Probing for peptidic drugs (2–10 kDa) in doping control blood samples

Abstract

Bioactive peptides with a molecular mass between 2 and 10 kDa represent an important class of substances banned in elite sports, which has been recognized with an increasing number and variety of substances by anti-doping organizations. Also, the annually renewed list of prohibited substances of the World Anti-Doping Agency (WADA) explicitly mentions more and more of these peptides, and efficient testing procedures are required. Even under simplified sample preparation conditions, liquid chromatography coupled to high-resolution mass spectrometry (with resolution properties > 100,000 full width at half maximum) offers suitable conditions for this task and can therefore be used as an initial testing procedure. In contrast to urine, blood analysis essentially relies on the detection of intact peptide hormones, and the expected concentrations are commonly higher in blood samples than in urine. This facilitates the analysis, and a generic sample preparation by means of mixed-mode solid-phase extraction could be realized in this study. Co-extraction and analysis of several different peptides such as insulins (human, lispro, aspart, glulisine, tresiba, detemir, glargine, bovine insulin and porcine insulin), growth hormone releasing hormones (sermorelin, CJC-1295 and tesamorelin), insulin-like growth factors (long-R3-IGF-I, R3-IGF-I and Des1-3-IGF-I) and mechano growth factors (human MGF and MGF-Goldspink) with criteria that fulfil the requirements of the WADA documents (TD2022 MRPL) for doping controls. The proof of principle was shown by the analysis of post administration samples after treatment with synthetic insulin analogues.

Full text

235 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
Received: 12 July 2022
Revised: 5 August 2022
Accepted: 6 August 2022
Probing for peptidic drugs (2–10 kDa) in doping control blood
samples
Andreas Thomas1Sam Thilmany1Amelie Hofmann1Mario Thevis1,2
1Institute of Biochemistry, Center for
Preventive Doping Research, German Sport
University Cologne, Cologne, Germany
2European Monitoring Center for Emerging
Doping Agents, Cologne, Germany
Correspondence
Andreas Thomas, Institute of Biochemistry,
Center for Preventive Doping Research,
German Sport University Cologne, Am
Sportpark Müngersdorf 6, 50933 Cologne,
Germany.
Email: a.thomas@biochem.dshs-koeln.de
Abstract
Bioactive peptides with a molecular mass between 2 and 10 kDa represent an impor-
tant class of substances banned in elite sports, which has been recognized with an
increasing number and variety of substances by anti-doping organizations. Also, the
annually renewed list of prohibited substances of the World Anti-Doping Agency
(WADA) explicitly mentions more and more of these peptides, and efficient testing
procedures are required. Even under simplified sample preparation conditions, liquid
chromatography coupled to high-resolution mass spectrometry (with resolution prop-
erties >100,000 full width at half maximum) offers suitable conditions for this task and
can therefore be used as an initial testing procedure. In contrast to urine, blood anal-
ysis essentially relies on the detection of intact peptide hormones, and the expected
concentrations are commonly higher in blood samples than in urine. This facilitates
the analysis, and a generic sample preparation by means of mixed-mode solid-phase
extraction could be realized in this study. Co-extraction and analysis of several differ-
ent peptides such as insulins (human, lispro, aspart, glulisine, tresiba, detemir, glargine,
bovine insulin and porcine insulin), growth hormone releasing hormones (sermorelin,
CJC-1295 and tesamorelin), insulin-like growth factors (long-R3-IGF-I, R3-IGF-I and
Des1-3-IGF-I) and mechano growth factors (human MGF and MGF-Goldspink) with
criteria that fulfil the requirements of the WADA documents (TD2022 MRPL) for dop-
ing controls. The proof of principle was shown by the analysis of post administration
samples after treatment with synthetic insulin analogues.
KEYWORDS
high-resolution mass spectrometry, LC-MS/MS, sports drug testing
Abbreviations: FWHM, full width at half maximum; GH-RH, growth hormone-releasing hormone; IGF, insulin-like growth factor; ISTD, internal standard; LOD,limit of detection; MGF, mechano
growth factor; MRPL, minimum required performance level; PS, particle size; SPE, solid phase extraction;TD, technical document; tSIM, targeted single ion monitoring; WADA, world anti-doping
agency.
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited and is not used for commercial purposes.
© 2022 The Authors. Analytical Science Advances published by Wiley-VCHGmbH.
