Development of a Predictive Multiple Reaction Monitoring (MRM) Model for High-Throughput ADME Analyses Using Learning-to-Rank (LTR) Techniques.

Multiple Reaction Monitoring (MRM) is an important MS/MS technique commonly used in drug discovery and development, allowing for the selective and sensitive quantification of compounds in complex matrices. However, compound optimization can be resource intensive and requires experimental determination of product ions for each compound. In this study, we developed a Learning-to-Rank (LTR) model to predict the product ions directly from compound structures, eliminating the requirement for MRM optimization experiments. Experimentally determined MRM conditions for 5757 compounds were used to develop the model. Using the MassChemSite software, theoretical fragments and their mass-to-charge ratios were generated, which were then matched to the experimental product ions to create a data set. Each possible fragment was ranked based on its intensity in the experimental data. Different LTR models were built on a training split. Hyperparameter selection was performed using 5-fold cross validation. The models were evaluated using the Normalized Discounted Cumulative Gain at top k (NDCG@k) and the Coverage at top k (Coverage@k) metrics. Finally, the model was applied to predict MRM conditions for a prospective set of 235 compounds in high-throughput Caco-2 permeability and metabolic stability assays, and quantification results were compared to those obtained with experimentally acquired MRM conditions. The LTR model achieved a NDCG@5 of 0.732 and Coverage@5 of 0.841 on the validation split, and its predictions led to 97% of biologically equivalent results in the Caco-2 permeability and metabolic stability assays.

Rapid Compound Integrity Assessment for High-Throughput Screening Hit Triaging.

Hits from high-throughput screening (HTS) assays are typically evaluated using cheminformatics and/or empirical approaches before a decision for follow-up (activity confirmation and/or sample resynthesis) is made. However, the compound integrity (i.e., identity and purity) of these hits often remains largely unknown at this stage, since many compounds in the screening collection could undergo various changes such as degradation, polymerization, and precipitation during storage over time. When compound integrity is actually assessed for HTS hits postassay to address this issue, the process often increases the overall cycle time by weeks due to the reacquisition of the samples and the lengthy liquid chromatography-ultraviolet/mass spectrometric analysis time. Here we present a novel approach where compound integrity data are collected concurrently with the concentration-response curve (CRC) stage of HTS, with both assays occurring either in parallel on two distributions from the same liquid sample or serially using the original source liquid sample. The rapid generation of compound integrity data has been enabled by a high-speed ultra-high-pressure liquid chromatography-ultraviolet/mass spectrometric platform capable of analyzing ~2000 samples per instrument per week. From this parallel approach, both compound integrity and CRC potency results for screening hits become available to medicinal chemists at the same time, which has greatly enhanced the decision-making process for hit follow-up and progression. In addition, the compound integrity results from recent hits provide a real-time and representative "snapshot" of the sample integrity of the entire compound collection, and the data can be used for in-depth analyses of the screening collection.

Ultrahigh-Throughput and Chromatography-Free Bioanalysis of Polar Analytes with Acoustic Ejection Mass Spectrometry.

Bioanalysis of polar analytes using liquid chromatography-tandem mass spectrometry (LC-MS/MS) remains a significant challenge because of their poor chromatographic retention on the commonly used reversed-phase LC columns and the resulting severe ionization suppression from coeluting matrix components. Here we present a novel approach to perform ultrahigh-throughput and chromatography-free bioanalysis of polar compounds using a prototype acoustic ejection mass spectrometer (AEMS) platform. Previously developed for direct analysis of solid or liquid samples by MS, the open port interface (OPI) has recently been modified and coupled to an acoustic nanoliter dispenser to enable high-speed direct MS analysis from 384-well plates with a reported speed as fast as 0.5 s/sample. Ionization suppression was reduced due to the >1000 fold dilution of the original sample by the carrier solvent in the AE-OPI-MS operation. Taking full advantage of the chromatography-free and suppression-reducing features of this prototype instrument, we successfully demonstrated the ultrahigh-throughput bioanalysis of metformin, a small polar substrate commonly used in high-throughput in vitro transporter inhibition assays in the early ADME profiling space in drug discovery. The AEMS platform achieved a speed of 2.2 s/sample using only 10 nL of sample volume. Similar bioanalytical and biological results from actual assay samples were obtained by AEMS when compared to those obtained by the fastest LC-MS/MS method previously reported, along with a 15-fold speed advantage and ∼500-fold less sample consumption to enable future assay miniaturization. The general applicability of this novel approach to bioanalysis of several classes of polar analytes including ethambutol, isoniazid, ephedrine, and gemcitabine in biological matrices was further demonstrated.

