Combined use of algorithms for peak picking, peak tracking and retention modelling to optimize the chromatographic conditions for liquid chromatography–mass spectrometry analysis of fluocinolone acetonide and its degradation products

doi:10.1016/j.aca.2011.07.047

Analytica Chimica Acta

Volume 704, Issues 1–2, 17 October 2011, Pages 180-188

https://doi.org/10.1016/j.aca.2011.07.047 Get rights and content

Abstract

A strategy for rapid optimization of liquid chromatography column temperature and gradient shape is presented. The optimization as such is based on the well established retention and peak width models implemented in software like e.g. DryLab and LC simulator. The novel part of the strategy is a highly automated processing algorithm for detection and tracking of chromatographic peaks in noisy liquid chromatography–mass spectrometry (LC–MS) data. The strategy is presented and visualized by the optimization of the separation of two degradants present in ultraviolet (UV) exposed fluocinolone acetonide. It should be stressed, however, that it can be utilized for LC–MS analysis of any sample and application where several runs are conducted on the same sample. In the application presented, 30 components that were difficult or impossible to detect in the UV data could be automatically detected and tracked in the MS data by using the proposed strategy. The number of correctly tracked components was above 95%. Using the parameters from the reconstructed data sets to the model gave good agreement between predicted and observed retention times at optimal conditions. The area of the smallest tracked component was estimated to 0.08% compared to the main component, a level relevant for the characterization of impurities in the pharmaceutical industry.

Graphical abstract

Highlights

► Highly automated algorithms for peak detection, peak tracking, and retention modelling have been combined. ► The strategy was exemplified with a typical application of LC–MS method development for stability studies of drug substances. ► Fast and efficient generation of accurate results was obtained at levels relevant for pharmaceutical products.

Introduction

The development of liquid chromatography (LC) methods for the determination of impurities and degradation products in drug products is a central task within the pharmaceutical industry. All impurities and degradation products present at levels above 0.1% of the active substance must be quantified, identified and qualified. The most commonly used techniques are LC combined with ultraviolet (UV)-detection for quantitative analysis and LC combined with mass spectrometry (MS) for identification [1]. MS detection is also frequently used during the development of LC–UV methods to facilitate peak tracking, i.e. identify how peaks move as the chromatographic conditions change.

Several efforts have been made to reduce the time needed for LC analysis as well as the LC method development itself [2], [3]. Faster separations can be achieved by improvements in the instrumentation such as the use of monolithic columns or systems employing columns packed with sub 2 μm porous or 2.7 μm superficially porous particles in combination with elevated temperature and back pressures (600–1200 bar). A reduction of the time needed for method development has also been achieved by the introduction of automated column and solvent selection valves that reduces user intervention. A number of method development strategies have been developed to decrease the total number of trials needed to find optimal conditions for the determination of degradation products and/or impurities [4], [5], [6], [7], [8], [9], [10], [11], [12], [13].

Optimization/prediction software has been developed that, based on a small number of experiments, can guide the analyst towards optimal conditions. DryLab [14], [15], LC simulator/Autochrom [16], ChromSword [10], [17], Osiris [18] and LabExpert [9] are examples on data programs that have been developed for these tasks and in some cases even take control over the instrument and is capable to make decisions on what chromatographic parameters to try next to reach optimal conditions. Many of the optimization programs use simple but efficient relations between variables and responses that require a minimum number of analytical runs. Similar strategies based upon experimental design and multivariate evaluation has also been described in the literature, for example by Moberg et al. [19], [20], [21] and Popovic et al. [4]. Also, a combination of these approaches has been reported [6], [7].

