3166 J. Sep. Sci.

2011, 34, 3166–3177

Julia Martı́n Research Article

Dolores Camacho-Muñoz
Juan L. Santos
Irene Aparicio
Esteban Alonso
Department of Analytical
Chemistry, High Polytechnic
School, University of Seville,
Seville, Spain
Received May 25, 2011
Revised July 12, 2011
Accepted August 22, 2011
Simultaneous determination of a selected
group of cytostatic drugs in water
using high-performance liquid
chromatography–triple-quadrupole mass
spectrometry
In recent years, an increasing concern has risen about the presence of pharmaceuticals in
the aquatic environment. Despite their toxicity, increasing consumption and release into
the municipal sewage, only a few studies have been focused on cytostatic drugs, mainly
due to the lack of methods for their simultaneous analysis. In this work, a method, based
on solid-phase extraction prior to high-performance liquid chromatography–triple quad-
rupole mass spectrometry determination, was optimized and validated for the simulta-
neous determination of some (14) of the most widely used cytostatic drugs in river water,
influent and effluent wastewater. Process efficiency was in the range between 41 and 99%
in real samples, except for cytarabine (24%), docetaxel (17%) and methotrexate (30%), due
to suppression effects; precision values were o11%, except for gemcitabine (up to 19%);
and detection limits were in the range between 0.1 and 38 ng/L. Cytarabine, doxorubicin,
etoposide, gemcitabine, iphosphamide and vinorelbine were found at concentration levels
up to 14 ng/L in influent and effluent wastewater, showing an insignificant decrease
during sewage treatment; cytarabine and gemcitabine were found in effluent wastewater
and were also detected in river water associated with effluent discharges

Keywords: Cytostatic drugs / Emerging pollutants / Liquid chromatography /

Solid-phase extraction / Triple quadrupole mass spectrometry
DOI 10.1002/jssc.201100461

1 Introduction antibiotics, hormones and anti-inflammatory drugs [1, 7],

Nowadays, an increasing concern about the presence of
pharmaceutical compounds in different environment
compartments has happened due to their toxicity, resulting
in a large number of published studies [1, 2]. In fact,
pharmaceuticals have been designed to exhibit biological
activity in humans and may, in principle, cause adverse
effects to aquatic organisms.
Although many research studies about their occurrence
and ecotoxicity in wastewater [3], surface and groundwater
water [4], sewage sludge [5], soils and sediments [6] can be
found extensively in the literature, they are limited to certain
therapeutic groups. Special attention have received the
Correspondence: Dr. Esteban Alonso, Department of Analytical
Chemistry, High Polytechnic School, University of Seville,
C/ Virgen de África, 7, E-41011 Seville, Spain
E-mail: ealonso@us.es
Fax: 134-9-5428-2777
Abbreviations: IS, internal standard; MDL, method detection
limit; MRM, multiple reaction monitoring; QqQ-MS, triple
quadrupole mass spectrometry; WWTP, wastewater
treatment plant
while only few studies have been focused on the family of
cytotoxic (also known as antineoplastic) chemotherapy
drugs, which were designed to kill rapidly growing cells
such as those found in cancer tumors. Methods reported for
the determination of some cytostatic drugs such as irinote-
can, mitomycin, paclitaxel or vinorelbine are restricted to
their determination in biological matrices [8–10] and in
pharmaceutical preparations [11, 12]. For most of these
agents, the environmental burden is unknown, because the
required analytical procedures have not been established
yet. A few exceptions include the cytotoxics cyclophos-
phamide and iphosphamide [13, 14].
Due to the highly potent mechanism of action of these
compounds (including carcinogenic, mutagenic or embryo-
toxic properties), they are considered of great environmental
concern, even though consumption rates and
expected concentrations in the environment may be
comparatively low [13]. In developed countries, demand
for chemotherapy treatment continues to increase at
around 10% per year [15, 16]. Nowadays, there are over 50
cytotoxic drugs used routinely in chemotherapy in hospitals
([15], www.fda.gov/cder/cancer/druglistframe.htm). Table 1
shows the data of consumption per year in a single hospital,
data obtained from an hospital in Andalusia (South of

