Poster Presentations for Poster Session A : Wed 9:15 - 10:30

Smart Solutions for High Analyte Concentrations and Carryover in LC-MS Monitoring Andrew Clark (Presenter) Thermo Fisher Scientific

Poster #1a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
In clinical toxicology laboratories, routine quantification of drugs of abuse (DoA) in biological samples using validated liquid chromatography–mass spectrometry (LC-MS) methods often encounters samples with analyte concentrations exceeding the upper limit of quantification. Such high concentrations can compromise the accuracy of quantification and increase the risk of carryover to subsequent injections. To mitigate these issues, it is essential for LC-MS data acquisition software to incorporate features that allow users to define thresholds, such as upper concentration limits, and evaluate each injection in real-time. This enables automated responses to prevent carryover and maintain the integrity and reliability of the dataset.

METHOD:
A mixture of cocaethylene, oxazepam, methadone, methamphetamine, and PCP, along with their corresponding internal standards, was spiked into 20-fold diluted blank human urine at concentrations of 0.5, 5, 100, 500, and 1000 ng/mL. A 1 µL mixture was directly injected onto a Thermo Scientific™ Vanquish™ Flex UHPLC system coupled to a Thermo Scientific™ Orbitrap Exploris™ 120 mass spectrometer equipped with the OptaMax NG HESI source. Analyte separation was achieved on a Thermo Scientific™ Accucore™ Biphenyl column at 0.5 mL/min flowrate with 0.1% formic acid in water as the mobile phase A and 0.1% formic acid in methanol as the mobile phase B. The calibration curves were generated using the samples with concentration 0.5, 5, and 100 ng/mL, and the sample with the concentration 500 and 1000 ng/mL was used to trigger the pre-defined Upper Limit of Quantification (ULOQ) of 100 ng/mL or Upper Peak Area (UPA) of 2e7 in System Suitability Test / Intelligent Run Control (SST/IRC) in Chromeleon. Three Custom Injection Variables in SST/IRC were setup and tested independently: 1) “AutoDilutionConc” using the injection volume determined from ULOQ, 2) "AutoDilutionArea" using the injection volume based on UPA, and 3) "InjectWash" to inject solvent blank with a column wash method.

RESULTS:
SST/IRC provide intelligent features that monitor various sample types across different test cases. These include the presence of analytes in blank matrix workup, whether analyte concentrations in quality control samples meet expected values, and most importantly, whether a sample with unknown analyte concentration exceeds a certain value that may cause unreliable quantification results or potential carryover. In SST/IRC, users can define peak area or calculated concentration threshold for each or selected compounds and choose appropriate intervention actions from options such as "Reinject," "AutoDilute,” and user-defined "Custom Injection Variables." We tested and triggered 1/5 or 1/10 injection volumes of the Unknown sample, 500 and 1000 ng/mL respectively, using the 'AutoDilution' feature, based on the fold difference in analyte concentrations or peak areas relative to ULOQ or UPA. A solvent blank was also injected using a different instrument method to efficiently wash the flow path.

CONCLUSION:
This study successfully demonstrated the setup and utilization of intelligent features in Chromeleon software to monitor analyte quantification results in real-time. The software allows users to select appropriate actions to adapt to their specific needs, ensuring reliable quantification and preventing carryover in clinical toxicology applications.

Evaluating the Recovery of 68Zn in Radiopharmaceutical Production Solutions Using an Agilent 8900 ICP-MS/MS Anna Bitzer (Presenter) Mayo Clinic

Poster #3a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Radiopharmaceuticals represent a promising and quickly evolving branch of precision medicine, in which drugs containing radioactive elements are delivered to targeted cells, including those that form cancerous tumors. The materials used to manufacture these pharmaceuticals however must meet certain standards of chemical purity and isotopic enrichment in order to be used.

OBJECTIVES:
The goal of this study was to create an accurate and precise method to quantitate the total concentration and isotopic zinc composition of solutions recovered from the production of cyclotron-created radiopharmaceuticals to determine the amount of 68Zn present in these solutions.