The development part of AstraZeneca has adopted a three-step strategy which appears to be common in the pharmaceutical industry. An initial scouting step where a suitable pH value and buffer are selected based on the structure and pK_a values of the drug substance, by prior knowledge or by screening mobile phases with different pH. The purpose with this step is to find buffers that produce a good peak shape as well as sufficient retention. A second step comprises a generic screening of different combinations of columns [22], organic modifiers and buffers with different pH values which has been selected to maximize selectivity differences. The purpose with this step is to determine which components are present in the samples as well as to find a good starting point for the subsequent optimization. A third and final step is the actual optimization in which the temperature of the column and the shape of the gradient are simultaneously optimized. This optimization is based on well known retention time and peak width models [23], which have been implemented in the commercial software like DryLab [14], LC simulator/Autochrom [16], ChromSword [10] and Osiris [18].

These software and most method development strategies require peak tracking. In most cases a challenging and time consuming part of the method development process. Unfortunately, not much has been reported on how the time needed for peak tracking can be reduced. Diode array detection (DAD) is often of limited use since impurities and degradants related to the active component typically are present at low level where the quality of the UV spectrum is poor. An additional complication is that related impurities tend to have a spectrum very similar to the active component. Peak tracking based on UV peak area is often not reliable due to poor signal to noise ratio, which results in an uncertain determination. Another complication is that degradation may change peak areas over time [24]. MS detection is probably the most reliable tool for peak tracking. It is, however, often a challenge to locate chromatographic peaks at relevant levels in the relatively noisy LC–MS data matrices. Another problem is the large amount of data produced, which is quite time consuming to process manually. This is currently considered to be one of the largest bottlenecks in the method development process. Hence, there is a need for automated processing methods for detection and tracking of chromatographic peaks.

In this work such an automated processing method is described. Its application is exemplified by the optimization of a LC–MS method for the determination of two degradants present in fluocinolone acetonide after exposure to UV light [25], [26]. The objective is to show how our previously developed processing methods for signal enhancement, peak picking [27] and peak tracking [28] can be combined to facilitate retention modelling and the optimization of chromatographic conditions.

Section snippets

Sample preparation

A solution of fluocinolone acetonide (CAS 67-73-2, Sigma–Aldrich, St. Louis, MO) was prepared at 1 mg mL⁻¹ in 50/50 (v/v) acetonitrile/water. This solution was then exposed to the ambient light and temperature in a transparent glass vessel for 24 h to accelerate degradation of the steroid. The final solution was stored in darkness at −18 °C pending analysis.

Chemical analysis

The LC–DAD–MS system used comprised two Micro Series 2000 LC pumps (PerkinElmer, Inc., Wellesley, MA), an autoinjector (G1367A, Agilent

Results and discussion

The scope of this article is to demonstrate how our previously developed processing methods for signal enhancement, peak picking [27] and peak tracking [28] can be combined to facilitate retention modelling and the optimization of chromatographic conditions, i.e. the final step in a typical method development strategy. It should be stressed, however, that these processing methods are equally applicable for the detection of relevant impurities in the screening step.

Conclusions

In this paper it has been shown how our recently developed algorithms for highly automated data processing for detection and tracking of chromatographic peaks in noisy LC–MS data can be applied to facilitate LC–MS method development. The algorithms were shown to be successful in the optimization of a separation of components present in a challenging sample containing a large number of components present at low level and with similar or, in some cases identical, mass spectra impeding the

Acknowledgements

We would like to thank AstraZeneca, Analytical Development, R&D Lund, Sweden and European regional development fund of the European Union for financial support.

References (32)

S. Baertschi
Trends Anal. Chem.
(2006)
B.A. Olsen et al.
Trends Anal. Chem.
(2006)
I. Molnár et al.
J. Chromatogr. A
(2010)
W. Li et al.
J. Chromatogr. A
(2003)
R.M. Krisko et al.
J. Chromatogr. A
(2006)
E.F. Hewitt et al.
J. Chromatogr. A
(2006)
M. Dumarey et al.
Anal. Chim. Acta
(2009)
I. Molnar
J. Chromatogr. A
(2002)
M.R. Euerby et al.
J. Chromatogr. A
(2003)
N. Lundell et al.
J. Chromatogr.
(1993)

T. Teutenberg

Anal. Chim. Acta

(2009)

M.J. Fredriksson et al.