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 3166–3177
Spain), of the 14 cytostatic drugs most frequently used. The
consumption of cytostatic drugs per hospital and year was in
the range from 11.6 kg of 5-fluorouracil to 0.024 kg of
mitomycin in 2010. While some compounds, as X-ray
contrast media, are controlled once administered in hospi-
tals, patients treated with cytostatic drugs, which are among
the most toxic pharmaceuticals of common use, do not
follow any control. Some of the amounts administered in
hospitals are excreted at home reaching municipal sewage
via urine and feces of patients [17]. Therefore, they are
assumed to be environmentally relevant compounds.
Most of the analytical methods published in literature
are limited to individual determinations of the most
consumed anticancer drugs: cyclophosphamide and iphos-
phamide [13, 18–21]. In hospital effluents, Steger-Hartmann
et al. [18] determined the antineoplastics cyclophosphamide
and iphosphamide at concentration levels of 146 and 24
ng/L, respectively, while higher concentration levels of
iphosphamide (109 ng/L) were reported by Kümmerer et al.
[19]. Garcı́a-Ac et al. [20] determined these compounds in
influent wastewater reporting concentration levels of 9 and
59 ng/L for cyclophosphamide and methotrexate, respec-
tively. Buerge et al. [13] and Zuccato et al. [21] detected
concentration levels up to 0.07 and 10 ng/L, respectively, of
cyclophosphamide in surface water. Mahnik et al. developed
two methodologies to determine 5-fluorouracil [22] and
doxorubicin and epirubicin [23] in hospital effluents,
reporting concentration levels that reached 124 and 1.35
mg/L, of 5-fluorouracil and doxorubicin, respectively. Other
authors [4, 20, 24, 25] have published analytical methods to
determine one or two of these cytostatic drugs within
multiresidue methods. To our knowledge, only one analy-
tical method for the simultaneous determination of ten
cytotoxic drugs has been developed [11], but not for the
analysis of environmental samples. Besides, it requires
16 min to separate ten cytotoxic drugs, without including
one of the most used as 5-fluorouracil because it was poorly
retained. The disadvantage of that limitation was justified
because 5-fluorouracil has been considered as ‘not classifi-
able as to its carcinogenicity to humans’ by the International
Agency for Research on Cancer [26]. Although it has been
demonstrated that high-performance liquid chromatog-
raphy (HPLC) with diode array absorbance and fluorescence
detection may be a low-cost technology suitable in some
cases for the analysis of pharmaceuticals in surface water
and wastewater [3], the implementation of MS detection is
out of question for trace analysis of cytostatic drugs. Triple
quadrupole mass spectrometry (QqQ-MS) provides a high
certainty in peak identification due to the possibility of
multiple reaction monitoring (MRM) [27].
This work describes a rapid, specific, reliable and
sensitive analytical method for the simultaneous determi-
nation of 14 of the most widely and frequently used cyto-
static drugs (Table 1) in influent and effluent wastewater
and river water. The analytical method involves sample pre-
treatment by solid-phase extraction (SPE) and determination
by HPLC/QqQ-MS. Method applicability was tested by the
& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Liquid Chromatography 3167
determination of the cytostatic drugs in river, influent and
effluent wastewater. The selection of cytostatic drugs was
based on data from a hospital in Andalusia (Spain) of the 14
cytostatic agents most consumed (Table 1). Platinum
compounds need other detection techniques such as ICP-
MS or voltammetry, because of that cis-platinum was
eliminated from the original list [11]. Moreover, although
cis-platinum was one of the first cytostatic drugs used,
nowadays it is among the less common cytostatic drugs
used (0.047 kg/year).
2 Materials and methods
2.1 Materials and reagents
HPLC-grade water, acetonitrile and methanol were supplied
by Romil (Barcelona, Spain). HPLC-grade acetone and
analytical grade sulfuric acid and formic acid 98% were
obtained from Panreac (Barcelona, Spain). Ammonium
formate was purchased from Sigma-Aldrich (Steinheim,
Germany). Tert-butyl methyl ether was supplied by Scharlab
(Barcelona, Spain). L-Amethopterin hydrate (methotrexate,
498%), cytosine-b-D-arabinofuronoside (cytarabine), 5-
fluorouracil (499%), iphosphamide and mitomycin C from
Streptomyces caespitosus were supplied by Sigma-Aldrich
(Steinheim). Cyclophosphamide monohydrate was
purchased from Fluka (Steinheim, Germany). Docetaxel
trihydrate, doxorubicin hydrochloride, epirubicin hydro-
chloride, etoposide, gemcitabine, paclitaxel and vinorelbine
tartrate were purchased from the European Pharmacopoeia
Reference Standards (Strasbourg, France). Irinotecan hydro-
chloride trihydrate was obtained by Tokyo Chemical
Industry (Tokyo, Japan). Internal standard (IS) phenacetin-
ethoxy-1-13C was obtained by Cambridge Isotope Labora-
tories (Andover, MA, USA). The chemical structures of each
drug are shown in Table 1. SPE cartridges, the ones packed
with 60 mg of Oasis HLB and the others packed with 60 mg
of Oasis MCX, were purchased from Waters (Milford, MA,
USA). SPE cartridges, packed with Chromabond tetracycline
(500 mg), were supplied by Machery-Nagel (Düren,
Germany).
Stock solutions of each cytostatic drug at a concentra-
tion level of 1000 mg/L were prepared in methanol and
stored at 41C. Calibration solutions, in the concentration
range from 0.05 to 1000 mg/L, were prepared by diluting the
stock standard solutions in methanol.
2.2 Sample collection
Influent and effluent wastewater samples used to test
method applicability were collected in January 2011 from a
wastewater treatment plant (WWTP) sited in Seville (South
of Spain). Twenty-four-hour composite samples of influent
and effluent wastewater were taken with an automatic
device. River samples were collected in January 2011 from
www.jss-journal.com
3168 J. Martı́n et al. J. Sep. Sci. 2011, 34, 3166–3177