Anal Sci Adv. 2022;3:235–243. wileyonlinelibrary.com/journal/ansa 235

236 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
1INTRODUCTION
The World Anti-Doping Agency (WADA) has acted on the increasingly
important role of peptidic drugs by recognizing a considerable num-
ber of these in the annually updated prohibited list and, in accordance
with criteria applicable to urine samples, has also introduced mini-
mum required performance levels (MRPLs) for blood samples (serum
or plasma). These MRPLs are set to sub ng/ml levels for synthetic
insulins (0.3 ng/ml) and growth hormone releasing hormone analogues
(0.3 ng/ml), while higher concentrations apply in the case of IGF-I
analogues (2 ng/ml).1,2 In general, the analysis of blood samples is
favoured because for most (if not all) banned peptides the pharma-
cologically relevant concentrations in blood are established, whereas
the renal clearance and urinary concentrations in many cases are
largely unknown and sometimes unidentified degradation products
(metabolites) warrant consideration. Additionally, the stability of most
peptide hormones in blood samples (e.g. ethylenediaminetetraacetic
acid [EDTA] plasma or serum) is presumably superior compared to
urine samples, especially in situations when frozen conditions (e.g.
sample transport) are not guaranteed. Extensive degradation of the
prohibited peptides synachten (synthetic adrenocorticotropic hor-
mone analogue) and long-R3-IGF-I in urine was described previously
when stored at room temperature or 4◦C.3–6
Although the expected concentrations of the target peptides in
blood samples are generally higher than in urine samples, the available
volume compensates for this. With regular urine samples usually avail-
able in 100 ml containers, blood samples are commonly limited to less
than 2 ml of serum or plasma. The volumes required for blood analyt-
ical assays usually range between 50 and 500 µl, while for urine most
assays are designed to operate with 1 to 5 ml. Nevertheless, the analy-
sis of doping control blood samples represents a promising approach,
especially to uncover the misuse of prohibited peptide hormones
(2–10 kDa) by means of liquid chromatography-mass spectrometry
(LC-MS). Besides the low concentration of the target analytes, the
coexisting and complex mixture of high-abundant proteins compli-
cates a simple analysis with established standard protocols. Existing
methods include time-consuming sample preparation steps such as
immunoaffinity purification before LC-MS.5,7–15 These methods are
very specific and selective due to the complementary combination
of immune extraction, liquid chromatographic separation and detec-
tion by (high-resolution/tandem) MS. But due to the laborious sample
preparation steps, recently also more simplified approaches were
developed, which also meet the criteria outlined in mandatory WADA
documents.16–19 Most of these assays focus on urine analysis or are
limited to one class of peptides (e.g. insulins) only; conversely, in the
present study the applicability of a mixed-mode anion-exchange solid-
phase extraction (SPE) was shown to allow for an effective sample
preparation adequate for subsequent LC-MS analysis of doping control
blood samples for several different prohibited peptides. Noteworthy,
not all target peptides were recovered satisfactorily with the chosen
SPE-based strategy and, thus, for example synacthen was not consid-
ered in this study. On the other hand, the method is per definition not
limited to the included target peptides and also other substances (e.g.
lower molecular mass peptidic drugs) are co-extracted.
2METHODS
2.1 Chemicals and reagents
Acetic acid (glacial), acetonitrile and methanol were obtained from
Merck (Darmstadt, Germany). Ammonium hydroxide solution (25%
in water) and formic acid were purchased from Sigma (Schnelldorf,
Germany). The mixed-mode SPE cartridges (HR-XA, 3 ml, 60 mg)
were from Macherey&Nagel (Düren, Germany). For all dilution steps
and preparation of aqueous solutions, ultra-pure water of MilliQ-
quality was used. Insulin analogues lispro (Humalog), aspart (Novolog),
glulisine (Apidra), detemir (Levemir) and degludec (Tresiba) were sup-
plied by Eli Lilly (Indianapolis, IN), Novo Nordisk (Princeton, NJ) and
Aventis (Kansas City, MO). Long-R3-IGF-I, porcine insulin and bovine
insulin were from Sigma (Schnelldorf, Germany). GH-RH1-29 (Geref)
was purchased from BMFZ (Düsseldorf, Germany). Tesamorelin, and
Acetyl-(Tyr1,D-Arg
2)-GRF1-29(ISTD2) were from Bachem (Bubendorf,
Switzerland). CJC-1295((D-Ala2,Gln
8,Ala
15,Leu
27)-GRF amide), CJC-
1293((D-Ala2)-GRF amide) and the metabolites of Geref (GRF3-29)
and CJC-1293((D-Ala2)-GRF2-29 amide, purity 91%) were custom-
synthesized by Centic Biotec (Jena, Germany). Des1-3 -IGF-I and
R3-IGF-I were obtained from IBT Biosystems (Reutlingen, Germany).
The glargine metabolite (DesB31-32 glargine) was obtained from IBA
(Warsaw, Poland, purity >90%) and the stable isotope-labelled insulin
internal standard [[2H10]-Leu B6,B11,B15,B17]-Insulin (human) (ISTD1)
was purchased from PeptaNova (Sandhausen, Germany). MGF human
and MGF “Goldspink” were obtained from Phoenix Pharmaceuticals,
Inc (Karlsruhe, Germany), and 15N-labelled IGF-I used as ISTD3 was
purchased from Prospec (Santa Clara, CA). All reference standards own
apurity>95% unless otherwise stated.