Current status and future directions of high-throughput ADME screening in drug discovery.

During the last decade high-throughput in vitro absorption, distribution, metabolism and excretion (HT-ADME) screening has become an essential part of any drug discovery effort of synthetic molecules. The conduct of HT-ADME screening has been "industrialized" due to the extensive development of software and automation tools in cell culture, assay incubation, sample analysis and data analysis. The HT-ADME assay portfolio continues to expand in emerging areas such as drug-transporter interactions, early soft spot identification, and ADME screening of peptide drug candidates. Additionally, thanks to the very large and high-quality HT-ADME data sets available in many biopharma companies, in silico prediction of ADME properties using machine learning has also gained much momentum in recent years. In this review, we discuss the current state-of-the-art practices in HT-ADME screening including assay portfolio, assay automation, sample analysis, data processing, and prediction model building. In addition, we also offer perspectives in future development of this exciting field.

Development, optimization and implementation of a centralized metabolic soft spot assay.

AIM: High clearance is a commonly encountered issue in drug discovery. Here we present a centralized metabolic soft spot identification assay with adequate capacity and turnaround time to support the metabolic optimization needs of an entire discovery organization. METHODOLOGY: An integrated quan/qual approach utilizing both an orthogonal sample-pooling methodology and software-assisted structure elucidation was developed to enable the assay. Major metabolic soft spots in liver microsomes (rodent and human) were generated in a batch mode, along with kinetics of parent disappearance and metabolite formation, typically within 1 week of incubation. RESULTS & CONCLUSION: A centralized metabolic soft spot identification assay has been developed and has successfully impacted discovery project teams in mitigating instability and establishing potential structure-metabolism relationships.

Recent developments in software tools for high-throughput in vitro ADME support with high-resolution MS.

The last several years have seen the rapid adoption of the high-resolution MS (HRMS) for bioanalytical support of high throughput in vitro ADME profiling. Many capable software tools have been developed and refined to process quantitative HRMS bioanalysis data for ADME samples with excellent performance. Additionally, new software applications specifically designed for quan/qual soft spot identification workflows using HRMS have greatly enhanced the quality and efficiency of the structure elucidation process for high throughput metabolite ID in early in vitro ADME profiling. Finally, novel approaches in data acquisition and compression, as well as tools for transferring, archiving and retrieving HRMS data, are being continuously refined to tackle the issue of large data file size typical for HRMS analyses.

Sample reduction strategies in discovery bioanalysis.

It is a constant challenge to provide timely bioanalytical support for the evaluation of drug-like properties and PK/PD profiles for the ever-increasing numbers of new chemical entities in a cost-effective manner. While technological advancement in various aspects of LC-MS/MS analysis has significantly improved bioanalytical efficiency, a number of simple sample reduction strategies can be employed to reduce the number of samples requiring analysis, and as a result increase the bioanalytical productivity without deploying additional instruments. In this review, advantages and precautions of common sample reduction strategies, such as sample pooling and cassette dosing, are discussed. In addition, other approaches such as reducing calibration standards and eliminating over-the-curve sample reanalysis will also be discussed. Taken together, these approaches can significantly increase the capacity and throughput of discovery bioanalysis without adding instruments, and are viable means to enhance the overall productivity of the bioanalytical laboratory.

Recent development in software and automation tools for high-throughput discovery bioanalysis.

Bioanalysis with LC-MS/MS has been established as the method of choice for quantitative determination of drug candidates in biological matrices in drug discovery and development. The LC-MS/MS bioanalytical support for drug discovery, especially for early discovery, often requires high-throughput (HT) analysis of large numbers of samples (hundreds to thousands per day) generated from many structurally diverse compounds (tens to hundreds per day) with a very quick turnaround time, in order to provide important activity and liability data to move discovery projects forward. Another important consideration for discovery bioanalysis is its fit-for-purpose quality requirement depending on the particular experiments being conducted at this stage, and it is usually not as stringent as those required in bioanalysis supporting drug development. These aforementioned attributes of HT discovery bioanalysis made it an ideal candidate for using software and automation tools to eliminate manual steps, remove bottlenecks, improve efficiency and reduce turnaround time while maintaining adequate quality. In this article we will review various recent developments that facilitate automation of individual bioanalytical procedures, such as sample preparation, MS/MS method development, sample analysis and data review, as well as fully integrated software tools that manage the entire bioanalytical workflow in HT discovery bioanalysis. In addition, software tools supporting the emerging high-resolution accurate MS bioanalytical approach are also discussed.

Cassette incubation followed by bioanalysis using high-resolution MS for in vitro ADME screening assays.

