tendency to be improved when methanol content was increased
in a mobile phase (data not shown). The great impediment of
assay development was the stability issue of CPT derivatives in
biological matrix because of the inter-conversion phenomenon
occurring between lactone and carboxylate forms. CPT derivatives
were mostly changed to the carboxylate forms in plasma or tis-
sue extracts. The results had shown that their lactone forms had
remained of only 2030% within 3 h when samples were not pre-
treated with acidication (data not shown). The acidication of
various samples was able to be carried out by adding with 2 M
formic acid (nal: 0.3 0.03 M) to the acidity range of about pH
2.53.0 in plasma or tissue extracts immediately after obtaining
plasma and tissue homogenates. All of the acidied solutions were
weakly alkalinized in the range of pH 8.08.5 with 2 M ammo-
nium hydroxide just before LLE because the extraction efciency
of CPT-11 was the largest in the range of weak-alkali but those
of SN-38 and CPT were relatively high in the range from strong
acid to weak alkali (Fig. 2). In conclusion, extraction efciencies
Fig. 2. Liquid-liquid extraction efciencies proles of camptothecin derivatives
(1000 ng/ml) according to the pH variation of rabbit plasma.
150 D.J. Park et al. / J. Chromatogr. B 962 (2014) 147152

Fig. 3. Chromatograms of camptothecin derivatives in plasma and VX2 tumors of rabbits obtained 12 h after injection of CPT-11 (6 mg/kg). The levels of camptothecin
derivatives were determined as 5.3 ng/ml for CPT-11 and less than LLOQ for SN-38 in plasma (A). Their concentrations are 22.3 ng/ml and 6.5 ng/ml for CPT-11 and SN-38,
respectively, in tumors (B). Any interfering peaks were not detected at the specic run times of CPT-11 (1.8 min), SN-38 (3.0 min) and CPT (3.1 min) in blank samples.

distributed within the range of 4070% (CPT-11:44.048.4%; SN-
38:57.768.1%; CPT:51.056.4%).
3.2. LOD, LLOQ, linearity and selectivity
It was found that CPT-11 had the LOD of 1 ng/ml (S/N = 3.7) and
the LLOQ of 5 ng/ml. In the case of SN-38, the LOD was determined
as 0.2 ng/ml (S/N = 4.8) and the LLOQ was xed at 1 ng/ml. The
retention time of CPT-11 was about 1.8 min. SN-38 had a similar
retention time with CPT (3.0 min).
Several chromatograms for CPT derivatives in plasma and tumor
extracts were shown in Fig. 3. There were no interfering peaks of
biological ingredients in the chromatograms of bank plasma and
tumor extracts. All of the standard curves were well established
and had the correlation coefcients of r2 0.995. Their ranges of
linearity were 51000 ng/ml for CPT-11 and 1250 ng/ml for SN-38
in both plasma and tissue extracts.
solutions, QCs and all samples in both plasma and tumor extracts
(Table 3). When QCs had been kept at a room temperature for
6 h in acidied samples and then extracted after alkalization, the
mean concentrations of CPT-11 and SN-38 were determined to be
in the range of 91.9101.8% in plasma and 88.899.6% in tumor
extracts relative to QC nominal concentrations. In the assay of
post-preparative stability, their mean concentrations were within
the range of 102.1110.7% in plasma and 92.0105.6% in tumor
extracts. In addition, the levels of CPT-11 and SN-38 were shown
to have no signicant degradation in the freeze-thaw stability tests
(90% < measured concentrations < 110%). The levels of CPT as one
concentration of 100 ng/ml were determined to have the range
of 86.7112.1% in all stability assay (data not shown). QCs of
CPT-11 and SN-38 were stable in 31 days when stored at 20 C
(106.3111.6%)