2.2 Sample preparation
All target peptides were purified from 200 µl of plasma or serum by
means of mixed-mode anion-exchangeSPE. Two hundred microliters of
plasma (or serum) were fortified with 10 µl of ISTD solution (containing
0.5 ppm of 2H-labelled human insulin (ISTD 1), 15N-labelled IGF-I
(ISTD 3) and acetyl-(Tyr1,D-Arg
2)-GRF1-29(ISTD 2) and 10 µl of ammo-
nium hydroxide solution (5% in water). After vortex, the samples were
precipitated with 550 µl of a mixture of ice-cold methanol/acetonitrile
(50/50, v:v) and centrifuged at 17,000 g for 10 min. The supernatant
was diluted with 1 ml of water in a new Eppendorf tube and trans-
ferred to the mixed-mode solid-phase cartridge (HR-XA), which was
preconditioned with 1 ml of methanol and 1 ml of water. After sample
loading, the cartridge was washed with 2 ml of water and 2 ml of
methanol/water (50/50, v:v). Elution into an Eppendorf tube followed
using 1.2 ml of methanol (acidified with 5% of formic acid). After

237 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
evaporation in a vacuum centrifuge (approx. 60 min at 40◦C), the
samples were reconstituted in 70 µl of formic acid (3%) and injected
with 20 µl into the LC-MS.
2.3 Blood specimens and administration samples
The validation was carried out with EDTA plasma and serum samples
from 10 healthy male and female volunteers without any medication
within the last 24 h. Additionally, a commercially available plasma pool
(Octaplas, Octapharma GmbH, Langenfeld, Germany) was used for
selected validation parameters (e.g. recovery) due to the absence of
endogenous insulin. Post-administration serum samples were obtained
from insulin-dependent diabetics who regularly administer different
synthetic insulin analogues (male volunteer, diabetes mellitus type I,
subcutaneous injection via insulin-pen, insulin aspart: 8 IU, 2 h before
sample collection, insulin detemir: 17 IU, 3 hours before sample col-
lection). Written consent of the volunteers and approval by the local
ethics committee (DSHS No.: 139/2021) was obtained for this study.
2.4 Liquid chromatography
The chromatographic separation of the peptides was performed by
means of high-performance LC using a Vanquish system (Thermo, Bre-
men, Germany). The system was equipped with a dual-pump set-up and
initial trapping of the target analytes on an Accucore Phenyl/Hexyl, 3 ×
10 mm, 2.7 µm PS (Thermo) trapping column using water with formic
acid (0.1%, solvent A1) and acetonitrile (with 0.1% formic acid, solvent
B1). Trapping was performed for 2 min at 99% of solvent A1before
switching the flow to the analytical column (Poroshell C18 3 ×50 mm
(Agilent, Karlsruhe, Germany)). As solvent buffers, A2and B2for the
gradient, water with formic acid (0.1%) and dimethyl sulfoxide (DMSO,
1%) as solvent A2and acetonitrile with DMSO (1%) and formic acid
(0.1%) (solvent B2)wereused.Theflowwassetto400µl/minandthe
gradient started at 95% of A2and decreased to 60% of A2within 8 min.
Within the next 2 min, the gradient decreased to 20% of A2for cleaning
the column. Finally, the system was re-equilibratedfor 5 min at starting
conditions. The resulting overall run time was 15 min, and the injection
volume was 20 µl. The column compartment was set to 25◦Candthe
autosampler cooled to 10◦C.
2.5 Mass spectrometry
High-resolution MS was performed using an Orbitrap Exploris 480
(Thermo) equipped with a heated electrospray ion source. The instru-
ment operated in positive ionization mode acquiring data in full scan
mode (mass-to-charge ratio (m/z)=400–1700, resolution 60,000 full
width at half maximum [FWHM]) and targeted single ion monitoring
(tSIM) with data dependent MS2by means of an inclusion list. tSIM
experiments were performed for the multiply protonated molecules of
the target peptides (see Table 1) with a resolution at 120,000 FWHM
and multiplexed 5 times with a quadrupole isolation window of 3 m/z
units. The data-dependently triggered targeted MS2experiments were
acquired with a resolution of 15,000 FWHM and a quadrupole isola-
tion window of 2 Da. The instrument was calibrated according to the
manufacturer‘s recommendations using a calibration mixture (consist-
ing of caffeine, the tetrapeptide MRFA and Ultramark). The gas supply
consisted of nitrogen (N2-generator; CMC, Eschborn, Germany). Ion-
ization in positive mode was accomplished at a voltage of 3 kV, and the
temperature of the ion transfer tube was adjusted to 320◦C. Xcalibur
Software used was: Foundation 3.1 SP7 QF1 and Xcalibur 4.4 (Thermo).