BACKGROUND: High-resolution MS (HRMS) has recently received a considerable interest in quantitative bioanalysis using full-scan acquisition mode. The benefits include complete elimination of compound-specific MS method development, and simultaneous collection of mass spectral data on both targeted and non-targeted components. One additional advantage that has not been widely discussed is its suitability for simultaneous quantitation of, theoretically, an unlimited number of compounds, which is not possible with selected reaction monitoring (SRM) on a triple quadrupole mass spectrometer. MATERIALS & METHODS: We took advantage of this unique bioanalytical capability of HRMS and developed a novel in vitro ADME workflow of cassette incubation of as many as 32 compounds, followed by quantitative bioanalysis using full-scan acquisition on an Orbitrap HRMS. The workflow was evaluated for a serum protein-binding assay and a parallel artificial membrane permeability (PAMPA) assay. RESULTS: The bioanalytical assay displayed acceptable sensitivity, selectivity and linearity for all compounds in the cassettes, and the biological results obtained using this approach were similar to those from discrete incubation and analysis, demonstrating the feasibility of the workflow. Additional benefits of this platform include a saving of analysis time due to the reduced sample numbers from the cassette approach, as well as cost saving due to the reduction in the required assay reagents. CONCLUSION: Cassette incubation with bioanalysis using HRMS is a feasible approach for high-throughput in vitro ADME assays evaluated in this study.

Ultrafast mass spectrometry based bioanalytical method for digoxin supporting an in vitro P-glycoprotein (P-gp) inhibition screen.

The evaluation of interactions between drug candidates and transporters such as P-glycoprotein (P-gp) has gained considerable interest in drug discovery and development. Inhibition of P-gp can be assessed by performing bi-directional permeability studies with in vitro P-gp-expressing cellular model systems such as Caco-2 (human colon carcinoma) cells, using digoxin as a substrate probe. Existing methodologies include either assaying (3)H-digoxin with liquid scintillation counting (LSC) detection or assaying non-labeled digoxin with liquid chromatography-tandem mass spectrometric (LC-MS/MS) analysis at a speed of several minutes per sample. However, it is not feasible to achieve a throughput high enough using these approaches to sustain an early liability screen that generates more than a thousand samples on a daily basis. To address this challenge, we developed an ultrafast (9 s per sample) bioanalytical method for digoxin analysis using RapidFire™, an on-line solid-phase extraction (SPE) system, with MS/MS detection. A stable isotope labeled analog, d3-digoxin, was used as internal standard to minimize potential ionization matrix effect during the RF-MS/MS analysis. The RF-MS/MS method was more than 16 times faster than the LC-MS/MS method but demonstrated similar sensitivity, selectivity, reproducibility, linearity and robustness. P-gp inhibition results of multiple validation compounds obtained with this RF-MS/MS method were in agreement with those generated by both the LC-MS/MS method and the (3)H-radiolabel assay. This method has been successfully deployed to assess P-gp inhibition potential as an important early liability screen for drug-transporter interaction.

An integrated bioanalytical platform for supporting high-throughput serum protein binding screening.

Quantification of small molecules using liquid chromatography/tandem mass spectrometry (LC/MS/MS) on a triple quadrupole mass spectrometer has become a common practice in bioanalytical support of in vitro adsorption, distribution, metabolism and excretion (ADME) screening. The bioanalysis process involves primarily three indispensable steps: MS/MS optimization for a large number of new chemical compounds undergoing various screening assays in early drug discovery, high-throughput sample analysis with LC/MS/MS for those chemically diverse compounds using the optimized MS/MS conditions, and post-acquisition data review and reporting. To improve overall efficiency of ADME bioanalysis, an integrated system was proposed featuring an automated and unattended MS/MS optimization, a staggered parallel LC/MS/MS for high-throughput sample analysis, and a sophisticated software tool for LC/MS/MS raw data review as well as biological data calculation and reporting. The integrated platform has been used in bioanalytical support of a serum protein binding screening assay with high speed, high capacity, and good robustness. In this new platform, a unique sample dilution scheme was also introduced. With this dilution design, the total number of analytical samples was reduced; therefore, the total operation time was reduced and the overall throughput was further improved. The performance of the protein binding screening assay was monitored with two controls representing high and low binding properties and an acceptable inter-assay consistency was achieved. This platform has been successfully used for the determination of serum protein binding in multiple species for more than 4000 compounds.

Recent development in high-throughput bioanalytical support for in vitro ADMET profiling.