Table 1. Chemical structures and consumption rates of the cytostatic drugs

Cytostatic drug Chemical structure kg/yeara)

5-Fluorouracil O 11.6
F
HN

O N
H
Gemcitabine NH 1.7

HO
N O
F
O

OH F
Cyclophosphamide Cl 1.2

HN
N
Cl P
O
O
Iphosphamide O H 0.909
O N
P Cl

N
Cl
Cytarabine NH 0.571

HO
N O
OH
O

OH
Methotrexate H N N N 0.390
N N
N O
H
NH N
OH

O OH
Paclitaxel O O OH 0.204
O CH
O

O
Ph NH O

O
Ph O O
OH
OH O O

Ph
Etoposide OH 0.101
HO O

O
O O

O
O

O O

OH

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 3166–3177 Liquid Chromatography 3169

Table 1. Continued.
Cytostatic drug Chemical structure kg/yeara)

Irinotecan 0.087
O
N N
O N O
N

OH
O
Docetaxel O OH 0.074
HO
O

O
O NH O

O
Ph O O
OH
OH O O

Ph
Epirubicin O 0.065

O O OH O OH

NH

OH
OH
O OH O
Doxorubicin O OH O
OH
0.041

OH

OCH O OH O NH

O
OH

CH

Vinorelbine 0.031
N

H
N
N
H CO

O
N
O
HO
O

OCH

Mitomycin C O 0.024
NH

O
O

H N OCH

N NH

a) Consumption rate of each cytostatic drug per year and hospital (data of consumption from an hospital in Andalusia, Spain).

Guadalquivir River, which is the receiving stream of the Prior to extraction, 250 mL of water samples was filtered
above-mentioned WWTP. Samples were transferred to through a 1.2-mm membrane filter (Whatman, Mainstone,
amber glass bottles and stored at 41C until analysis. UK) to remove suspended matter. The IS was added to

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
3170 J. Martı́n et al. J. Sep. Sci. 2011, 34, 3166–3177

Table 2. Optimized MRM parameters of the QqQ-MS determination

Compound Retention time (min) Precursor ion (m/z) MRM 1 (quantification) MRM 2 (confirmation) Fragmentor (V) CE (V)

Cytarabine 0.85 266 [M1Na]1 2664134 – 180 12

Fluorouracil 1.25 129 [M–H] 129442.1 – 100 15
Gemcitabine 1.45 264 [M1H]1 2644112 264495.0 180 12
Methotrexate 4.40 455 [M1H]1 4554308 4554175 140 16
Mitomycin C 4.85 357 [M1Na]1 3574242 3574274 140 24
Irinotecan 6.06 587 [M1H]1 5874124 5874167 180 36
Iphosphamide 6.26 261 [M1H]1 261492 2614154 138 24
Doxorubicin 6.29 544 [M1H]1 5444397 5444379 135 10
Epirubicin 6.43 542 [M–H]– 5424395 – 145 8
Cyclophosphamide 6.48 261 [M1H]1 2614140 2614106 131 20
Phenacetin (IS) 6.66 180 [M1H]1 1804110 1804163 89 16
Etoposide 6.79 606 [M1NH4]1 6064229 6064185 114 16
Vinorelbine 7.36 779 [M1H]1 7794658 7794323 175 20
Docetaxel 9.58 830 [M1Na]1 8304549 8304304 135 20
Paclitaxel 9.64 876 [M1Na]1 8764308 8764474 180 24

CE, collision energy.