2.6 Validation
The method was validated according to the requirements of WADA,
described in the international standards for laboratories and the tech-
nical document for the MRPL considering an initial testing procedure
for non-threshold substances.2,20 The parameters were specificity (10
blank plasma samples), reliability at the MRPL (10 different plasma
samples fortified at the respective MRPL), the limit of detection (LOD,
six different samples at MRPL 50%, 25% and 10%), carryover and sta-
bility in the autosampler (24 h). In addition to these recommended
parameters, also the recovery was determined. Here, the loss of the
respective peptides during the sample preparation is characterized
by analysing six technical replicates fortified at the MRPL before the
preparation in comparison to six replicates fortified after the sample
preparation just before the inj ection. The carryover of a highly concen-
trated sample (4 ×MRPL20) to a following blank sample was evaluated
with three repetitions. The chromatograms of the blank sample after
injection of the high concentrated sample (4 ×MRPL) was evaluated for
the presence of occurring peaks in the respective retention time win-
dow. Fortesting the stability after preparation, 10 samples at the MRPL
were reinjected after 24 h storage in the autosampler. The robust-
ness was tested for serum instead of plasma with six different serum
samples analysed as blank and fortified at the respective MRPL.
3RESULTS AND DISCUSSION
3.1 Validation
Due to the recently introduced MRPLs for several prohibited peptide
hormones in blood samples, all validation parameters were performed
in accordance with the WADA requirements fixed in the interna-
tional standard for doping control laboratories.20 The main results
aresummarizedalsoinTable2. Chromatograms were investigated at
the respective retention times (window ∼30 s) with a mass window
of 10 ppm. The detection rate in all blank samples was 0/10 except
for human insulin and IGF-I, which was present as endogenous hor-
mones in all samples (10/10). The specificity was fulfilled accordingly.
The same set of 10 blank samples was also fortified at the respective
MRPLs of the different peptide classes (see Table 2) and here all pep-
tides were detected with 100% detection rate (10/10 samples). Thus,

238 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
TAB L E 1 Amino acid sequences, masses, precursor ions and retention times of target peptides
Peptide Amino acid sequence
Monoisotopic
mass
(Da)
Precursor
tSIM
(m/z)
Dominant
charge
state
∼Ret. time
(min)
Human insulin GIVEQCCTSICSLYQLENYCN –
FVNQHLCGSHLVEALYLVCGERGFFYTPKT
5803.6 1162/1452 4+/5+6.9
Bovine insulin GIVEQCCASVCSLYQLENYCN-
FVNQHLCGSHLVEALYLVCGERGFFYTPKA
5729.6 1147 5+6.8
Lispro GIVEQCCTSICSLYQLENYCN –
FVNQHLCGSHLVEALYLVCGERGFFYTKPT
5803.6 1162 5+6.9
Aspart GIVEQCCTSICSLYQLENYCN –
FVNQHLCGSHLVEALYLVCGERGFFYTDKT
5821.6 1166 5+6.9
Glulisine GIVEQCCTSICSLYQLENYCN –
FVKQHLCGSHLVEALYLVCGERGFFYTPET
5818.6 1166 5+6.9
Glargine Met GIVEQCCTSICSLYQLENYCG –
FVNQHLCGSHLVEALYLVCGERGFFYTPKT
5746.6 1151 5+6.9
Porcine insulin GIVEQCCTSICSLYQLENYCN –
FVNQHLCGSHLVEALYLVCGERGFFYTPKA
5773.6 1156 5+6.9
Detemir GIVEQCCTSICSLYQLENYCN-
FVNQHLCGSHLVEALYLVCGERGFFYTPK-Myr
5912.8 1479 4+8.9
Degludec GIVEQCCTSICSLYQLENYCN –
FVNQHLCGSHLVEALYLVCGERGFFYTPK-γ-L-Glu-Pal
6099.8 1527 4+8.1
Geref YADAIFTNSYRKVLGQLSARKLLQDIMSR-NH23355.8 840 4+7.2
CJC-1293 YdADAIFTNSYRKVLGQLSARKLLQDIMSR-NH23355.8 840 4+7.2
CJC-1295 YANAIFTQSYRKVLAQLSARKLLQDILSR-NH23365.9 842 4+7.3