IMPORTANCE OF THE FIELD: High-throughput in vitro ADMET (absorption, distribution, metabolism, excretion and toxicity) profiling has become an important common practice in the pharmaceutical industry to assess compound liability early in the drug discovery process. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is the bioanalytical method of choice for ADMET profiling assays that require compound-specific detection. However, the in vitro ADMET profiling environment, with its unique bioanalytical requirements of analyzing many samples generated from many discrete compounds in a high-throughput fashion, poses significant challenges for the traditional LC-MS/MS technology and process workflow, which were originally designed and optimized for single-compound bioanalysis. AREAS COVERED IN THIS REVIEW: This article reviews advances made during the last several years in both conventional high-throughput LC-MS/MS approaches and a number of promising novel MS-based technologies specifically developed to address the unique challenges in an ADMET environment. The advantages and limitations of each technology are also discussed. In addition, software solutions to enable and integrate these hardware improvements into the high-throughput ADMET workflow are reviewed. WHAT THE READER WILL GAIN: The reader will gain an updated knowledge of the state-of-the-art technologies and practices, as well as promising novel MS-based methodologies in the field of ADMET bioanalysis. TAKE HOME MESSAGE: Recent advances such as automated MS/MS optimization, high-speed and multiplexed LC separation, and integrated software support have significantly increased the speed and quality of ADMET bioanalysis using LC-MS/MS. Emerging novel technologies in front-end sample introduction, ionization and mass analysis are expected to further push the current throughput limit and potentially transform the existing bioanalytical paradigm in the future.

A high-throughput bioanalytical platform using automated infusion for tandem mass spectrometric method optimization and its application in a metabolic stability screen.

Liquid chromatography/tandem mass spectrometry (LC/MS/MS) is the bioanalytical method of choice to support plate-based, in vitro early ADME (Absorption, Distribution, Metabolism and Excretion) screens such as metabolic stability (Metstab) assessment. MS/MS method optimization has historically been the bottleneck in this environment, where samples from thousands of discrete compounds are analyzed on a monthly basis, mainly due to the lack of a high-quality commercially available platform to handle the necessary MS/MS method optimization steps for sample analysis by selected reaction monitoring (SRM) on triple quadrupole mass spectrometers. To address this challenge, we recently developed a highly automated bioanalytical platform by successfully integrating QuickQuan 2.0, a unique high-throughput solution featuring MS/MS method optimization by automated infusion, with a customized in-house software tool in support of a Metstab screen. In this platform, a dual-column setup running parallel chromatography was also implemented to reduce the bioanalytical cycle time for LC/MS/MS sample analysis. A set of 45 validation compounds was used to demonstrate the speed, quality and reproducibility of MS/MS method optimization, sample analysis, and data processing using this automated platform. Metstab results for the validation compounds in microsomes from multiple species (human, rat, mouse) showed good consistency within each batch, and also between batches conducted on different days. We have achieved and maintained a monthly throughput of 1300 compound assays representing 500 discrete compounds per instrument per month on this platform, and it has been used to generate metabolic stability data for more than 25 000 compounds to date with an overall success rate of more than 95%.

Complete profiling and characterization of in vitro nefazodone metabolites using two different tandem mass spectrometric platforms.

This paper describes the complete profiling and characterization of in vitro metabolites of the antidepressant agent nefazodone (NEF) generated by human liver microsome (HLM). Two new metabolic pathways (biotransformation) for NEF have been discovered by the characterization of three new metabolites, including two new metabolites (M24, M25) formed due to the N-dealkylation reaction that occurred between the triazolone and propyl units, and one new metabolite (M26) formed due to the O-dearylation reaction that occurred on the phenoxyethyl unit. These metabolites were initially detected by a 4000 Q-Trap instrument and then confirmed by exact mass measurement using an LTQ-Orbitrap. Both instruments proved to be capable of providing complete in vitro metabolite information in a single liquid chromatography/tandem mass spectrometry (LC/MS/MS) analysis, although each had its advantages and disadvantages. One noticeable disadvantage of the 4000 Q-Trap was the reduced quality of isotopic pattern in the enhanced mass scan (EMS) spectrum when it was used as survey scan to trigger multiple dependent product ion scans. The problem was especially exacerbated for minor metabolites with low signal intensity. On the other hand, the LTQ-Orbitrap maintained excellent isotopic pattern when used as a full scan survey scan. Twenty-six metabolites were detected and identified. The formation of these new metabolites was also confirmed by analyzing duplicate incubations at different time points.

'N-in-one' strategy for metabolite identification using a liquid chromatography/hybrid triple quadrupole linear ion trap instrument using multiple dependent product ion scans triggered with full mass scan.