filtered samples to achieve a final concentration, in samples
prior to extraction, of 0.2 mg/L.
2.3 Sample treatment
Sample treatment was carried out by SPE. Prior to sample
treatment, Oasis HLB SPE cartridges were conditioned with
3 mL methanol and 3 mL of deionized water at a flow rate of
about 3 mL/min. Samples (250 mL) were percolated
through the cartridges at a flow rate of approximately
15 mL/min using a vacuum pump. The loaded cartridges
were rinsed with 3 mL of deionized water at a flow rate of
about 3 mL/min. The elution was carried out using four
successive aliquots of 1 mL of methanol at a flow rate of
about 1 mL/min. The combined aliquots were evaporated to
dryness by a gentle nitrogen stream. The residues were
reconstituted in 1 mL of methanol, filtered through a 0.45-
mm nylon filter and 20 mL was injected into the HPLC
instrument.
2.4 Instrumentation
2.4.1 High-performance liquid chromatography
Chromatographic analyses were performed on an Agilent
1200 series HPLC (Agilent, USA) consisting of a vacuum
degasser, a binary pump, an autosampler and a thermo-
stated column compartment.
Separation was carried out using a Zorbax Eclipse XDB-
C18 Rapid Resolution HT (4.6 mm  50 mm id; 1.8 mm)
analytical column (Agilent). Analytes were separated by
gradient elution with (A) acetonitrile (containing 0.1% v/v
formic acid) and (B) aqueous 15 mM ammonium formate
solution (containing 0.1% v/v formic acid). The linear
gradient elution program was: from 5 to 60% of organic
phase in 7 min, and then increased to 90% in 1 min and
held for 1 min, finally, back to 5% of organic phase in 3 min.
Elution was performed at a flow rate of 0.6 mL/min with the
column temperature kept at 251C.
2.4.2 Mass spectrometry
MS measurements were done with a 6410 QqQ instrument
equipped with an electrospray ionization source (Agilent).
Ionization of analytes was carried out using the following
settings: MS capillary voltage 3000 V, drying-gas flow rate
9 L/min, drying-gas temperature 3501C and nebulizer
pressure 40 psi.
For each compound, two MRM transitions were moni-
tored, except for cytarabine, epirubicin and 5-fluorouracil for
which only one MRM transition was chosen. The MRM
transitions were measured in periods to obtain higher
signals. In period 1 (0–2.5 min), cytarabine, 5-fluorouracil
and gemcitabine were measured with a dwell time of
100 ms. In period 2 (2.5–5.5 min), methotrexate and mito-
mycin were measured with a dwell time of 140 ms. In
period 3 (5.5–7.2 min), cyclophosphamide, doxorubicin,
epirubicin, etoposide, iphosphamide, irinotecan and the I.S.
were measured with a dwell time of 100 ms except for
cyclophosphamide and iphosphamide for which a dwell
time of 200 ms was chosen to improve the sensitivity. In
period 4 (7.2–8.7 min), vinorelbine was measured with a
dwell time of 180 ms. In the last period (8.7–10.5 min),
docetaxel and paclitaxel were measured with a dwell time of
180 ms. Table 2 summarizes the optimized QqQ-MS
conditions for the analysis of the cytostatic drugs.
Instrument control and data acquisition were
performed with MassHunter software (Agilent).

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 3166–3177
2.5 Method performance
Calibration curves were constructed at concentration levels
ranging from 0.05 to 1000 mg/L. Precision and accuracy of
the method were evaluated by means of spiked samples.
Method detection limits (MDLs) of the analytes were
calculated as the concentration of analyte corresponding to
a signal-to-noise ratio of 3 in influent and effluent waste-
water and in river water. Matrix effect was investigated in all
types of samples.
3 Results and discussion
3.1 HPLC-QqQ-MS optimization
In this work, the Zorbax Eclipse XDB-C18 Rapid Resolution
HT analytical column was selected for the separation in
10 min of the 14 cytostatic drugs studied. The use of this
column allowed the satisfactory separation of the studied
compounds with a significant reduction in analysis time
and, as consequence, in solvent consumption in comparison
with previously reported methods for the determination of
ten of these drugs [11].
Optimization of the QqQ-MS conditions of the selected
compounds was performed by the automatic optimization
function of the MassHunter software (Agilent) assisted by
direct injections in the mass spectrometer of 100 mg/L
standard solutions of each compound at a flow rate of
0.6 mL/min.
Negative and positive ion modes were monitored
simultaneously. Most of the studied compounds were
determined under positive ionization, except fluorouracil
and epirubicin which were determined under negative
ionization.
Detection was performed in MRM mode using the two
most intense and characteristic MRM transitions for all
studied compounds, except for cytarabine, 5-fluorouracil
and epirubicin, due to their poor fragmentation. For such
compounds, only one specific selected MRM transition was
used to avoid the loss of sensitivity that occurs when more
transitions are added.
3.2 SPE optimization
The SPE procedure was optimized using aliquots of 250 mL
of Milli-Q water spiked at a concentration level of 100 mg/L
in each compound. Three different sorbents (Chromabond
tetracycline, Oasis HLB and Oasis MCX), pH values (acid,
basic and neutral) and elution solvents with different
polarities (acetone, methanol and tert-butyl methyl ether)
were tested to find the most efficient extraction conditions.
Each analysis was done in triplicate. The results are
summarized in Fig. 1.
The selection of solid-phase material was carried out in
the first step, using the three SPE cartridges at neutral pH.
& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Liquid Chromatography 3171
Methanol, which seems to be the most efficient solvent for
the elution of contaminants from different SPE cartridges
according to previously reported methods [4, 13, 28], was
chosen as the elution solvent for the selection of the SPE
cartridge. The highest recoveries were achieved using Oasis
HLB cartridges (mean recovery 90%). Moreover, some of the
studied cytostatic drugs (doxorubucin, irinotecan and
paclitaxel) as well as the IS showed poor recoveries using
both Oasis MCX and Chromabond tetracycline SPE
cartridges. According to these results, Oasis HLB SPE
cartridges were selected for sample extraction.
The study of the influence of the sample pH was carried
out using Oasis HLB cartridges and methanol as the eluent.
The highest recoveries were achieved at pH 7 (90%),
followed by pH 2 and 8.5 (mean recoveries 76 and 68%,
respectively). Moreover, at pH 7 recoveries ranged from 69
to 105% while some cytostatic drugs were not extracted
under acid (5-fluorouracil, mitomycin and vinorelbine) and
basic (doxorubicin, methotrexate, paclitaxel and vinorelbine)
conditions. These results are in accordance with those found
in literature that use Oasis HLB due to its capacity for the
simultaneous extraction of acid, base and neutral pharma-
ceutical compounds without the necessity of adjusting the
pH [29].
Oasis HLB cartridges and sample pH 7 were selected for
elution solvent optimization. Recoveries achieved using
different elution solvents are shown in Fig. 1C. Methanol
showed the highest recoveries for all the studied compounds
with mean recovery of 90%. Acetone showed slightly lower
recoveries (81%), while mean recovery using tert-butyl
methyl ether was significantly lower (53%).
3.3 Method performance
Method accuracy, expressed as recovery percentage, and
precision, expressed as repeatability (intra-day) and repro-
ducibility (between-day) in terms of relative standard
deviation (% RSD), were evaluated by recovery studies of
the target compounds carried out in spiked influent and
effluent wastewater and in surface water samples. Moreover,
the matrix effect was evaluated. Samples were spiked with a
mixed standard solution at a concentration level of 200 ng/L
in each compound in quadruplicate (n 5 4). Recoveries were
calculated by comparison of the peak areas obtained from
spiked samples with the peak areas from the same samples
without standard solution addition (blanks) and, finally,
with the areas obtained by direct injection of a standard
solution at the concentration level expected after sample
treatment. Results are shown in Table 3. Recoveries
achieved for most of the analytes were ranged from 41 to
99% in the studied water samples. Cytarabine, docetaxel and
methotrexate were the compounds with the lowest recovery
values (21–34% from river water, 11–32% from influent
wastewater and 20–25% from effluent wastewater).
Intra-day precision of most of the cytostatic drugs in
terms of % RSD was in the range from 1 to 8% in river
www.jss-journal.com
3172 J. Martı́n et al. J. Sep. Sci. 2011, 34, 3166–3177