METHODS:
An Agilent 8900 ICP-MS/MS was used to analyze the five stable isotopes of zinc (64Zn, 66Zn, 67Zn, 68Zn, 70Zn). All isotopes were measured on-mass, using oxygen as a reaction gas, and 185Re as an internal standard. Each sample acquisition included 5 seconds of stabilization time, a total of 7 replicates, 100 sweeps per replicate, and an integration time of 0.1 seconds per isotope. Concentrations of all isotopes were summed to determine the total zinc concentration. A calibration of the analytical measurement range of 50-5000 mcg/L total zinc was performed daily. National Institute of Standards and Technology (NIST) 1643f aqueous samples were measured to monitor the accuracy of the total zinc. An isotopically enriched 68Zn solution produced and certified by Oak Ridge National Laboratory was used to monitor the accuracy of the zinc isotope ratios. All radiopharmaceutical samples were analyzed in triplicate, bracketed by NIST aqueous samples and Oak Ridge isotopically enriched solution to monitor accuracy throughout the run.

RESULTS:
Total Zn results were found to have a slight high bias, with an average value of NIST 1643f at 98.2 ng/mL over n=9 runs compared to the expected value of 74.4 ng/mL, which was an acceptable level of accuracy for the purposes of this assay. Total Zn inter-assay precision was monitored over n= 9 runs also deemed acceptable, with a %CV of 12.3% for NIST 1643f (98.2 ng/mL) and 7.3% for the Oak Ridge enriched solution (904.3 ng/mL). The total zinc intra-triplicate precision was acceptable as well over n=9 runs, with a %CV of 2.5% for NIST 1643f and 0.6% for the Oak Ridge enriched solution. The 68Zn intra-triplicate precision specifically for the Oak Ridge enriched solution had a %CV of 0.4%. Finally, the enrichment of 68Zn in the Oak Ridge enriched solution was monitored over n=9 runs and found to have an average enriched abundance of 94.4% (natural abundance = 18.8%), with inter-assay precision of 19.5% and an average intra-triplicate precision of 0.3%.

CONCLUSION:
A precise, relatively accurate, and quick method was developed to quantitate both the total concentration and the isotopic composition of solutions recovered from the production of cyclotron-created Zn radiopharmaceuticals. By utilizing ICP-MS already present in the clinical laboratory, isotopic methods can be created to help in the research, development, and implementation of radiopharmaceutical protocols.

The Analysis of Tricyclic Antidepressant Drugs in Plasma for Clinical Research Padhraic Rossiter (Presenter) Waters Corporation

Poster #4a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
The accurate measurement of tricyclic antidepressants (TCAs) in plasma for clinical research, is vital in order to undertake pharmacokinetic studies. Waters has developed a clinical research method using LC-MS/MS which allows for efficient measurement of 15 TCAs: amitriptyline, clomipramine, clozapine desipramine, doxepin, imipramine, maprotiline, norclomipramine, norclozapine, nordoxepin, normaprotiline, nortrimipramine, nortriptyline, protriptyline and trimipramine in human plasma.

METHODS:
Calibrators and quality control samples were prepared using pooled human plasma. Samples (50 µL) were extracted with internal standard (150 µL) in acetonitrile, vortex mixed, centrifuged (2 minutes) and the supernatant was diluted in water and mixed. Using an ACQUITY™ UPLC™ I-Class FTN System, samples were injected onto a XSelect™ Premier HSS T3 Column using a water/methanol/formic acid/ammonium formate gradient elution profile, with detection using a Xevo™ TQ-S micro Mass Spectrometer.