3.3. Accuracy and precision 3.5. Assay application to the pharmacokinetic study

The precision and accuracy data from QCs were within accept-
able limits in the within- and between-day assay (Table 2). The CVs
of the precision were less than 10% in all assays for CPT-11 or SN-38.
The accuracy data of CPT-11 or SN-38 had the range of 89.8114.3%
in comparison to the nominal QC concentrations in both plasma
and tumor extracts. It was found that the validation results of lin-
earity, accuracy and precision were also acceptable in the range of
10010,000 ng/ml for CPT-11 (data not shown).
3.4. Stability
All of CPT derivatives showed acceptable stabilities when the
method adopted the acidication pretreatment of standard curve
Newly validated method was successfully applied to deter-
mine CPT-11 and SN-38 in plasma during 24 h after intravenous
administration of CPT-11 (6 mg/kg) in the pharmacokinetic study
(Fig. 4). In plasma, the concentration of SN-38, major metabolite
of CPT-11, showed a tendency to be about 1/100th of the parents
concentration (Fig. 4A). In the case of tissues disposition, the lev-
els of CPT-11 and SN-38 were mostly higher in VX2 tumors than
plasma or normal muscles. Especially, the concentration of SN-38,
major active metabolite, was the largest in VX2 tumors at near
12 h after injection, at which SN-38 was little detected in plasma
and muscle tissues. It was considered that active metabolite,
SN-38 could effectively exhibit cytotoxicity and anticancer effect
(Fig. 4B).
 D.J. Park et al. / J. Chromatogr. B 962 (2014) 147152 151

Table 2
Accuracy and precision of the determination method forCPT-11 and SN-38 in rabbit plasma and tumor extracts.

Nominal concentration Plasma Tumor extracts

(ng/ml)
Concentration Accuracya Precisionb Concentration Accuracy Precision
(Mean SD, ng/ml) (%) (CV, %) (Mean SD, ng/ml) (%) (CV, %)

Within-day (n = 5)
CPT-11>
5 5.0 0.34 99.7 6.8 4.6 0.07 91.7 1.6
100 97.9 8.20 97.9 8.4 96.1 7.25 96.1 7.5
1000 942.6 42.86 94.3 4.5 919.1 27.23 91.9 3.0
SN-38
1 1.0 0.07 95.7 7.2 1.1 0.03 106.1 3.0
50 50.0 2.33 99.9 4.7 51.3 1.96 102.7 3.8
250 242.8 9.39 97.1 3.9 272.0 8.89 108.8 3.3

Between-day (n = 5)
CPT-11
5 5.1 0.32 101.9 6.4 4.8 0.25 96.7 5.2
100 97.2 4.82 97.2 5.0 98.8 3.88 98.8 3.9
1000 1017.2 59.41 101.7 5.8 977.5 54.50 97.8 5.6
SN-38
1 1.0 0.06 100.2 5.8 1.0 0.05 98.6 4.7
50 51.6 2.95 103.2 5.7 47.5 2.16 95.0 4.6
250 253.2 9.46 101.3 3.7 246.3 14.44 98.5 5.9
a
(Mean concentration found/nominal concentration) 100(%).
b
Standard deviation/mean concentration 100(%).

Table 3
Stabilities of CPT-11 and SN-38 in rabbit plasma and tumor extracts.

Stability test CPT-11 SN-38

Concentration Stabilitya Concentration Stability

(ng/ml, Mean SD) (%) (ng/ml, Mean SD) (%)

Rabbit plasma
Short-term 5 4.7 0.08 94.1 1 0.9 0.10 91.9
(6 h at RT) 1000 1018.0 40.47 101.8 250 247.82 5.10 99.1
Post-preparative 5 5.2 0.28 103.3 1 1.1 0.12 110.7
(6 h at 10 C) 1000 1034.4 79.59 103.4 250 255.3 6.18 102.1
3 Freeze-thaw cycles 5 4.8 0.57 95.7 1 1.0 0.12 99.0
(20 C) 1000 1004.7 77.66 100.5 250 247.1 26.31 98.8
Long-term 5 5.3 0.64 106.3 1 1.1 0.11 108.3
(31 days at 20 C) 1000 1076.8 145.08 107.7 250 266.3 27.24 106.5

Tumor extracts
Short-term 5 4.9 0.38 98.3 1 0.9 0.11 88.8
(6 h at RT) 1000 996.3 61.79 99.6 250 241.1 8.25 96.4
Post-preparative 5 5.0 0.39 99.0 1 0.9 0.10 92.0
(6 h at 10 C) 1000 986.1 32.97 98.6 250 263.9 9.89 105.6
3 Freeze-thaw cycles 5 4.8 0.63 95.5 1 1.1 0.15 105.3
(20 C) 1000 951.0 107.41 95.1 250 247.0 27.14 98.8
Long-term 5 5.4 0.48 108.5 1 1.1 0.12 111.6
(31 days at 20 C) 1000 1104.6 159.02 110.5 250 273.5 32.11 109.4
a
(Mean concentration found/nominal concentration) 100(%).

