2022 Volume 70 Issue 3 Pages 226-229
Quantitative proton NMR (qHNMR) methodology was employed for the stoichiometric (free base and the corresponding counterion) assessment of a ticagrelor process impurity, also referred to in the United States Pharmacopeia (USP), Pharmacopeial Forum as Ticagrelor Related Compound A (RC A), [(1R,2S)-2-(3,4-difluorophenyl)cyclopropan-1-amine (R)-mandelate], also called as Tica amine mandelate, a critical impurity that, when present during manufacturing, has a limit of not more than 0.0008%. The Tica amine is also a listed impurity E in the Ticagrelor monograph, in European Pharmacopeia. Because there was no existing NMR spectroscopic method in the literature specific to quantify the counterion (mandelic acid) in Ticagrelor RC A, this study aimed to fill the gap. Accurate stoichiometric measurement of this impurity serves to enhance product quality in the manufacturing of the ticagrelor active pharmaceutical ingredient (API). Using ethylene carbonate as an internal standard (IS), the qHNMR analysis on Ticagrelor impurity, revealed many key characteristics of the test mixture composition, including (free base and counterion). The results demonstrate that qHNMR has great potential for addressing several key quality attributes associated with chemical analyses such as detection, identification, quantification, and purity determination, as well as deriving molecular stoichiometry, all from the single proton spectrum.
Ticagrelor is the first approved reversible oral antiplatelet agent that also works to prevent vasospasm and increases myocardial blood flow.1–3) The principal pharmacopeial means of its assessment4) include identification by IR spectroscopy and quantitative analysis by HPLC; the latter permitting separation of ticagrelor from its product- and process-related impurities. Several published synthetic routes of ticagrelor include a key intermediate, referred to as Tica amine, comprised of the difluoro phenyl ring with a cyclopropanamine moiety; in the proposed United States Pharmacopeia (USP) monograph, it is referred to as Ticagrelor Related Compound A (RC A).4) As a critical impurity, if present in the ticagrelor active pharmaceutical ingredient, RC A has a limit of not more than 0.0008%. The Tica amine is also a listed impurity E in the Ticagrelor monograph of European Pharmacopeia and has a limit of 8 ppm.5) Ticagrelor RC A is synthesized and isolated as a salt of mandelic acid (Fig. 1). The use of (R)-mandelic acid improves the stability and allows the selective crystallization and purification of the 1R,2S Tica amine isomer as a key chiral building block commonly used in the ticagrelor synthesis.6–11)
The use of quantitative techniques within the realm of NMR spectroscopy has been gaining momentum.12) Quantitative proton NMR (qHNMR) has recently gained significant traction and attracted the attention of analytical community.13–15) Examples of recent applications include assessment of pharmaceuticals.16–22) Accurate stoichiometry determination is paramount to proper pharmaceutical manufacturing, dosing, and subsequent quality control. Typically achieved by means of chromatographic separation, assay of a drug substance and its structurally closely related molecules, commonly seen as product- and process-related impurities, and degradation products, is the most sought-after analytical data during quality control testing in pharmaceutical testing.
Free base compounds that are often difficult to isolate after synthesis are frequently rendered as stable salts upon neutralization with acids.23) Thus, counterions in pharmaceuticals are common to obtain molecular forms of the active substances that are suitable for pharmaceutical purposes with enhanced stability, solubility, and other physical properties.24–27) Chiral counter ions also aid in the isolation and purification of the desirable pharmacologically active enantiomers. Stoichiometric excesses or defects in the ratio of counterions to free bases are quite possible during pharmaceutical manufacturing. Conventional approach for the counterion determination in active pharmaceutical ingredient (API) and drug product is titrimetry, however, in recent decades the analysis has been shifting toward a combination of chromatography and qHNMR.28,29) Because it is critical to accurately determine the presence of the counterion in the Ticagrelor RC A, the present study aimed to fully characterize the critical quality attributes of mandelic acid such as detection, identification, quantification, purity determination and stoichiometry verification, using the method developed by qHNMR test. In short, qHNMR provides the ability to analyze pharmaceutical salt mixtures exhaustively and accurately.