This paper describes a new strategy that utilizes the fast trap mode scan of the hybrid triple quadrupole linear ion trap (QqQ(LIT)) for the identification of drug metabolites. The strategy uses information-dependent acquisition (IDA) where the enhanced mass scan (EMS), the trap mode full scan, was used as the survey scan to trigger multiple dependent enhanced product ion scans (EPI), the trap mode product ion scans. The single data file collected with this approach not only includes full scan data (the survey), but also product ion spectra rich in structural information. By extracting characteristic product ions from the dependent EPI chromatograms, we can provide nearly complete information for in vitro metabolites that otherwise would have to be obtained by multiple precursor ion scan (prec) and constant neutral loss (NL) analysis. This approach effectively overcomes the disadvantages of traditional prec and NL scans, namely the slow quadrupole scan speed, and possible mass shift. Using nefazodone (NEF) as the model compound, we demonstrated the effectiveness of this strategy by identifying 22 phase I metabolites in a single liquid chromatography/tandem mass spectrometry (LC/MS/MS) run. In addition to the metabolites reported previously in the literature, seven new metabolites were identified and their chemical structures are proposed. The oxidative dechlorination biotransformation was also discovered which was not reported in previous literature for NEF. The strategy was further evaluated and worked well for the fast discovery setting when a ballistic gradient elution was used, as well as for a simulated in vivo setting when the incubated sample (phase I metabolites) was spiked to control human plasma extract and control human urine.

Advantages of using tetrahydrofuran-water as mobile phases in the quantitation of cyclosporin A in monkey and rat plasma by liquid chromatography-tandem mass spectrometry.

A new analytical method is described here for the quantitation of anti-inflammatory drug cyclosporin A (CyA) in monkey and rat plasma. The method used tetrahydrofuran (THF)-water mobile phases to elute the analyte and internal standard, cyclosporin C (CyC). The gradient mobile phase program successfully eluted CyA into a sharp peak and therefore improved resolution between the analyte and possible interfering materials compared with previously reported analytical approaches, where CyA was eluted as a broad peak due to the rapid conversion between different conformers. The sharp peak resulted from this method facilitated the quantitative calculation as multiple smoothing and large number of bunching factors were not necessary. The chromatography in the new method was performed at 30 degrees C instead of 65-70 degrees C as reported previously. Other advantages of the method included simple and fast sample extraction-protein precipitation, direct injection of the extraction supernatant to column for analysis, and elimination of evaporation and reconstitution steps, which were needed in solid phase extraction or liquid-liquid extraction reported before. This method is amenable to high-throughput analysis with a total chromatographic run time of 3 min. This approach has been verified as sensitive, linear (0.977-4000 ng/mL), accurate and precise for the quantitation of CyA in monkey and rat plasma. However, compared with the usage of conventional mobile phases, the only drawback of this approach was the reduced detection response from the mass spectrometer that was possibly caused by poor desolvation in the ionization source. This is the first report to demonstrate the advantages of using THF-water mobile phases to elute CyA in liquid chromatography.

A novel approach to perform metabolite screening during the quantitative LC-MS/MS analyses of in vitro metabolic stability samples using a hybrid triple-quadrupole linear ion trap mass spectrometer.

In vitro metabolic stability experiments using microsomes or other liver preparations are important components in the discovery and lead-optimization stages of compound selection in the pharmaceutical industry. Currently, liquid chromatography-tandem mass spectrometric (LC-MS/MS) support of in vitro metabolic stability studies primarily involves the monitoring of disappearance of parent compounds, using selected reaction monitoring (SRM) on triple-quadrupole instruments. If moderate to high turnover is observed, separate metabolite identification experiments are then conducted to characterize the biotransformation products. In this paper, we present a novel method to simultaneously perform metabolite screening in addition to the quantitative stability measurements, both within the same chromatographic run. This is accomplished by combining SRM and SRM-triggered, information-dependent acquisition (IDA) of MS/MS spectra on a hybrid triple-quadrupole linear ion trap (QqQLIT) mass spectrometer. Microsomal stability experiments using model compounds, bufuralol, propranolol, imipramine, midazolam, verapamil and diclofenac, were used to demonstrate the applicability of our approach. This SRM + SRM-IDA approach generated metabolic stability results similar to those obtained by conventional SRM-only approach. In addition, MS/MS spectra from potential metabolites were obtained with the enhanced product ion (EPI) scan function of LIT during the same injection. These spectra were correlated to the spectra of parent compounds to confirm the postulated structures. The time-concentration profiles of identified metabolites were also estimated from the acquired data. This approach has been successfully used to support discovery programs.