Quantification of Gd species in a tetrameric Gd-based contrast agent using HPLC-ICP-MS
Journal of Chromatography A, Volume 1746, 12 April 2025, 465808: Graphical abstract
This study introduces a high-performance liquid chromatography-inductively coupled plasma-mass spectrometry (HPLC-ICP-MS) method for quantifying Gd complexes in the macrocyclic MRI contrast agent Gadoquatrane. Designed to detect multimeric Gd species formed during synthesis, this method offers high sensitivity and accurate quantification using external calibration with a generic Gd standard, providing a reliable alternative to conventional HPLC-UV in pharmaceutical analysis.
The method enables the detection of six different multimeric Gd-containing by-products at trace levels, with a limit of quantification of 38 nmol/L, corresponding to 0.004 mass% in the final product. This approach allows for effective assessment of unknown Gd-containing by-products, facilitating synthesis process optimization and drug substance control without requiring extensive reference compound synthesis, ultimately accelerating development timelines.
The original article
Quantification of Gd species in a tetrameric Gd-based contrast agent using HPLC-ICP-MS
Sonja Weishaupt, Wiebke Holkenjans, Sandra Balzer, Anne Jeremias, Michael Sperling, Martin Vogel, Uwe Karst
Journal of Chromatography A, Volume 1746, 12 April 2025, 465808
https://doi.org/10.1016/j.chroma.2025.465808
licensed under CC-BY 4.0
Selected sections from the article follow. Formats and hyperlinks were adapted from the original.
1. Introduction
Magnetic resonance imaging (MRI) has become a widely used tool in clinical diagnostics, whereby patients are routinely administered Gd-based MRI contrast agents (GBCAs) to enhance the contrast of magnetic resonance (MR) images. Due to its paramagnetic properties, Gd(III) decreases the relaxation times of surrounding protons [1]. Therefore, prior injection of a GBCA can improve the diagnosis, with the main determinants of the contrast in an MR image being proton relaxation times T1 and T2 [2]. For this purpose, solutions of inert chelated organic Gd complexes are used as contrast agents, since the free Gd(III) ion as a Ca(II) antagonist has a toxic effect on the human organism due to the inhibition of Ca channels [1,3,4].
The pharmaceutical company Bayer AG has developed a new macrocyclic GBCA with a tetrameric structure, high relaxivity, and high stability named Gadoquatrane. Via covalent linkage of four Gd-GlyMe-DOTA complexes across amide bridges, Gadoquatrane exhibits a 2–3-fold higher R1 relaxivity per Gd atom of established macrocyclic GBCAs. This increased relaxivity can be attributed to the larger molecular size and resulting rigidity of the molecule, which reduces rotational movement [5,6]. Increased relaxivity results in lower T1 and T2 relaxation times and thus more effective contrast enhancement. Hence, there is either the benefit of a lower dose for the same effect or an enhanced signal for the same dose, which can reduce the risk of adverse health effects or improve the detection or delineation of tissue lesions [7].
As medicinal products, paramagnetic GBCAs are subject to regulatory guidance. The International Conference of Harmonisation (ICH), requires the reporting of organic impurities in a drug substance exceeding a content threshold of 0.03 % and the identification of those exceeding a content threshold of 0.05 % (assuming a maximum daily dose > 2 g) [8]. In order to achieve those regulatory purity limits for a new drug substance candidate, a detailed analytical understanding of all synthesis by-products and degradation products has to be gained during the development process. Therefore, it is common practice to study early process development samples or even mother liquor from crystallization steps and to characterize and quantify all by-products of relevance. The Gd-containing by-products in Gadoquatrane can be analysed by high-performance liquid chromatography (HPLC) coupled with electrospray ionisation-high resolution mass spectrometry (ESI-HRMS). Based on the detection of accurate masses and the additional performance of HPLC-ESI tandem mass spectrometry (MS/MS) fragmentation experiments, information on the molecular formula and structure of the detected compounds can be obtained [9]. For quantification purposes, HPLC-UV is widely used in the pharmaceutical industry because of its simplicity, robustness and cost-efficiency [10,11]. The main disadvantage in quantification using HPLC-UV are species-dependent response factors due to different UV absorption behaviour. To enable a quantification of different by-products, species-specific reference compounds are necessary, which are then characterized by ESI-HRMS and NMR [12]. However, nuclear magnetic resonance (NMR) spectroscopy, a widely used analytical technique for the identification and structural elucidation of organic molecules, cannot be used to identify Gd complexes [12,13]. This is due to the large relaxation effects of Gd(III), which prevent the detection of paramagnetic NMR effects within a Gd complex [14]. These difficulties in the quantitative assessment of Gd species can be overcome by using HPLC - inductively coupled plasma - mass spectrometry (ICP-MS). Since most of the organic by-products of the GBCA are Gd-containing compounds, HPLC-ICP-MS offers a powerful alternative method to the routinely used HPLC-UV analysis in pharmaceutical industry for quantifying possible by-products [[15], [16], [17]].