4. Discussion acidication-alkalization process of sample solutions, which

impacted on both stabilities and extraction efciencies.
The LLE method of CPT derivatives was optimized to It was reported that there was a signicant loss of CPT deriva-
increase the extraction efciencies of all CPT derivatives through tives over time in the various solutions including phosphate

Fig. 4. Application of the developed LC-MS/MS method to the pharmacokinetic study of CPT-11 and SN-38 in plasma and tissues after injection of CPT-11 (6 mg/kg) in rabbits
bearing VX2 tumors: plasma concentration-time proles (A) and comparison of distribution in various tissues after injection of CPT-11 and SN-38 (B).
152 D.J. Park et al. / J. Chromatogr. B 962 (2014) 147152
buffered system, human plasma and RPMI 1640 (cell culture
medium) all of which had the same acidity of pH 7.4 (physiological
condition). However, the lactone forms of CPT derivatives were rel-
atively unchanged at pH 6.0 or below. Furthermore, the acidity of
pH 3.0 was able to transform carboxylate to lactones forms [12]. We
have found that 7080% of CPT derivatives are converted into car-
boxylate forms within 3 h at room temperature in rabbit plasma or
tissue extracts. Therefore, it has been still insufcient to ascertain
the real concentration of CPT derivatives in vivo because of time dis-
crepancy between sampling and quantication. Generally, it was
more frequently used as a sample preparation method of acidi-
cation and protein precipitation, followed by HPLC-uorescence
detection to determine CPT-11 and SN-38 because such meth-
ods could make analysts get back the loss of carboxylate forms
of CPT derivatives produced by the inter-conversion [8]. De Bruijn
determined the lactone forms and total forms of CPT derivatives
by different sample pretreatment methods as the following: for
the determination of lactone forms, LLE method was used with
organic layer of acetonitrile-n-butyl chloride which had extracted
more lipophilic lactone forms than carboxylate forms and for the
determination of the total forms, acidication and simple protein
precipitation were used by mixing with the cold (20 C) perchlo-
ric acidwatermethanol (1:20:20) of which perchloric acid was
able to convert the carboxylate forms to the actone forms [13]. In
other studies, the acidication and protein precipitation of sam-
ples were achieved with methanol or acetonitrile mixtures acidied
with mineral acids such as perchloric acid, hydrochloric acid and
ortho-phosphoric acid when they were connected with HPLC sys-
tem [1416]. However, we found that their incubation times of
530 min were not enough to complete the conversion of all car-
boxylate forms into the lactone forms but it needed at least 1 h
of incubation time at pH 3. There were a few LCMS/MS meth-
ods to determine CPT derivatives, all of which were related with
protein precipitation procedures but not liquid-liquid extraction
[17,18]. There are still bothersome problems in protein precipi-
tation method, especially when we determine CPT derivatives by
the delicate LCMS/MS system in dirty biological samples like as
tissues. For that reason, we adopted the liquid-liquid extraction
method as a preparation method in order to eliminate massive