Geref Met AIFTNSYRKVLGQLSARKLLQDIMSR-NH23121.7 781 4+7.1
CJC-1293 Met dADAIFTNSYRKVLGQLSARKLLQDIMSR-NH23192.7 799 4+7.1
Tesamorelin Hex-YADAIFTNSYRKVLGQLSARKLLQDIMSRQ
QGESNQERGARARL-NH2
5133.7 734 7+7.3
LongR3-IGF-I MFPAMPLSSLFVNGPRTLCGAELVDALQFVCGDRGFYFNK
PTGYGSSSRRAPQTGIVDECCFRSCDLRRLEMYCAPLKPAKSA
9105.4 1307 7+7.4
R3-IGF-I GPRTLCGAELVDALQFVCGDRGFYFNKPTGYGSS
SRRAPQTGIVDECCFRSCDLRRLEMYCAPLKPAKSA
7670.6 1099 7+6.4
IGF-I GPETLCGAELVDALQFVCGDRGFYFNKPTGYGSS
SRRAPQTGIVDECCFRSCDLRRLEMYCAPLKPAKSA
7643.6 1093 7+6.6
Des1-3-IGF-I TLCGAELVDALQFVCGDRGFYFNKPTGYGSSSRRAPQ
TGIVDECCFRSCDLRRLEMYCAPLKPAKSA
7360.5 1053 7+6.5
MGF human YQPPSTNKNT KSQRRKGSTF EERK 2866.5 574 5+3.2
MGF “Goldspink” YQPPSTNKNT KSQRRKGSTF EEHK 2846.5 570 5+3.2
the reliability at the MRPL is given. The recoveries are calculated in
a range between <10% (e.g. hMGF) to >80% (e.g. insulins). The LOD
for each target peptide is defined with a detection rate of 95%. It was
found that for 12 out of the 18 target peptides the LOD range at 25%
of the MRPL and for 16 out of the 18 target peptides at 50% of the
respective MRPL. Noteworthy, hMGF and insulin detemir were reli-
ably detectable at 100% MRPL only. At 10% of the MRPL all target
peptides were detected only sporadically. Reinjection of the samples
fortified at the MRPL after storage in the autosampler (set to 10◦C) for
24 h showed good stability with a detection rate of 10/10. No carryover
in the chromatographic system from a sample fortified at 4 ×MRPL
level to the next blank sample was observed for all target analytes. The
serum samples tested for robustness showed no interfering signals in
the blank samples (0/6) and 100% detection rate at the MRPL (6/6).
Noteworthy, the here presented method is designed as initial testing
procedure to enable an effective first analysis of doping control sam-
ples. The final identification of the prohibited peptides according to the
technical document WADA TD_IDCR2021.21
3.2 Liquid chromatography-MS
The obtained results indicate sufficient selectivity, specificity and sen-
sitivity of the approach, which fulfils the criteria for state-of-the-art

239 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
TAB L E 2 Validation results
Peptide Specificity
Recovery
(%)
MRPL
(ng/ml)
Detection
rate at
MRPL
at MRPL
50%
at MRPL
25%
at MRPL
10%
Carryover
(%)
Stability
detection
rate after
24 h at
MRPL
Human insulin 10/10 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.
Bovine 0/10 82 0.3 10/10 6/6 6/6 6/3 n.o. 10/10
Lispro 0/10 78 0.3 10/10 6/6 6/6 6/5 n.o. 10/10
Aspart 0/10 86 0.3 10/10 6/6 6/6 6/3 n.o. 10/10
Glulisine 0/10 86 0.3 10/10 6/6 6/6 6/3 n.o. 10/10
Glargine Met 0/10 88 0.3 10/10 6/6 6/6 6/2 n.o. 10/10
Porcine 0/10 84 0.3 10/10 6/6 6/6 6/1 n.o. 10/10
Detemir 0/10 79 0.3 10/10 6/4 6/1 6/0 n.o. 10/10
Degludec 0/10 96 0.3 10/10 6/6 6/6 6/2 n.o. 10/10
Geref/CJC-1293 0/10 35 0.3 10/10 6/6 6/4 6/0 n.o. 10/10
CJC-1295 0/10 56 0.3 10/10 6/6 6/6 6/2 n.o. 10/10
Geref Met 0/10 31 0.3 10/10 6/6 6/6 6/1 n.o. 10/10
CJC-1293 Met 0/10 35 0.3 10/10 6/6 6/6 6/1 n.o. 10/10
Tesamorelin 0/10 21 0.3 10/10 6/6 6/4 6/0 n.o. 10/10
LongR3-IGF-I 0/10 43 2 10/10 6/6 6/5 6/1 n.o. 10/10
R3-IGF-I 0/10 15 210/10 6/6 6/2 6/0 n.o. 10/10
IGF-I 10/10 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.
Des1-3-IGF-I 0/10 33 210/10 6/6 6/6 6/2 n.o. 10/10
MGF human 0/10 5 2*10/10 6/4 6/1 6/0 n.o. 10/10
MGF “Goldspink” 0/10 25 2*10/10 6/6 6/6 6/3 n.o. 10/10
n.d.: not determined; n.o.: not observed.