The study summary shown in Fig. 2 outlines the representative quality attributes of chemical analyses by qHNMR and the expansions to show the selected quantitative signal regions refer to Supplementary Figs. S1–S3.
All the materials, methods, reagents, and instrumentation details are provided in the supplementary material S1 Experimental Details. The purity value was calculated based on the average of five sample preparations with five replicate determinations (5 × 5) for each sample, using the published Excel worksheet.13) The five sample preparations addressed sample homogeneity because the analyte was taken from different parts of the vial. Notably, all test sample solutions were stable (0–24 h). The signals used for the quantitation were those associated with ethylene carbonate at about 4.53 ppm, free base Tica amine at about 2.76 ppm with a minimum signal to noise (S/N) of 4374, and the aromatic methine protons of the counterion mandelate at about 7.49 ppm with a S/N of 2130.
The method development is comprised of part (a) sample preparation, and part (b) NMR optimization. For successful qHNMR method development, the key criterion is to have a test sample with a compatible solvent system that completely dissolves both the analyte and the IS.
Part (a) Methanol-d4 was used as the NMR solvent and ethylene carbonate as the (IS), facilitated stable test sample solutions suitable for quantitative analysis. The structure verification was performed in dimethyl sulfoxide (DMSO)-d6 a priori to quantitation. The one dimensional (1D) and 2D-NMR experimentation allowed complete assignment of 1H-NMR, 13C, 15N and 19F signals of Ticagrelor RC A (without internal standard) and are reported in Supplementary Table S1 (structural verification, see Supplementary Figs. S4–S10).
Part (b). The optimized 400 MHz NMR instrumental set-up consists of acquisition time (AQ) of 4.0 s; pulse width (P1) of 8.6 µs; receiver gain adjusted (RGA) used; and temperature (TE), 298 K. The method recovery (100.66%) and repeatability (99.82%) are shown in Supplementary Table S2. An example of the method precision (99.83%) is shown in Supplementary Table S3 for one of the representative test samples. The molar ratio of Ticagrelor RC A to IS determined by qHNMR versus the gravimetric molar ratio of Ticagrelor RC A to IS, is shown in Fig. 3. Linear regression yielded a correlation coefficient of 1.000 and a regression line of y=1.0001x−0.0004. The gravimetric moles were obtained using the weighed amounts of Ticagrelor RC A, and ethylene carbonate and their corresponding molecular weights. The theoretical moles were based on the weighed amounts and then factoring the purity by mass balance of both the Ticagrelor RC A and ethylene carbonate.
The specificity and selectivity of the method were evaluated based on the absence of overlap of at least one signal of each component with the IS. The accuracy and precision results are summarized in Supplementary Table S3. The purity of 99.82% determined using qHNMR was comparable to the orthogonally derived value of 99.01% obtained from [100 −(0.50 + 0.38 + 0.1 + 0.01)] by the mass balance (gravimetric) method. The mass balance value has been established by subtracting the contents of organic impurities (HPLC), water content (Karl Fischer), residual solvents (GC), and inorganic residue (Residue on Ignition) determined using the proposed USP Ticagrelor monograph method and relevant USP compendial methodologies; the results were 0.50, 0.38, 0.10, and 0.01%, respectively. While HPLC procedures are quantitative with good precision, the generous use of organic solvents, additional reagents, and specific columns compatible to chemistries involved, and time demanding method development works to achieve high accuracy and precision often demand extensive resources with respect to chromatographic means of identification and quantification. The lack of universal detector, wavelength dependent responses, maintenance and conditioning of columns, recurring cost of replacement columns, also add to the longer-term cost considerations. Likewise, titration methods are also generous on the reagents used and accuracy may be impacted by human error towards end point determinations. Unlike chromatography detectors, probes in NMR spectroscopy the universal experiment to begin with for any organic molecule is the standard proton NMR. Additionally, the qHNMR test is in a unique position to decipher the purities of the individual components, the free base and the counterion, as shown below. This information is of significant value and is apparent from the landscape of a simple proton spectrum of RC A; to yield the same information, an elaborate chromatographic analysis is typically employed. Quantitative chromatographic procedures for individual components [e.g., free base (a) and counterion (b)] are often not readily available and are largely dependent on the availability of chemically identical primary standards and custom-developed analytical procedures. Lastly, the overall 0.06% relative standard deviation (RSD) for the purity determination (Supplementary Table S2) highlights the exemplary analytical figures of merit typical for a well-developed qHNMR method. The content of Tica amine (free base) and mandelic acid (counterion) in Ticagrelor RC A, as determined by qHNMR, are shown in Supplementary Table S4.