J. Chromatogr. A

(2010)

C. Liu et al.

J. Chromatogr. A

(1999)

B.O. Keller et al.

Anal. Chim. Acta

(2008)

G. Chen et al.

J. Mass Spectrom.

(2007)

I. Popovic et al.

Chromatographia

(2008)

Cited by (13)

Automated method development in high-pressure liquid chromatography
2024, Journal of Chromatography A
Method development in liquid chromatography is a crucial step in the optimization of analytical separations for various applications. However, it is often a challenging endeavour due to its time-consuming, resource intensive and costly nature, which is further hampered by its complexity requiring highly skilled and experienced scientists. This review presents an examination of the methods that are required for a completely automated method development procedure in liquid chromatography, aimed at taking the human out of the decision loop. Some of the presented approaches have recently witnessed an important increase in interest as they offer the promise to facilitate, streamline and speed up the method development process. The review first discusses the mathematical description of the separation problem by means of multi-criteria optimization functions. Two different strategies to resolve this optimization are then presented; an experimental and a model-based approach. Additionally, methods for automated peak detection and peak tracking are reviewed, which, upon integration in an instrument, allow for a completely closed-loop method development process. For each of these approaches, various currently applied methods are presented, recent trends and approaches discussed, short-comings pointed out, and future prospects highlighted.
Computer-driven optimization of complex gradients in comprehensive two-dimensional liquid chromatography
2023, Journal of Chromatography A
Method development in comprehensive two-dimensional liquid chromatography (LC × LC) is a complicated endeavor. The dependency between the two dimensions and the possibility of incorporating complex gradient profiles, such as multi-segmented gradients or shifting gradients, renders method development by “trial-and-error” time-consuming and highly dependent on user experience. In this work, an open-source algorithm for the automated and interpretive method development of complex gradients in LC × LC-mass spectrometry (MS) was developed. A workflow was designed to operate within a closed-loop that allowed direct interaction between the LC × LC-MS system and a data-processing computer which ran in an unsupervised and automated fashion. Obtaining accurate retention models in LC × LC is difficult due to the challenges associated with the exact determination of retention times, curve fitting because of the use of gradient elution, and gradient deformation. Thus, retention models were compared in terms of repeatability of determination. Additionally, the design of shifting gradients in the second dimension and the prediction of peak widths were investigated. The algorithm was tested on separations of a tryptic digest of a monoclonal antibody using an objective function that included the sum of resolutions and analysis time as quality descriptors. The algorithm was able to improve the separation relative to a generic starting method using these complex gradient profiles after only four method-development iterations (i.e., sets of chromatographic conditions). Further iterations improved retention time and peak width predictions and thus the accuracy in the separations predicted by the algorithm.
Algorithm for tracking peaks amongst numerous datasets in comprehensive two-dimensional chromatography to enhance data analysis and interpretation
2023, Journal of Chromatography A
Analytical data processing often requires the comparison of data, i.e. finding similarities and differences within separations. In this context, a peak-tracking algorithm was developed to compare multiple datasets in one-dimensional (1D) and two-dimensional (2D) chromatography. Two application strategies were investigated: i) data processing where all chromatograms are produced in one sequence and processed simultaneously, and ii) method optimization where chromatograms are produced and processed cumulatively. The first strategy was tested on data from comprehensive 2D liquid chromatography and comprehensive 2D gas chromatography separations of academic and industrial samples of varying compound classes (monoclonal-antibody digest, wine volatiles, polymer granulate headspace, and mayonnaise). Peaks were tracked in up to 29 chromatograms at once, but this could be upscaled when necessary. However, the peak-tracking algorithm performed less accurate for trace analytes, since, peaks that are difficult to detect are also difficult to track. The second strategy was tested with 1D liquid chromatography separations, that were optimized using automated method-development. The strategy for method optimization was quicker to detect peaks that were still poorly separated in earlier chromatograms compared to assigning a target chromatogram, to which all other chromatograms are compared. Rendering it a useful tool for automated method optimization.