ICP-MS provides an analytical technique that is highly selective and sensitive for Gd with a wide dynamic range [18]. Since molecules are almost completely atomised in the high-temperature plasma, the signal for an element will theoretically be independent of the molecular form of the element, its chemical species. In the absence of species-selective interferences, ICP-MS therefore enables accurate quantification of different Gd-species based on compound-independent calibration (CIC) using a generic Gd standard [19]. This is particularly useful for analytical profiling, as it eliminates the need to synthesize each compound individually as reference standard. In addition to the conventional use of ICP-MS for determination of inorganic impurities, this opens up a further application for this highly sensitive and accurate elemental analysis technique [20,21].
2. Experimentals
2.1. HPLC separation
Investigations were performed using an UltiMate 3000 UHPLC system of Thermo Fisher Scientific (Bremen, Germany). The reversed-phase (RP) UHPLC column ACQUITY BEH Phenyl (100 mm, 3.0 mm, 1.7 µm) of Waters™ GmbH (Eschborn, Germany) with a phenyl-modified stationary phase as well as a gradient elution was used for all investigations. The gradient started from 0 % eluent B (0–2 min), increased to 7 % B (2–32 min), ramped to 100 % B (32.1–37 min) for column equilibration, and re-equilibrated at 0 % B (37.1–47 min). For the HPLC-ICP-MS analyses of the Gadobutrol (Gadovist®, Bayer AG; Leverkusen, Germany) calibration series only, the HPLC method was adapted to three minutes run time under isocratic conditions with 100 % eluent A. To protect the mass analyser, the eluent was discharged into the waste after 34 min for 10 min via an additional 6-way valve at the end of the column. Furthermore, for all investigations, a flow rate of 0.4 mL/min, an injection volume of 5 µL and a column oven temperature of 35 °C were selected. The sample solutions were stored at 8 °C in the autosampler. For mass spectrometric detection, an aqueous solution of 0.1 vol% formic acid (FA), 0.1 % ammonium formate (HCOONH4) and 0.3 vol% acetonitrile (ACN) was used as eluent A and a mixture of an aqueous solution of 0.1 vol% FA and ACN (2:3, v/v) was used as eluent B. For UV detection, a NH4H2PO4 buffer solution with a pH value of 2.4 was prepared by dissolving 1.15 g NH4H2PO4 with 680 µL concentrated H3PO4 (85 %) in 1 L double-distilled (dd)H2O. A solution of 15 mL ACN filled up with the NH4H2PO4 buffer to 500 ml was used as eluent A, and a mixture of ACN and NH4H2PO4 buffer (3:2, v/v) was used as eluent B.
2.2. ESI-HRMS and ESI-MS/MS
For HPLC-ESI-HRMS investigations, an Exactive™ Orbitrap MS system of Thermo Fisher Scientific (Bremen, Germany) including the atmospheric pressure ionization (API) source of Exactive™ with a probe for electrospray ionization (ESI) was used. Data were collected with a m/z range of 400 to 1500, a scan rate of 0.9 spectra/sec, a resolution of 100,000, a minimum automatic gain control (AGC) target of 5∙105, a maximum inject time of 250 ms, a capillary temperature of 380 °C, an auxiliary gas flow rate of 5 L/min, a sheath gas flow rate of 40 a.u., a sweep gas flow rate of 1 a.u. and a spray voltage of 4 kV using the positive ionisation modus.
For HPLC-ESI-MS/MS investigations, a Q Exactive Orbitrap MS system of Thermo Fisher Scientific (Bremen, Germany) with a heated electrospray ionization (HESI)-II probe was used. A spray voltage of 4.0 kV in the positive ionisation mode, a sheath gas flow rate of 40.0 a.u., an auxiliary gas flow rate of 5.0 a.u. and a capillary temperature of 380 °C were used. Data were collected using the full MS/ dd-MS2 (Top N = 5) mode, with full MS parameters of a resolution of 140,000, a minimum AGC target of 3∙106, a maximum inject time of 200 ms and a scan range of 300 to 2000 m/z. For the dd-MS2 mode, parameters of a resolution of 17,500, an AGC target of 1∙105, a maximum inject time of 100 ms, an isolation window of 1.0 m/z and of 4.0 m/z, a scan range of 200 to 2000 m/z, a collision energy of 30 u.a., a minimum AGC target of 5.0∙103, an intensity threshold of 5.0∙104 and a dynamic exclusion of 10.0 s were selected.