*MRPL not explicitly shown in TD2022 MRPL.
doping control analysis. Figure 1shows the chromatograms of a blank
sample fortified at the respective MRPLs. Additionally, in Figure 2a
blank sample is shown. Nevertheless, considering established hybrid
assays combining immuno-extraction with subsequent LC-MS analy-
sis, the herewith yielded extracts are much more complex with a lower
degree of purification. This has a direct impact on the HRMS data with
a high number of interfering (or at least visible) signals in the tSIM mass
spectra at the respective retention time. This phenomenon was already
described earlier for insulin-specific assays.14 Especially in low concen-
trated samples (at MRPL or less), the evaluation of the mass spectra
and the corresponding extracted ion chromatograms are less straight
forward compared to the hybrid assay data reported previously.11,15
In order to enable specific detection, the resolution of the orbitrap
analyser was set to 120,000 FWHM in the SIM experiments. Thus,
extraction of very narrow mass ranges (∼2 ppm) were used for data
evaluation, and employing the multiplex option for the SIM precursors
allowed for the generation of a sufficient number of data points per
peak (with a typical peak width of 10 s). In case of confirmatory analysis,
extraction with specific antibodies is strongly recommended in order to
avoid result misinterpretations.
The growth hormone releasing hormones Geref (Sermorelin) and
CJC-1293 differ by the exchange of one D-amino acid at position
2(dAla
2) only. With the present approach it is not possible to dif-
ferentiate both peptides, due to identical mass, retention time, and
product ion spectrum (see also Table 1).22 This was shown already in
earlier studies and the differentiation might be enabled by diagnos-
tic metabolite pattern.15,23,24 The metabolites of Geref and CJC-1293
were included in the validation accordingly. For the differentiation of
human insulin and insulin lispro, the ddMS2 spectra enable the iden-
tification of the synthetic insulin analog with the diagnostic product
ion at m/z 217(corr. to (B)y2-ion, see Figure 1).11,15,17 The proof-of-
principle was shown with post administration samples obtained from a
patient suffering from diabetes mellitus following the regular treatment
with two synthetic insulin analogues. In Figure 3the extracted ion chro-
matograms for all synthetic insulin analogues are shown with abundant
signals for the short acting insulin aspart (at 6.9 min) and long acting
insulin detemir (at 8.9 min). Additionally, the corresponding mass spec-
trum from the SIM experiment for insulin aspart (showing the 5-fold
protonated precursor at m/z 1166) and the triggered ddMS2 spectrum
for insulin detemir (derived from the 4-fold protonated precursor at

240 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
FIGURE 1 Extracted ion chromatograms (from targeted single ion monitoring [tSIM]) of a blank plasma sample fortified at the respective
minimum required performance levels (MRPLs) of different insulins (0.3 ng/ml), GH-RH (0.3 ng/ml), MGF (2 ng/ml) and IGF-I analogues (2 ng/ml)
FIGURE 2 Extracted ion chromatograms (from targeted single ion monitoring [tSIM]) of a blank plasma (Octaplas) with the diagnostic ion
traces of the different insulins, GH-RH, mechano growth factor (MGF) and insulin-like growth factor (IGF)-I analogues

241 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
FIGURE 3 Extracted ion chromatograms (left, from targeted single ion monitoring [tSIM]) showing a sample from a diabetic patient after
administration (regular treatment) of the fast-acting synthetic insulin aspart and the long-acting insulin detemir. Mass spectra (right) of insulin
aspart (SIM, top), showing 5-fold protonated precursor ion and of insulin detemir (ddMS2, bottom) with two diagnostic product ions including the
attached myristic fatty acid
m/z 1489) are shown on the right. Under evaluation of the acquired
data, the presence of these two synthetic insulins is clearly indicated.
3.3 Doping control aspects
Especially in sports drug testing, the simplicity and the speed of the
method represent a clear benefit when analysing a high number of
samples in a short time. Due to the low number of adverse analyti-
cal findings in doping controls, confirmatory reanalysis of suspicious
samples is rare and, thus, effective (fast and sensitive) initial testing
methods are favoured. Here, the presented approach offers obvious
progress to established assays. Even with respect to the low MRPL
values, bioactive peptides are detectable after parental administration
for several hours up to a few days only.8,10–12,14,24 Thus, frequent out-
of-competition sampling represents a more promising doping control
approach, compared to classical in- (or after) competition testing. This
is especially true with most of the prohibited peptides providing no
direct benefit when used during the competition. Also, the expected
dosages in cheating athletes and the respective plasma concentrations
are very hard to foresee, because some of the target peptides lack med-
ical approval yet (e.g. GHRH- or IGF-I analogues) or produce direct
life-threatening effects in case of high dosages (e.g. insulins). Thus, as
long as no additional information about the misused peptides is avail-
able, the recommended MRPLs were accepted for effective doping
control analysis.