Implementation of analytical quality by design and green chemistry principles to develop an ultra-high performance liquid chromatography method for the determination of Fluocinolone Acetonide impurities from its drug substance and topical oil formulations
2022, Journal of Chromatography A
Citation Excerpt :
The following assay methods were reported for various formulations by using UV [3–4]; HPLC [5–14]; high-performance thin layer chromatography (HPTLC) [15]; and UPLC/ Tandem MS [16]. A study was conducted by Fredriksson et al. to determine the impurities by LC-MS [17], but no impurities were specified. Since no impurities method was reported for the drug product, we aimed to develop a suitable single reversed phase (RP)-UPLC method to determine impurities in the drug substance and its topical oil formulation.
An anti-inflammatory skin condition is treated with fluocinolone acetonide (FLA), a synthetic corticoid. The current study aims to develop a stability-indicating UPLC method for the determination of impurities present in fluocinolone acetonide and its topical oil formulation. The method development was performed by implementing Analytical Quality by Design (AQbD) and green chemistry principles. A detailed risk assessment was conducted based on the cause-and-effect relationship. d-optimal split-plot design was employed to screen the critical method parameters (CMPs). The central composite design (CCD) was employed to optimize the final method conditions. p-values for the model and lack of fit were <0.0001 and >0.05, respectively, which indicates the best fit statistical model for the studied responses (peak resolutions R1 – R5). The critical method attributes (CMAs) and CMPs such as the ratio of ACN: Water in mobile phase-B as 600:400 (v/v), the ratio of mobile phase-A & B in initial gradient program as 60:40, flow rate as 0.3 mL min⁻¹, and column oven temperature as 50 °C were optimized from the CCD. The best possible separation among all components was achieved with a gradient elution using Waters Acquity UPLC HSS C₁₈, 100 mm × 2.1 mm, 1.8 µm analytical column. The optimized gradient program is time (min)/%B: 0.0/40, 1.5/40, 6.0/60, 8.0/70, 9.0/80, 12.0/100, 15.0/100, 15.1/40 & 18.0/40. Optimization of diluent is highly critical for any oil-based formulations. The experimental results show that acetonitrile is the most suitable diluent for the current study. The method validation was executed in compliance with ICH and USP 〈1225〉 guidelines. Mean recovery of the impurities ranged between 95.7 and 105.7%, the correlation coefficient(r) was> 0.999, the RSD values (n = 6) ranged between 0.9 – 3.2% across the range for LOQ – 150% levels. The peaks from the specificity study did not interfere with the known and active analyte peaks. The major degradation products were identified as Imp-C, B, and A, and established their degradation pathways from FLA based on the stress studies. The method greenness was evaluated using GAPI, AGREE and analytical eco scale and found that the method is green.
Peak-tracking algorithm for use in comprehensive two-dimensional liquid chromatography – Application to monoclonal-antibody peptides
2021, Journal of Chromatography A
Citation Excerpt :
When comparing chromatograms of different samples, compounds that cannot be tracked may be impurities in specific samples, which may be highly interesting for many applications. Peak-detection methods were developed for 1D-LC [24–26] and 2D-LC [27] and subsequently peak-tracking algorithms have been written for LC-DAD [28,29] and LC-MS data [25,30], including our own algorithm [31]. The application of peak tracking to LC × LC-MS data, however, is accompanied by multiple challenges.