ingredients and prevent the contamination of LCMS/MS system.
However, when CPT derivatives were extracted with organic sol-
vent from acidied samples, it found that SN-38 and CPT were
well extracted but CPT-11 was not. CPT-11 had reached the sum-
mit of extraction efciencies in the range of pH 8.09.0, whereas
extraction efciencies of both SN-38 and CPT had begun to decline
in that range. Therefore, we extracted all CPT derivatives from bio-
logical matrix with organic extraction solution in optimized range
of pH 8.08.5 by alkalization just before adding extraction solu-
tion. In conclusion, we developed and validated new LCMS/MS
method capable of determining both carboxylate and lactone forms
and successfully applied to the determination of CPT derivatives in
the pharmacokinetic study of rabbit plasma and tissues.
Acknowledgement
This study was supported by a grant (KMS2011) from College of
Medicine, Kosin University, South Korea.
References
[1] M.F. Wall, M.C. Wani, C.F. Cook, K.H. Palmer, A.T. McPhail, G.A. Sim, J. Am. Chem.
Soc. 88 (1966) 38883890.
[2] M. Yoshida, N. Boku, A. Ohtsu, M. Muto, F. Nagashima, S. Yoshida, Gastric Cancer
4 (2001) 144149.
[3] C.S. Fuchs, M.R. Moore, G. Harker, L. Villa, D. Rinaldi, J.R. Hecht, J. Clin. Oncol.
21 (2003) 807814.
[4] R.H. Mathijssen, R.J. van Alphen, J. Verweij, W.J. Loos, K. Nooter, G. Stoter, A.
Sparreboom, Clin. Cancer Res. 7 (2001) 21822194.
[5] M.L. Rothenberg, J.R. Eckardt, J.G. Kuhn, H.A. Burris 3rd, J. Nelson, J.R. Eckatdt,
M. Tristan-Morales, S.G. Hilsenbeck, G.R. Weiss, L.S. Smith, G.I. Rodrigez, M.K.
Rock, D.D. Von Hoff, J. Clin. Oncol. 11 (1993) 21942204.
[6] L. Stewart, M.R. Redinbo, X. Qiu, W.G. Hol, J.J. Champoux, Science 279 (1998)
15341541.
[7] G. Boyd, J.F. Smyth, D.I. Jodrell, J. Cummings, Anal. Biochem. 297 (2001) 1524.
[8] M. Ramesh, P. Ahlawat, N.R. Srinivas, Biomed. Chromatogr. 24 (2010)
104123.
[9] S. Bardin, W. Guo, J.L. Johnson, S. Khan, A. Ahmad, J.X. Duggan, J. Ayoub, I. Ahmad,
J. Chromatogr. A 1073 (2005) 249255.
[10] H. Jiang, J. Zeng, N. Zheng, H. Kandoussi, Q. Peng, J.L. Valentine, R.W. Lange, M.E.
Arnold, Anal. Chem. 83 (2011) 62376244.
[11] J.H. Kim, J.Y. Oh, B.H. Park, D.E. Lee, J.S. Kim, H.E. Park, M.S. Roh, J.E. Je, J.H. Yoon,
S.H. Thorne, D. Kirn, T.H. Hwang, Mol. Ther. 14 (2006) 361370.
[12] K. Sano, M. Yoshikawa, S. Hayasaka, K. Satake, Y. Ikegami, H. Yoshida, T.
Ishikawa, S. Sawada, S. Tanabe, J. Chromatogr. B. Analyt. Technol. Biomed. Life.
Sci. 795 (2003) 2534.
[13] P. de Bruijn, J. Verweij, W.J. Loos, K. Nooter, G. Stoter, A. Sparreboom, J. Chro-
matogr. B. Biomed. Sci. Appl. 698 (1997) 277285.
[14] A. Sparreboom, P. de Bruijn, M.J. de Jonge, W.J. Loos, G. Stoter, J. Verweij, K.
Nooter, J. Chromatogr. B. Biomed. Sci. Appl. 712 (1998) 225235.
[15] D.F. Chollet, L. Goumaz, A. Renard, G. Montay, L. Vernillet, V. Arnera, D.J. Mazzo,
J. Chromatogr. B. Biomed. Sci. Appl. 718 (1998) 163175.
[16] J. Escoriaza, A. Aldaz, C. Castellanos, E. Calvo, J. Giraldez, J. Chromatogr. B.
Biomed. Sci. Appl. 740 (2000) 159168.
[17] S. Khan, A. Ahmad, I. Ahmad, Biomed. Chromatogr. 17 (2003) 493499.
[18] S. Khan, A. Ahmad, W. Guo, Y.F. Wang, A. Abu-Qare, I. Ahmad, J. Pharm. Biomed.
Anal. 37 (2005) 135142.