Not all prohibited peptides were purified with the same recovery
and the method represents a compromise between comprehensive-
ness on the one hand and selectivity on the other hand. All insulins were
well recovered, while the GH-RHs, IGFs and MGFs show significantly
lower values for recovery. Nevertheless, all WADA requirements are
met. In principle, the method is open also for further target peptides or
metabolites, but, noteworthy, also some prohibited peptides (e.g. syn-
acthen, isoelectric point [pI] =11) yielded inadequate results and were
not considered accordingly. The extraction quality is largely correlated
to the number of acidic respectively basic amino acids in the peptide
sequence and the resulting pI. On the other hand, also peptides <2kDa,
such as gonadoliberin and analogues are potentially co-extracted with
the present approach. Data for gonadoliberin (LH-RH, MW: 1181 Da,
not included in this communication) yielded recoveries at approx. 30%
and an estimated LOD <0.1 ng/ml. Hence, the method is potentially
expandable to other prohibited peptides (of lower molecular mass)
without changes in the sample preparation procedure.
4CONCLUSION
Along with the development of increasingly powerful mass spectrome-
ters with very high resolution (>100,000 FWHM), it is now possible to
simplify sample preparation to the extent that less pure extracts can be
analysed without jeopardizing the required specificity.25 Samples pre-
pared in this way are generally not suitable for analysis with nano-LC

242 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
systems, but modern normal-flow systems have demonstrated suffi-
cient sensitivity and robustness to achieve the mandatory MRPLs in
doping controls.7,8,14,16,19 The present method is designed as an initial
testing procedure, which ideally covers different (if not all) prohibited
peptides in one analytical approach. Although this was not entirely
achieved, the developed assay offers a clear improvement considering
simplicity and speed without compromising the analytical sensitivity.
This straightforward approach (without the need for handling antibod-
ies) can be readily implemented and allows for meeting mandatory
WADA requirements for initial testing procedures; however, for con-
firmatory analyses (which may result in formal adverse analytical
findings), methods employing immunoaffinity purification prior to MS
are still recommended.
AUTHOR CONTRIBUTIONS
Andreas Thomas was associated with method development and con-
ceptualization, and wrote the original draft and supervision. Sam
Thilmany and Amelie Hofmann were associated with methodology,
formal analysis and investigations. Mario Thevis was associated with
supervision and reviewed and edited the final manuscript.
ACKNOWLEDGEMENTS
The study was carried out with the support of the Manfred Donike
Institute for Doping Analysis (Cologne, Germany) and the Federal Min-
istry of the Interior, Community and Building of the Federal Republic of
Germany (Bonn, Germany).
Open access funding enabled and organized by Projekt DEAL.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
DATA AVAILABILITY STATEMENT
The datasets generated and analysed during the current study are
available from the corresponding authors on reasonable request.
REFERENCES
1. https://www.wada-ama.org/sites/default/files/2022-01/2022list_
final_en_0.pdf.https://www.wada-ama.org/en/prohibited-list?
Accessed 7 Aug 2022
2. https://www.wada-ama.org/sites/default/files/resources/files/
td2022mrpl_v1.0_final_eng.pdf https://www.wada-ama.org/
sites/default/files/resources/files/td2022mrpl_v1.0_final_eng.pdf
Accessed 7 Aug 2022
3. Chaabo A, de Ceaurriz J, Buisson C, Tabet JC, Lasne F. Simultaneous
quantification and qualification of synacthen in plasma. Anal Bioanal
Chem. 2011;399(5):1835-1843. doi:10.1007/s00216-010- 4565-z
4. Thomas A, Kohler M, Schänzer W, Kamber M, Delahaut P, Thevis M.
Determination of Synacthen in urine for sports drug testing by means
of nano-ultra-performance liquid chromatography/tandem mass spec-
trometry. Rapid Commun Mass Spectrom. 2009;23(17):2669-2674.
doi:10.1002/rcm.4176
5. Thomas A, Walpurgis K, Delahaut P, Fichant E, Schänzer W, Thevis
M. Determination of LongR(3)-IGF-I, R(3)-IGF-I, Des1-3 IGF-I and
their metabolites in human plasma samples by means of LC-
MS. Growth Horm IGF Res. 2017;35:33-39. doi:10.1016/j.ghir.2017.
06.002
6. Tretzel L, Thomas A, Geyer H, Delahaut P, Schänzer W, Thevis M.
Determination of Synacthen((R)) in dried blood spots for doping con-
trol analysis using liquid chromatography tandem mass spectrometry.
Anal Bioanal Chem. 2015;407(16):4709-4720. doi:10.1007/s00216-
015-8674- 6
7. Judak P, Coppieters G, Lapauw B, Van Eenoo P, Deventer K. Urinary
detection of rapid-acting insulin analogues in healthy humans. Drug
Test Anal. 2020;12(11-12):1629-1635. doi:10.1002/dta.2817
8. Judak P, Van Eenoo P, Deventer K. Utilizing ELISA-plate based
immunopurification and liquid chromatography-tandem mass spec-
trometry for the urinary detection of short- and long acting human
insulin analogues. J Pharm Biomed Anal. 2018;153:76-81. doi:10.1016/
j.jpba.2018.02.024
9. Mazzarino M, Senofonte M, Martinelli F, de la Torre X, Botre F. Detec-
tion of recombinant insulins in human urine by liquid chromatography-
electrospray ionization tandem mass spectrometry after immunoaffin-
ity purification based on monolithic microcolumns. Anal Bioanal Chem.