A peak-tracking algorithm was developed for use in comprehensive two-dimensional liquid chromatography coupled to mass spectrometry. Chromatographic peaks were tracked across two different chromatograms, utilizing the available spectral information, the statistical moments of the peaks and the relative retention times in both dimensions. The algorithm consists of three branches. In the pre-processing branch, system peaks are removed based on mass spectra compared to low intensity regions and search windows are applied, relative to the retention times in each dimension, to reduce the required computational power by elimination unlikely pairs. In the comparison branch, similarity between the spectral information and statistical moments of peaks within the search windows is calculated. Lastly, in the evaluation branch extracted-ion-current chromatograms are utilized to assess the validity of the pairing results. The algorithm was applied to peptide retention data recorded under varying chromatographic conditions for use in retention modelling as part of method optimization tools. Moreover, the algorithm was applied to complex peptide mixtures obtained from enzymatic digestion of monoclonal antibodies. The algorithm yielded no false positives. However, due to limitations in the peak-detection algorithm, cross-pairing within the same peaks occurred and six trace compounds remained falsely unpaired.
Rational design of mixtures for chromatographic peak tracking applications via multivariate selectivity
2019, Analytica Chimica Acta: X
Citation Excerpt :
Whether the goal is to parameterize retention or simply screen retention behavior for a large number of solutes, the analyst will want to group the solutes in as few mixtures as possible while retaining the ability to determine which peaks correspond to which compounds that are present in the mixtures. The ability to identify peaks in chromatograms over a range of experimental conditions is generally referred to as peak tracking in the literature [5,7,21–23]. In some cases, this task can be fairly straightforward; for example, if the set of target solutes consists of one or more homologous series of compounds, the members of each homologous series can be prepared in a single mixture, as the retention order of the compounds is known.
Chromatographic characterization and parameterization studies targeting many solutes require the judicious choice of operating conditions to minimize analysis time without compromising the accuracy of the results. To minimize analysis time, solutes are often grouped into a small number of mixtures; however, this increases the risk of peak overlap. While multivariate curve resolution methods are often able to resolve analyte signals based on their spectral qualities, these methods require that the chromatographically overlapped compounds have dissimilar spectra. In this work, a strategy for grouping compounds into sample mixtures containing solutes with distinct spectral and, optionally, with distinct chromatographic properties, in order to ensure successful solute resolution either chromatographically or with curve resolution methods is proposed. We name this strategy rational design of mixtures (RDM). RDM utilizes multivariate selectivity as a metric for making decisions regarding group membership (i.e., whether to add a particular solute to a particular sample). A group of 97 solutes was used to demonstrate this strategy. Utilizing both estimated chromatographic properties and measured spectra to group these 97 analytes, only 12 groups were required to avoid a situation where two or more solutes in the same group could not be resolved either chromatographically (i.e., they have significantly different retention times) or spectrally (i.e., spectra are different enough to enable resolution by curve resolution methods). When only spectral properties were utilized (i.e., the chromatographic properties are unknown ahead of time) the number of groups required to avoid unresolvable overlaps increased to 20. The grouping strategy developed here will improve the time and instrument efficiency of studies that aim to obtain retention data for solutes as a function of operating conditions, whether for method development or determination of the chromatographic parameters of solutes of interest (e.g., k_w).

View all citing articles on Scopus

View full text

Combined use of algorithms for peak picking, peak tracking and retention modelling to optimize the chromatographic conditions for liquid chromatography–mass spectrometry analysis of fluocinolone acetonide and its degradation products

Abstract

Graphical abstract

Highlights

Introduction

Section snippets

Sample preparation

Chemical analysis

Results and discussion

Conclusions

Acknowledgements

Trends Anal. Chem.

Trends Anal. Chem.

J. Chromatogr. A

J. Chromatogr. A

J. Chromatogr. A

J. Chromatogr. A

Anal. Chim. Acta

J. Chromatogr. A

J. Chromatogr. A

J. Chromatogr.

Anal. Chim. Acta

J. Chromatogr. A

J. Chromatogr. A

Anal. Chim. Acta

J. Mass Spectrom.

Chromatographia