2.3. ICP-MS
For HPLC-ICP-MS analysis, an ICP-MS 7700 system of Agilent Technologies (Ratingen, Germany) equipped with a one-piece quartz torch with narrow bore injector (1.0 mm) was used.
For the determination of the ideal internal standard, a 1 ppb standard solution of a lanthanoid mixture and Rh in 4.5:2:93.5 v/v/v ACN/HNO3/H2O was used, which was continuously introduced to the HPLC effluent via a t-piece using a peristaltic pump with an uptake speed of 0.5 rps through a tube with an inner diameter of 0.254 mm. After injection of a blank solution (double-distilled (dd)H2O) HPLC-ICP-MS data were collected by detecting the m/z 103 (Rh), 139 (La), 141 (Pr), 158 (Gd), 159 (Tb), 165 (Ho), 169 (Tm) and 175 (Lu).
For quantification, a 1 ppb standard solution of Tb in 2 % v/v HNO3/H2O was used as internal standard, which was also continuously introduced to the HPLC effluent via a t-piece using a peristaltic pump with an uptake speed of 0.5 rps through a tube with an inner diameter of 0.254 mm. HPLC-ICP-MS data were collected by detecting the m/z 158 and 160 of Gd as well as 159 of Tb with an integration time of 0.30 sec. Instrument parameters were optimized by maximizing the detected Gd signals from a 1 ppb Gd standard solution in 5:2:93 v/v/v ACN/HNO3/H2O. Based on this, for all investigations of the GBCA, a radio frequency (RF) power of 1550 W, an RF matching of 1.70 V, a sample depth of 10.0 mm, a nebulizer gas flow of 0.50 L/min and an option gas (O2) flow of 37.0 % were selected.
2.4. UV detection
UV absorption was recorded with the UltiMate 3000 RS variable wavelength detector of Thermo Fisher Scientific (Bremen, Germany) at a wavelength of 198 nm and a sampling rate of 5 Hz.
2.7. Software
For the data evaluation, the software packages of OriginPro 2021b (9.8.5.212) and the Qual Browser of Thermo Xcalibur (3.1.66.10), IntelliJ IDEA (2021.3.1) were used.
3. Results and discussion
3.1. Quantification of multimeric Gd-complexes via HPLC-ICP-MS
In pharmaceutical industry, HPLC-UV is routinely used to quantify organic by-products, requiring the use of species-specific reference standards. Due to the characteristic element Gd of Gadoquatrane and possible synthesis by-products, the established elemental analysis technique HPLC-ICP-MS offers another analytical approach via the detection of the most abundant isotope 158Gd, enabling the quantification by CIC. In order to allow for CIC, plasma ionization conditions should be constant during the whole analysis. Since the plasma is significantly modified by the presence of organic solvents, changing the plasma solvent load during gradient elution often violates the requirements for CIC. To compensate for possible matrix effects caused by the applied elution gradient, the use of an internal standard is required whose signal exhibits comparable behaviour to the changing matrix. Due to complex interplay of several factors, such as ionization properties and relative atomic mass of the elements, a prediction of the ideal internal standard is not possible, and an empirical determination is required [22]. Furthermore, in order to quantitatively determine Gd-containing compounds, it is necessary to use an external calibration standard that contains Gd and also exhibits retention on the used RP-HPLC column, ensuring comparable experimental properties. To achieve a similar retention behaviour to that of the GBCA and its by-products, it is recommended to use a Gd chelate complex comparable to the one used in the GBCA as an external standard.
For the selection of a suitable internal standard, the signals of all monoisotopic lanthanides as well as of Rh along the used elution gradient up to an organic content of 4.5 % were investigated in comparison to 158Gd. The lighter elements, 103Rh, 139La and 141Pr show a significant increase in signal intensity with increasing ACN content in the eluent (Supplementary data Fig. A1). In contrast, the heavier elements 159Tb, 169Tm and 175Lu show a similarly constant signal as 158Gd, whereby 158Gd shows a significantly lower signal intensity compared to the monoisotopic elements due to the lower isotopic abundance. Only 165Ho shows a signal increase along the elution gradient used, similar to the lighter elements. 159Tb, 169Tm and 175Lu can therefore all be classified as suitable internal standards. For the present application, 159Tb is used as internal standard. Moreover, it can be concluded that the signal response of Gd is not significantly influenced by the solvent gradient of ACN, based on the constant signal behaviour of Gd along the entire retention range.
HPLC-ICP-MS can be used to obtain quantitative information on Gd-containing compounds in terms of mol Gd/L concentration even if their structure has not been determined. In order to be able to infer the concentration in mol/L from the Gd signal, the number of complexed Gd atoms of the by-products must be taken into account in the quantification by HPLC-ICP-MS. Accordingly, the detected Gd signal of the tetrameric, macrocyclic Gd complex Gadoquatrane (see Fig. 1) is to be corrected by a factor of 0.25 in order to be comparable with the detected signal of a monomeric Gd complex, such as Gadobutrol, with regard to quantitative conclusions.