Figure 1. Influence of (A) solid-

phase material (using pH 7 and
methanol as elution solvent), (B)
pH of sample and (C) elution
solvent on the extraction recov-
eries of the SPE procedure.

Table 3. Process efficiency and precision (% RSD) of the optimized method in river water, influent and effluent wastewater (n 5 4)

Cytostatic drug River water Effluent wastewater Influent wastewater

Process RSD (%) Process RSD (%) Process RSD (%)

efficiency (%) efficiency (%) efficiency (%)
Intra-day Between-day Intra-day Between-day Intra-day Between-day

Cytarabine 27 2 5 20 3 5 24 7 7
Cyclophosphamide 91 5 9 85 4 10 84 4 9
Docetaxel 21 9 8 20 3 8 11 9 10
Doxorubicin 84 7 10 68 4 6 69 3 3
Epirubicin 73 4 4 67 7 8 57 9 11
Etoposide 91 8 15 105 7 14 101 7 12
5-Fluorouracil 46 8 10 28 2 6 50 11 14
Gemcitabine 77 18 20 64 5 16 77 19 22
Iphosphamide 90 5 6 84 4 6 78 7 10
Irinotecan 43 2 4 50 1 4 58 4 6
Methotrexate 34 5 7 25 3 3 32 5 5
Mitomycin C 80 5 5 72 4 8 61 2 5
Paclitaxel 66 1 3 61 2 5 41 5 10
Vinorelbine 90 6 6 74 5 5 67 5 9

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 3166–3177 Liquid Chromatography 3173

Table 4. Linear ranges and MDLs and MQLs of the method in river water, influent and effluent wastewater samples

Cytostatic drug Linearity River water Effluent wastewater Influent wastewater

Linear range (ng/L) r2 MDL (ng/L) MQL (ng/L) MDL (ng/L) MQL (ng/L) MDL (ng/L) MQL (ng/L)

Cytarabine 4.8–200 0.9981 1.4 4.8 1.5 4.9 1.5 4.9

Cyclophosphamide 5.5–80 0.9999 1.7 5.5 2.3 7.7 2.1 7.1
Docetaxel 5.0–200 0.9989 1.5 5.0 1.7 5.5 1.9 6.3
Doxorubicin 14–400 0.9999 5.3 18 4.3 14 4.2 14
Epirubicin 2.4–200 0.9981 3.5 12 0.7 2.4 3.8 13
Etoposide 7.2–200 0.9995 2.2 7.2 3.0 9.8 3.4 11
5-Fluorouracil 70–4000 0.9914 34 114 21 70 38 128
Gemcitabine 3.3–200 0.9982 1.0 3.3 2.6 8.7 1.4 4.8
Iphosphamide 3.8–400 0.9984 1.3 4.4 1.1 3.8 1.7 5.8
Irinotecan 3.0–80 1.0000 0.9 3.0 1.0 3.5 1.1 3.7
Methotrexate 0.3–40 0.9988 0.1 0.3 0.1 0.3 0.1 0.3
Mitomycin C 5.5–200 0.9999 2.2 7.4 2.0 6.6 1.7 5.5
Paclitaxel 0.8–80 0.9989 0.3 0.8 0.2 0.8 0.3 0.9
Vinorelbine 13–200 0.9998 4.0 13 4.5 15 5.2 17