RESULTS:
Chromatographic separation was achieved for isobaric compounds nortriptyline and protriptyline (precursor m/z 264.1) and the interfering qualifier transitions of imipramine and nortrimipramine (m/z 281.2>86.1). There was no significant carryover observed from a high concentration plasma sample into a subsequent blank injection. A 1:5 dilution of a high concentration sample was performed that gave a mean bias of less than 10 % for all analytes. Analytical sensitivity results demonstrated precise quantification (<6.9 % bias and <10 % CV) at concentrations equal to or lower than the lowest calibrator. Total precision and repeatability were assessed (3 pools, 5 replicates, 5 days; n=25) and determined to be ≤8.0 % CV. Matrix effects studies evaluated at low and high concentration levels, demonstrated compensation by the internal standard for any signal enhancement or suppression observed. The method also demonstrated excellent linearity for all analytes, with calibration lines exhibiting coefficient of determination (r2) of > 0.995.

CONCLUSION:
The analysis of 15 tricyclic antidepressants with different polarities was achieved by a selective and robust UHPLC-MS/MS clinical research method and used Waters XSelect Premier HSS Technology to separate isobaric compounds. The method required only 50µL of sample with an analysis time less than 4.5 minutes.

For Research Use Only. Not for Use in Diagnostic Procedures.

ACQUITY, UPLC, Xevo, XSelect are trademarks of Waters Technologies Corporation.

Quantitative LC-MS/MS Assay for Ceramides in Plasma and Serum: Development, Validation, and Performance Evaluation Amol Bajaj (Presenter) ARUP

>> POSTER (PDF)

Poster #5a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

BACKGROUND:
Ceramides have emerged as novel diagnostic and prognostic biomarkers for cardiovascular disease (CVD), providing additional risk prediction beyond traditional lipid measures. Several studies have highlighted their potential utility as biomarkers for insulin resistance, type 2 diabetes mellitus, severity of coronary artery disease, atherosclerosis, dyslipidemia, malignancies, and neuroinflammation. We developed a quantitative LC-MS/MS method to measure Cer 16:0, Cer 18:0, Cer 24:0, and Cer 24:1 in plasma and serum and evaluated the method’s performance.

METHODS:
Aliquots of human plasma or serum samples (40 µL) were mixed with stable isotope-labeled internal standard (IS, deuterium labeled analogs of each ceramide, 40 µL), water (100 µL) and methyl tert-butyl ether (MTBE, 400 µL). The tubes were vortexed for 20 min, and then centrifuged for 5 min at 16,000 g. The supernatants were transferred to autosampler vials, and the samples were analyzed using a 6475 LC-MS/MS, equipped with a 1290 series HPLC (Agilent). Chromatographic separation was performed using a CSH C18 column (1.7 µm, 5 cm, 100 Å, 2.1 mm ID, Waters). Quantification was performed using a six-point calibration curve (0.02–2 µmol/L for Cer 16:0 and Cer 18:0; 0.2–20 µmol/L for Cer 24:0 and Cer 24:1), monitoring two mass transitions per analyte and IS. The injection volume was 5 µL, and data acquisition was performed in positive ion mode. Evaluation of the method’s performance included assessment of precision, sensitivity, linearity, accuracy, specificity, matrix effects, dilution integrity, carryover, robustness, and correlation with validated methods performed by other laboratories. Blood collection tube types, stability, and the adsorptive losses of ceramides during the sample preparation were also evaluated. All studies with samples from human subjects were approved by the Institutional Review Board of the University of Utah (Salt Lake City, UT).

RESULTS:
We developed and evaluated analytical performance of a method for quantifying four ceramides in plasma and serum samples by LC-MS/MS. The issues related to poor retention of Cer 16:0 and Cer 18:0, and adsorptive losses of ceramides were addressed with sample preparation methods and chromatographic separation. A CSH C18 column (Waters) provided the best HPLC performance. In a sample preparation method involving solvent extraction, evaporation, and reconstitution, a variation in the slopes of the calibration curves was observed using calibrators prepared in a Mass Spec Gold Serum (Golden West Diagnostics) matrix and those prepared in a pool of patient samples. However, when the evaporation and the reconstitution steps were omitted, and the extracts were directly injected for instrumental analysis, the variation in the slopes of the regression lines were comparable, suggesting adsorptive losses of the ceramide during evaporation and reconstitution steps of sample preparation.