2019;411(30):8153-8162. doi:10.1007/s00216-019- 02203-4
10. Nedelkov D, Niederkofler EE, Oran PE, Peterman S, Nelson RW. Top-
down mass spectrometric immunoassay for human insulin and its
therapeutic analogues. J Proteomics. 2018;175:27-33. doi:10.1016/j.
jprot.2017.08.001
11. Thomas A, Schänzer W, Delahaut P, Thevis M. Immunoaffinity purifica-
tion of peptide hormones prior to liquid chromatography-mass spec-
trometry in doping controls. Methods. 2012;56(2):230-235. doi:10.
1016/j.ymeth.2011.08.009
12. Thomas A, Schänzer W, Thevis M. Determination of human insulin
and its analogues in human blood using liquid chromatography cou-
pled to ion mobility mass spectrometry (LC-IM-MS). Drug Test Anal.
2014;6(11-12):1125-1132. doi:10.1002/dta.1710
13. Thomas A, Schänzer W, Thevis M. Immunoaffinity techniques coupled
to mass spectrometry for the analysis of human peptide hormones:
advances and applications. Expert Rev Proteomics. 2017;14(9):799-807.
doi:10.1080/14789450.2017.1362338
14. Thomas A, Thevis M. Recent advances in the determination of insulins
from biological fluids. In: Makowski GS, ed. Advances in Clinical Chem-
istry. Elsevier; 2019.
15. Thomas A, Walpurgis K, Tretzel L, et al. Expanded test method
for peptides >2 kDa employing immunoaffinity purification and LC-
HRMS/MS. Drug Test Anal. 2015;7(11-12):990-998. doi:10.1002/dta.
1868
16. Cox HD, Knussmann GN, Moore C, Eichner D. Detection of insulin
analogues and large peptides >2 kDa in urine. Drug Test Anal.
2022;14(7):1264-1272. doi:10.1002/dta.3249
17. Thomas A, Benzenberg L, Bally L, Thevis M. Facilitated qualitative
determination of insulin, its synthetic analogues, and C-peptide in
human urine by means of LC-HRMS. Metabolites. 2021;11(5):309.
doi:10.3390/metabo11050309
18. Thomas A, Krombholz S, Wolf C, Thevis M. Determination of ghre-
lin and desacyl ghrelin in human plasma and urine by means of
LC-MS/MS for doping controls. Drug Test Anal. 2021;13(11-12):1862-
1870. doi:10.1002/dta.3176
19. Thomas A, Yang R, Petring S, Bally L, Thevis M. Simplified quantifica-
tion of insulin, its synthetic analogues and C-peptide in human plasma
by means of LC-HRMS. Drug Test Anal. 2020;12(3):382-390. doi:10.
1002/dta.2765
20. https://www.wada-ama.org/sites/default/files/resources/files/
isl_2021.pdf 2021. Accessed 7 Aug 2022.
21. https://www.wada-ama.org/en/resources/lab-documents/
td2021idcr https://www.wada-ama.org/en/resources/lab-
documents/td2021idcr 2021.

243 Analytical Science Advances
Research Article
doi.org/10.1002/ansa.202200027
22. Thomas A, Thevis M. Stereoisomers in sports drug testing: analyti-
cal strategies and applications. J Chromatogr A. 2022;1674:463154.
doi:10.1016/j.chroma.2022.463154
23. Knoop A, Thomas A, Fichant E, Delahaut P, Schänzer W, Thevis
M. Qualitative identification of growth hormone-releasing hormones
in human plasma by means of immunoaffinity purification and
LC-HRMS/MS. Anal Bioanal Chem. 2016;408(12):3145-3153. doi:10.
1007/s00216-016- 9377-3
24. Jette L, Leger R, Thibaudeau K, et al. Human growth hormone-
releasing factor (hGRF)1-29-albumin bioconjugates activate the GRF
receptor on the anterior pituitary in rats: identification of CJC-1295
as a long-lasting GRF analog. Endocrinology. 2005;146(7):3052-3058.
doi:10.1210/en.2004-1286
25. Thevis M, VolmerDA. Recent instrumental progress in mass spectrom-
etry: advancing resolution, accuracy,and speed of drug detection. Drug
Test Anal. 2012;4(3-4):242-245. doi:10.1002/dta.344
How to cite this article: Thomas A, Thilmany S, Hofmann A,
Thevis M. Probing for peptidic drugs (2–10 kDa) in doping
control blood samples. Anal Sci Adv. 2022;3:235–243.
https://doi.org/10.1002/ansa.202200027