Figure 2. Matrix effect of river

water, influent and effluent
wastewater samples on the
determination of the cytostatic
drugs.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
3174 J. Martı́n et al. J. Sep. Sci. 2011, 34, 3166–3177

Table 5. Process efficiency and MDLs reported in published methods for the determination of cytostatic drugs in aquatic environmental
samples and those achieved by the proposed method

Cytostatic drug Sample Extraction method Determination Process efficiency (%) MDL (ng/L) References

Cyclophosphamide Surface water SPE (Oasis MCX) LC/MS-MS 70 0.02 [21]

Iphosphamide Hospital effluent SPE (RP-18) GC/MS 39 7 [19]
Cyclophosphamide Hospital effluent SPE (C18)1SPE (SiOH) GC/MS 30 6 [18]
Iphosphamide 39 7
Cyclophosphamide Effluent, tap water SPE (C18) LC/MS-MS 57 50–250 [25]
Iphosphamide 51 50–100
Cyclophosphamide Surface water, tap water SPE (PPL Bond-Elut) HPLC/MS-MS 71–102 10 [4]
Iphosphamide 73–87 4.2
Cyclophosphamide Influent, effluent, surface SPE (Bio-Beads SM) LC/MS-MS 74–94 0.02–1 [13]
Iphosphamide water, groundwater 75–102 0.02–2
Cyclophosphamide Hospital Influent SPE (Oasis HLB) LC/MS-MS 104 2 [14]
Iphosphamide 95
Cyclophosphamide Effluent SPE (Lichrolut EN) LC/MS-MS 106 1.90a) [24]
Methotrexate SPE (Oasis MCX) 76 0.83a)
Cyclophosphamide Influent, effluent, tap water SPE (Strata X) LC/MS-MS 80 5.0–6.0 [20]
Methotrexate SPE (Strata C8) 69 11.0–16
5-Fluorouracil Hospital effluent SPE ENV1 HPLC/CE 88 1700a) [22]
Doxorubicin Hospital effluent SPE (C8 columns) HPLC/FL 85 50 [23]
Epirubicin 86 60
Cyclophosphamide Influent, effluent SPE (HLB 1 WAX) UPLC/MS-MS 45–109 0.5–7.0 [31]
Doxorubicin
Etoposide
Iphosphamide
Methotrexate
Cytarabine Influent, effluent, river water SPE (Oasis HLB) HPLC/MS–MS 5–24 1.4–1.5 Proposed method
Cyclophosphamide 84–91 1.7–2.3
Docetaxel 11–21 1.5–1.9
Doxorubicin 68–84 4.2–5.3
Epirubicin 57–73 0.7–3.4
Etoposide 91–105 2.2–3.4
5-Fluorouracil 28–50 21–38
Gemcitabine 64–77 1.0–2.6
Iphosphamide 78–90 1.1–1.7
Irinotecan 43–58 0.9–1.1
Methotrexate 25–34 0.1–0.1
Mitomycin C 61–80 1.7–2.2
Paclitaxel 41–66 0.2–0.3
Vinorelbine 67–90 4.0–5.2

a) Method quantification limit.

water, from 2 to 11% in influent wastewater and from 1 to
7% in effluent wastewater (Table 3), except for gemcitabine
for which % RSD values up to 19% were achieved. Between-
day precision was lower than 15% for all the studied cyto-
static drugs, except for gemcitabine (up to 22%). These
values of % RSD can be considered satisfactory for the
analysis of environmental water samples.
Method linearity was evaluated for each cytostatic
compound by the injection of standard solutions at six
concentration levels ranging from 0.05 to 1000 mg/L. Cali-
bration curves were prepared by plotting the IS response
ratio (area of the analyte divided by area of the IS) measured
in each standard solution against its respective concentra-
tions. Linearity was well fitted by a linear expression with
coefficients of correlation (r2) higher than 0.998, except for
5-fluorouracil (r2 value up to 0.991). Results are shown in
Table 4.
MDLs and method quantification limits (MQLs) in the
range of 0.1–38 and 0.3–128 ng/L, respectively, were
achieved in the studied matrixes (Table 4). The highest MDL
corresponds to 5-fluorouracil. It could be mainly due to its
poor fragmentation by electrospray which leads to its low
sensitivity.
The matrix in which the analytes are found and the
extraction procedure are known to have a powerful effect on
electrospray MS by altering the extent to which analytes of