The assay demonstrated a total imprecision (CV) of <2-8.0% across the analytical measurement range for all the analytes included in the assay. In experiments on the evaluation of the methods’ accuracy, the observed concentrations agreed with the expected concentrations within 10% (imprecision among the replicates <5%). Slopes of linear regression lines (R2) for the agreement between the expected and the observed concentrations were 0.99 (0.996), 0.99 (0.995), 0.99 (0.995) and 0.99 (0.995) for Cer 16:0, Cer 18:0, Cer 24:0, and Cer 24:1, respectively. The lower limit of detection (LOD) was 0.01 µmol/L for Cer 16:0 and Cer 18:0, and 0.1 µmol/L for Cer 24:0 and Cer 24:1. The signal-to-noise ratio at the LOQ for the transitions of all the analytes was ≥10.
Over 600 neat patient samples (serum and plasma) have been analyzed by the assay; no peaks in the vicinity of the peaks of the analytes and the ISs were observed in the mass transitions used in the assay. Overestimation of the ceramide concentrations was observed in hemolyzed or lipemic samples, containing >300 mg/dL or >225 mg/dL of hemoglobin or intralipid, respectively. No carryover was observed after injection of samples containing 4 µmol/L for Cer 16:0 and Cer 18:0, and 40 µmol/L for Cer 24:0 and Cer 24:1. The method was compared to a validated LC-MS/MS method of another laboratory using NIST material (Metabolites in Frozen Human Plasma, SRM 1950) and patient plasma or serum pools (n=6). Median (range) for percent agreement between the developed method and the comparative method were: 132 (117-143)%, 152 (149-156)%, 101 (97-103)% and 93 (82-99)%; the R2 values for the linear regression curves were 0.978, 0.999, 0.994 and 0.987, for Cer 16:0, Cer 18:0, Cer 24:0 and Cer 24:1, respectively. Ceramide concentrations from serum (red top), EDTA plasma, and lithium heparin plasma were comparable, showing agreement within 15%.

CONCLUSION:
An LC-MS/MS method was developed for quantifying Cer 16:0, Cer 18:0, Cer 24:0, and Cer 24:1 in plasma and serum samples. The method demonstrated reliable analytical performance with adequate precision, accuracy, and robustness. We observed reasonably good inter-laboratory agreement in the measured ceramide concentrations.

Assessment of Serum 3-Epi-25-Hydroxyvitamin D3, Total 25-Hydroxyvitamin D3, and 25-Hydroxyvitamin D2 in the Korean Population with UPLC-MS/MS Sung-Eun Cho (Presenter) GCLabs ESAC

Poster #6a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
The liquid chromatography-tandem mass spectrometry method (LC-MS/MS) identified an epimeric form of 25-hydroxyvitamin D3 (3-epi-25-OH-D3) concentrations. The determination of 3-epi-25-OH-D3 has been nationally surveyed in the USA’s National Health and Nutrition Examination Survey (NHANES) since 2007 using LC-MS/MS method. In Korea, the 3-epi-25-OH-D3 was included for the first time in the Korea National Health and Nutrition Examination Survey (KNHANES) IX (from 2022 to 2024). We conducted this study and measured the 3-epi-25-OH-D3 concentrations, and now we evaluated the results of 3-epi-25-OH-D3 in KNHANES IX to find out its clinical status in Korea.

METHODS:
The UPLC-MS/MS was used for the measurement of 3-epi-25-OH-D3, 25-hydroxyvitamin D3 (25-OH-D3), and 25-hydroxyvitamin D2 (25-OH-D2). We used Nexera-X2-LC-30AD UPLC (Shimadzu, Japan) system equipped with a Kinetex XB C18 column (2.1X150 mm, 2.6 um; Phenomenex, USA) on a Triple Quad 4500MD (Sciex, USA) MS/MS system. This method had a longer run time of 16 minutes to separate 3-epi-25-OH-D3. We evaluated the above vitamin Ds based on gender and age group and then compared them with other markers.