Journal of Chromatography A, Volume 1746, 12 April 2025, 465808: Fig. 1. Molecular structure of the tetrameric macrocyclic GBCA Gadoquatrane with four Gd atoms and four stereogenic centers.
Two series of analyses of a mixture of Gadobutrol and Gadoquatrane and resulting linear regression curves (Supplementary data Fig. A2) show that a CIC-based quantification of Gd-containing complexes is possible following this procedure. The integrated peak areas of Gadoquatrane were corrected by a factor of 0.25. With slopes of 0.0079 L/nmol and 0.0082 L/nmol, the two regression lines show comparable values with a deviation below 3 %. During HPLC-ICP-MS analysis, Gadobutrol is detected in a retention range from 2.5 to 2.8 min, Gadoquatrane in a retention range from 23.5 to 27.0 min (Fig. 2a). Possible analytical artifacts based on matrix effects of changing eluent composition along the elution gradient can be excluded given the comparable results of Gadobutrol and Gadoquatrane. Furthermore, a satisfactory accuracy of ± 6% within a concentration range of 44 to 1800 nM is achieved for the quantitative determination of Gadoquatrane using the linear regression of Gadobutrol as calibration (Supplementary data Fig. A2). Therefore, it can be concluded that quantification based on CIC is achieved by applying correction factors related to the number of complexed Gd atoms and by assuming constant analytical conditions, along the used elution gradient, using the internal standard Tb.
Journal of Chromatography A, Volume 1746, 12 April 2025, 465808: Fig. 2. In a, the detected relative peak of 158Gd/159Tb is shown over the entire retention range of 34 min of an aqueous solution of Gadobutrol (5 µmol/L) and a Gadoquatrane sample from the early process development (500 µmol/L), analysed by HPLC-ICP-MS. In b, an enlarged section of the chromatogram in a is shown. The absolute (RT) and relative to first diastereomer of Gadoquatrane (RRT) retention times of the detected compounds are indicated (b), as well as the peaks of Gadobutrol and diastereomers of Gadoquatrane (GQ (D1, D2, D3)) (a).
Gadobutrol can be classified as a suitable standard substance for external calibration. The use of Gadobutrol instead of Gadoquatrane itself has several advantages. Gadobutrol shows only a single peak under the selected chromatographic conditions in contrast to Gadoquatrane, where three different peaks are detected due to its three diastereomers [23]. Since the identification of the different diastereomers is not relevant for the quantification of by-products, the signals are labelled as the first (D1), second (D2) and third (D3) diastereomer (Fig. 2). In addition, there is a considerable time saving, as Gadobutrol elutes after about 2.6 min under the initial isocratic elution conditions at 100 % eluent A (Fig. 2). Thus, the calibration series can be recorded at 100 % eluent A with an analysis time of only 3 min per individual run with the same set-up as for the samples (Supplementary Fig. A3).
4. Conclusions
The application of HPLC-ICP-MS to multimeric Gd-containing complexes provides a selective and sensitive approach for quantification of Gd-containing by-products in an early process development sample of the GBCA Gadoquatrane. HPLC-ICP-MS offers a species-independent method with significantly improved LOQs (38 nmol/L for monomeric, 19 nmol/L for dimeric, 13 nmol/L for trimeric and 10 nmol/L for tetrameric Gd complexes) compared with HPLC-UV. This enabled the detection and quantification of six different species of trimeric and tetrameric Gd complexes. Thus, compared to UV detection, which is routinely used in pharmaceutical industrial applications, HPLC-ICP-MS offers a sensitivity increase of over 15 times [11]. Additionally, HPLC-ICP-MS allows for quantification by CIC through the detection of the element Gd, whereas in HPLC-UV, the species might have different response factors due to varying absorption behaviour. Similar to highly sensitive HPLC-ESI-MS, HPLC-UV would require the use of species-specific reference standards for quantification [12].
The potential application of ICP-MS techniques in terms of analytical characterization for the tetrameric, macrocyclic GBCA Gadoquatrane can therefore be extended to the quantitative determination of mono- and polymeric Gd complexes in addition to the conventional determination of elemental impurities [20]. The expanded scope of HPLC-ICP-MS through the newly developed fit-for-purpose method enables a more accurate and comprehensive evaluation of a Gadoquatrane process development sample with respect to quantifying Gd-containing by-products than conventional routine methods, even in trace concentrations. Thus, the method has great potential for comparison of different process development samples. In combination with ESI-MS, by-products can be rapidly identified and quantified without the need for the time-consuming process of impurity synthesis for quantification purposes using species-specific standards.