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 3166–3177
interest are ionized. To investigate this phenomenon, the
chromatogram of a standard solution containing all the
cytostatic drugs (50 mg/L) was compared with the chroma-
togram of an extract of wastewater and river water spiked
after the extraction procedure at the same concentration as
the standard solution. The results were highly matrix
dependent (Fig. 2). Ion suppression was observed for most
of the cytostatics (cytarabine, docetaxel, irinotecan, mito-
mycin and paclitaxel), while ionization of docetaxel was
enhanced. The effect observed for etoposide, 5-fluorouracil
and gemcitabine was dependent on the type of sample.
Cyclophosphamide, epirubicin, iphosphamide and vinor-
elbine were largely unaffected by the matrix. To consider
matrix effect and extraction recoveries together, process
& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Liquid Chromatography 3175
Figure 3. HPLC-QqQ-MS chromatograms of
a standard solution containing 50 mg/L of
each cytostatic drug (left panel) and an
effluent sample of a WWTP (right panel).
efficiency was taken into account to quantify the presence of
the cytostatic drugs as proposed by Gracia-Lor et al. [30] for
these complex-matrix samples (Table 3).
The proposed method presents a significant improve-
ment compared with previously reported methods for the
analysis of cytostatic drugs in the environment (Table 5). As
can be seen in Table 5, process efficiency and MDL achieved
with the proposed method are similar to those reported by
other authors for most of the analyzed compounds.
However, in comparison with the previously published
methods, the proposed method allows the simultaneous
determination of 14 cytostatic drugs (Table 5) while
other methods described in the literature have been
designed for the determination of one or two cytostatic
www.jss-journal.com
3176 J. Martı́n et al. J. Sep. Sci. 2011, 34, 3166–3177

Table 6. Concentration levels of the selected cytostatic drugs in river water, influent and effluent wastewater (n 5 3)

Cytostatic drug River water Effluent wastewater Influent wastewater

Mean (ng/L) RSD (%) Mean (ng/L) RSD (%) Mean (ng/L) RSD (%)

Cytarabine 13 3.2 14 1.4 9.2 0.5

Cyclophosphamide oMDL – oMDL – oMDL –
Docetaxel oMDL – oMDL – oMDL –
Doxorubicin oMDL 0.3 oMDL 0.9 4.5 0.8
Epirubicin oMDL – oMDL – oMDL –
Etoposide oMDL – 3.4 0.1 15 0.1
5-Fluorouracil oMDL – oMDL – oMDL –
Gemcitabine 2.4 0.2 7.0 0.8 9.3 3.5
Iphosphamide oMDL 0.2 1.2 0.1 3.5 0.1
Irinotecan oMDL – oMDL – oMDL –
Methotrexate oMDL – oMDL – oMDL –
Mitomycin C oMDL – oMDL – oMDL –
Paclitaxel oMDL – oMDL – oMDL –
Vinorelbine oMDL 0.2 9.1 1.1 oMDL –