RESULTS:
The concentration ranges of 3-epi-25-OH-D3, 25-OH-D3 and 25-OH-D2 in 11,744 Korean participants were 0.50 (lower limit of quantification (LLOQ))-15.66, 2.32-116.78, 0.30 (LLOQ)-22.13 ng/mL, and mean±SD concentrations were 1.15±0.85, 22.60±10.60, and 0.44±0.62 ng/mL, respectively. The concentration of 3-epi-25-OH-D3 was not significantly different between males and females. The blood concentrations of 25-OH-D metabolites differed across age groups in both males and females (P < 0.0001). The concentrations of 25-OH-D2, 3-epi-25-OH-D3, 25-OH-D3, and total 25-OH-D (which is the sum of 25-OH-D2, 3-epi-25-OH-D3, and 25-OH-D3) were highest in the 60s, followed by higher concentrations in the age group of 70s and above. In those aged group under 60, the concentrations tended to increase according to the age increase. There was a highly significant correlation between 3-epi-25-OH-D3 and total 25-OH-D3 (sum of 3-epi-25-OH-D3 and 25-OH-D3) (Pearsons’s correlation coefficient r=0.74, R2=0.5404, slope=0.0587, P<0.0001). The minimum and maximum % 3-epi-25-OH-D3 were 1.35 and 31.47 %, respectively. Although the mean 3-epi-25-OH-D3 concentration increased linearly by quartile of serum total 25-OH-D3 concentration, the % 3-epi-25-OH-D3 was highest in the Q1 quartile of total 25-OH-D3 concentration (P<0.001). The inclusion of 3-epi-25-OH-D3 concentration in the estimate of serum total 25-OH-D reduced the prevalence of serum 25-OH-D deficiency, insufficiency, and suboptimal level by 1.3, 3.2, and 3.8 percentage points, respectively. The prevalence of serum total 25-OH-D > 70 ng/mL increased from 0.2 to 0.3 % when the 3-epi-25-OH-D3 concentration was included in the total estimate. Every marker except hsCRP had significant correlations with % 3-epi-25-OH-D3. Among them, PLT, BUN, Total cholesterol, HDL cholesterol, and LDL cholesterol showed a negative correlation with % 3-epi-25-OH-D3.

CONCLUSION:
This is the first large-scale study of 3-epi-25-OH-D3 serum concentrations across all age groups in the Korean population. Accurate measurement of 3-epi-25-OH-D3 concentration is necessary for the reliable results of vitamin D metabolites in KNHANES of Korea.

KEYWORDS: Tandem mass spectrometry, 3-epi-25-OH-D3, KNHANES

Distribution of Serum and Urinary 24,25(OH)₂D₃ and VMR and Their Association with PTH in an Adult Health Checkup Cohort Nau Ishimine (Presenter) Shinshu University Hospital

Poster #7a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

BACKGROUND:
Serum 25-hydroxyvitamin D (25[OH]D) is the standard marker of vitamin D status; however, it may not fully reflect functional vitamin D sufficiency. In particular, serum 25(OH)D shows a weak inverse correlation with parathyroid hormone (PTH) levels, suggesting that additional vitamin D metabolites are necessary to better reflect their biological effects. Two markers, 24,25-dihydroxyvitamin D₃ (24,25[OH]₂D₃) and the ratio of 24,25(OH)₂D₃ to 25(OH)D (vitamin D metabolite ratio, VMR), have emerged as potential improved indicators of vitamin D status. Non-invasive methods, such as urinary metabolite profiling, offer clinical advantages; however, they have not been thoroughly investigated. Herein, we aimed to investigate vitamin D metabolite distributions in both serum and urine in a large adult cohort and to examine their correlations with PTH to evaluate the utility of urine-based vitamin D markers compared with traditional serum measures.