Stoichiometry Verification by qHNMRThe adoption of quantitation by NMR has an immediate advantage of accurate confirmation of stoichiometry without the need of developing additional chromatographic procedures or establishing costly pharmacopeial reference standards.
More specific to counterion assessment, in a commonly employed titrimetric determination, one can encounter additional challenges such as standardization of the reagents used, compounded by methodological or instrumental errors. Crucially, stoichiometry can vary between multiple batches during manufacturing. While the absence of specific methods for counterion measurement in a specific salt mixture is a significant challenge, the lack of primary standards requisite for quantitative counterion evaluation is nearly insurmountable. Additionally, the conventional analytical evaluation of free base by chromatographic means and counter ion by titration method is a two-pronged approach, whereas here we demonstrate that by using a single method qHNMR we get both free base and counter ion evaluation. Thus, qHNMR offers a viable and green alternative to chromatographic separation and quantitation, where the use of large volumes of organic solvents is a prerequisite. The use of more resources such as columns, reagents to tailor the separation, multiple sample preparations and dilutions, lead to voluminous workflows which can be completed by passed by qHNMR adoption. In this study, the counterion mandelic acid content (47.26%) was consistent with the theoretical value (47.36%). The qHNMR-verified stoichiometric ratio and the average mandelic acid content are shown in Supplementary Table S5.
Ascertaining the molar ratio of these two components (1 : 1 by qHNMR) orthogonally supports mass balance calculations.
A rapid, precise, and accurate qHNMR method has been established for quantification of free base and its counterion in Ticagrelor RC A, a critical process impurity associated with the pharmaceutical manufacturing of ticagrelor. The developed method utilized qHNMR spectroscopy to verify the stoichiometric ratio of the multi-component pharmaceutical salt mixture. The acquisition of a naturally abundant and highly sensitive 1H spectrum under quantitative conditions yielded many quality attributes such as identification, quantification, purity assessment, and stoichiometry. These supply indisputable evidence of the advantages and insights into pharmaceutical manufacturing and pharmacopeial analysis to be realized from implementation of qHNMR. Additionally, such qHNMR method for estimation of mandelate, further facilitates determination of purity of enantiomer Tica-amine, during manufacturing of synthetic intermediates and crystallization processes. Such determinations by qHNMR have a potential that could lead to efficient quality control measures towards highly selective isolations attempted by crystallization processes, common in pharmaceutical manufacturing. While organic counter ions can possibly be tested by qNMR certain inorganic counter ions mainly represented by hydrochlorides, different techniques are employed.
The study underscores the role of qHNMR as a mature orthogonal platform for assessment of numerous quality attributes of the active pharmaceutical ingredients, associated process impurities and drug products, thereby facilitating development of the next generation pharmacopeial standards and ensuring an improved safety net for quality medications.
The authors would like to acknowledge the support and encouragement from the USP management. The authors are thankful to the USP Reference Standard Laboratories for providing support with mass balance analytical testing. Sincere thanks to USP internal scientific review comments by Drs. Gabriel Giancaspro, Dinesh Chalasani and Qun Xu, and Anton Bzhelyansky, and editorial assistance from Christina Chase. The NMR structural data was acquired by Dinesh Chalasani.
The authors declare no conflict of interest. This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
This article contains supplementary materials.