drugs. Only the method recently published by Yin et al. [31]
has been found for the determination of more than two
cytostatic drugs. However, to our knowledge, there is still a
lack of simple validated methods for the simultaneous
determination of different cytotoxic agents in aquatic
environmental matrices that allow the determination of
cytostatic drugs such as cytarabine, docetaxel, gemcitabine,
irinotecan, mitomycin, paclitaxel and vinorelbine. Moreover,
the proposed method reports low limits of detection and
satisfactory process efficiency for most of the studied cyto-
static compounds.
3.4 Analysis of real samples
The optimized method was applied to determine the
concentration levels of 14 cytostatic drugs in river water,
influent and effluent wastewater. Figure 3 illustrates MRM
chromatograms of a standard solution and an effluent
wastewater sample. Concentration levels found are shown
in Table 6. Six of the 14 cytostatic drugs (cytarabine,
doxorubicin, etoposide, gemcitabine, iphosphamide and
vinorelbine) were found at range concentration levels
ranging from 3.5 to 15 ng/L and from 1.2 to 14 ng/L in
influent and effluent wastewater, respectively, observing an
insignificant reduction during wastewater treatment.
Two of the cytostatic drugs found in effluent wastewater,
cytarabine and gemcitabine, were detected in river
samples at concentration levels of 13 and 2.4 ng/L,
respectively, associated with effluent wastewater discharges
to the receiving streams. These levels were several
orders of magnitude lower than those at which acute
ecotoxicological effects have been reported to occur (mg/L
range) [1, 2, 19].
4 Concluding remarks
A rapid, selective and sensitive analytical method, based on
SPE extraction and HPLC/QqQ-MS quantification, has been
optimized and validated for the simultaneous determination
of 14 cytostatic drugs in river water, influent and effluent
wastewater samples. The best process efficiencies (41–99%
in real samples, except cytarabine (24%), docetaxel (17%)
and methotrexate (30%)) were achieved using HLB as SPE
sorbent at neutral pH and methanol as elution solvent. The
method provided good MDL in the range from 0.1 to 38
ng/L in real samples. It was applied to the analysis of real
samples from a WWTP and its receiving stream from
Andalusia: 6 of the 14 drugs were detected at range
concentration levels up to 15 and 14 ng/L in influent and
effluent wastewater, respectively; 2 of them were found in
river water.
The authors thank the financial support from the Programa
de Becas de Formación de Profesorado Universitario (Ministerio
de Ciencia e Innovación, Spain).
The authors have declared no conflict of interest.
5 References
[1] Fent, K., Weston, A. A., Caminada, D., Aquatic Toxicol.
2006, 76, 122–159.
[2] Santos, L. H. M. L. M., Araújo, A. N., Fachini, A., Pena,
A., Delerue-Matos, C., Montenegro, M. C. B. S. M., J.
Hazard. Mater. 2010, 175, 45–95.
[3] Camacho-Muñoz, D., Martı́n, J., Santos, J. L., Aparicio,
I., Alonso, E., J. Sep. Sci. 2009, 32, 3064–3073.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 3166–3177
[4] Sacher, F., Lange, F. T., Brauch, H. J., Blankenhorn, I., J.
Chromatogr. A 2001, 938, 199–210.
[5] Martı́n, J., Santos, J. L., Aparicio, I., Alonso, E., J. Sep.
Sci. 2010, 33, 1760–1766.
[6] Dı́az-Cruz, M. S., López de Alda, M. J., Barceló, D.,
Trends Anal. Chem. 2003, 22, 340–351.
[7] Kümmerer, K., Chemosphere 2009, 75, 435–441.
[8] Baylatry, M. T., Joly, A. C., Pelage, J. P., Bengrine-
Lefevre, L., Prugnaud, J. L., Laurent, A., Fernandez, C., J.
Chromatogr. B 2010, 878, 738–742.
[9] Kim, S. C., Yu, J., Lee, J. W., Park, E. S., Chi, S. C.,
J. Pharm. Biomed. Anal. 2005, 39, 170–176.
[10] Xiong, X., Lim, B. A., Lat-Luna, M., Chew, P., Tan, D.,
J. Chromatogr. B 2011, 755, 65–72.
[11] Nussbaumer, S., Fleury-Souverain, S., Antinori, P.,
Sadeghipour, F., Hochstrasser, D. F., Bonnabry, P.,
Veuthey, J. L., Geiser, L., Anal. Bioanal. Chem. 2010,
398, 3033–3042.
[12] Li, C. L., Cui, J. X., Wang, C. X., Zhang, L., Li, Y. H.,
Zhang, L., Xiu, X., Li, Y. F., Wei, N., Int. J. Pharm. 2010,
391, 230–236.
[13] Buerge, I., Rudolfbuser, H., Poiger, T., Müller, M.,
Environ. Sci. Technol. 2006, 40, 7242–7250.
[14] Ort, C., Lawrence, M. G., Reungoat, J., Eaglesham, G.,
Carter, S., Keller, J., Water Res. 2010, 44, 605–615.
[15] Johnson, J. C., Jürgens, M. D., Williams, R. J.,
Kümmerer, K., Kortenkamp, A., Sumpter, J. P.,
J. Hydrol. 2008, 348, 167–175.
[16] Summerhayes, M., J. Oncol. Pharm. Pract. 2003, 9,
123–128.
[17] Kümmerer, K., Chemosphere 2001, 45, 957–969.
& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Liquid Chromatography 3177
[18] Steger-Hartmann, T., Kümmerer, K., Schecker, J., J.
Chromatogr. A 1996, 726, 179–184.
[19] Kümmerer, K., Steger-Hartmann, T., Meyer, M., Water
Res. 1997, 31, 2705–2710.
[20] Garcia-Ac, A., Segura, P. A., Viglino, L., Fürtös, A.,
Gagnon, C., Prévost, M., Sauvé, S., J. Chromatogr. A
2009, 1216, 8518–8527.
[21] Zuccato, E., Calamari, D., Natangelo, M., Fanelli, R.,
Lancet 2000, 355, 1789–1790.
[22] Mahnik, S., Rizovski, B., Fuerhacker, M., Mader, R., J.
Anal. Bioanal. Chem. 2004, 380, 31–35.
[23] Mahnik, S., Rizovski, B., Fuerhacker, M., Mader, R.,
Chemosphere 2006, 65, 1419–1425.
[24] Castiglioni, S., Bagnati, R., Calamari, D., Fanelli, R.,
Zuccato, E., J. Chromatogr. A 2005, 1092, 206–215.
[25] Ternes, T. A., Hirsch, R., Mueller, J., Haberer, K.,
Fresenius J. Anal. Chem. 1998, 362, 329–340.
[26] IARC, Lyon, 1987. Overall Evaluations of Carcinogeni-
city: An Updating of IARC Monographs. Vol. 1–42.
http://monographs.iarc.fr/ENG/Classification/index.php
[accessed 20 March 2010].
[27] Buchberger, W. W., J. Chromatogr. A 2011, 1218,
603–618.
[28] Petrovic, M., Gros, M., Barceló, D., J. Chromatogr. A
2006, 1124, 68–81.
[29] Weigel, S., Kallenborn, R., Hühnerfuss, H., J. Chroma-
togr. A 2004, 1023, 183–195.
[30] Gracia-Lor, E., Sancho, J. V., Hernández, F., J. Chro-
matogr. A 2011, 1218, 2264–2275.
[31] Yin, J., Yang, Y., Li, K., Zhang, J., Shao, B., J. Chroma-
togr. Sci. 2010, 48, 781–789.
www.jss-journal.com