METHODS:
We included 506 adults undergoing annual health checkups. Paired blood and spot urine samples were collected. Using liquid chromatography-tandem mass spectrometry (LC-MS/MS), concentrations of 25(OH)D₃, 25(OH)D2, and 24,25(OH)₂D₃ were measured in both serum and urine. Urine samples were pre-treated with β-glucuronidase to hydrolyze glucuronide conjugates. All samples were derivatized with DAP-PA (JeoQuantTM VD kit, JEOL, Japan) to enhance electrospray ionization. The derivatized metabolites were quantified by LC-MS/MS using isotope-labeled internal standards. Serum PTH was measured by immunoassays. Serum vitamin D status was classified as deficient (< 20 ng/mL, <50 nmol/L), insufficient (20–30 ng/mL, 50-75 nmol/L), or sufficient (> 30 ng/mL, > 75 nmol/L), based on thresholds widely used in clinical and epidemiological studies, including those of the Endocrine Society and Japanese guidelines. Furthermore, cut-offs were based on those commonly used in clinical and epidemiological studies. VMR was calculated using the following formula: VMR = [24,25(OH)₂D₃/25(OH)D] × 100. Correlations between vitamin D metabolites and VMR with PTH were analyzed.

RESULTS:
A total of 506 individuals (156 males, 350 females; median age 41 years) were included. Vitamin D deficiencies were identified in 39.5% of participants (serum 25[OH]D < 20 ng/mL), and 42.1% were insufficient (20–30 ng/mL). Thus, 81.6% of participants had suboptimal vitamin D status. The median serum 25(OH)D was 22.33 ng/mL and the median serum VMR was approximately 4.5%. Notably, 33.6% of participants had serum VMR < 4%, a cutoff proposed for low vitamin D metabolite profiles.
Differences by sex were found in serum 25(OH)D₃, 24,25(OH)₂D₃, and VMR. Urinary 24,25(OH)₂D₃ was the predominant metabolite, often exceeding urinary 25(OH)D₃ in concentration. Correlation analysis showed that urinary 25(OH)D and 24,25(OH)₂D₃ levels were significantly correlated with their corresponding serum metabolites (ρ > 0.6, p < 0.001). Urinary 24,25(OH)₂D₃ showed significantly stronger correlation with serum 25(OH)D than other urinary metabolites (p < 0.001, Steiger’s test). Both serum and urinary vitamin D metabolites showed significant inverse correlations with serum PTH. Notably, urinary 24,25(OH)₂D₃ and VMR demonstrated stronger correlations with PTH than their corresponding serum metabolites. Quartile analysis further confirmed that lower urinary metabolite levels were associated with higher PTH concentrations (p < 0.001).

CONCLUSIONS:
This study found a high prevalence of vitamin D insufficiency in this adult cohort. Moreover, low 24,25(OH)₂D₃ levels and low VMR were characteristic findings of vitamin D deficiency. We demonstrated that urinary vitamin D metabolites, particularly 24,25(OH)₂D₃ and urinary VMR, are more strongly correlated with PTH than serum 25(OH)D, suggesting that these non-invasive markers better reflect functional vitamin D status. These findings support the use of urinary vitamin D metabolite profiling as a practical and biologically meaningful tool. Its potential application in over-the-counter tests, such as self-health checkups, warrants further exploration.

24/7 Fully Automated Therapeutic Drug Analysis for Research Projects by LC/MS/MS Kohei Yoshikawa (Presenter) Shimadzu Corporation

Poster #8a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Simplifications in LCMS instrumentation have made MS a viable option for clinical research (specificity, accuracy, and reduced reagent costs). The ability to support various analysis methods on a single system is a key feature. Automation is an essential function in aiming for a better quality of results and better comfort for the users. However, lately, the automation of biological sample extraction, directly coupled to LCMS, has proven to be a challenge in the field.

METHODS:
A collaboration between the University Medical Center Göttingen (UMG) and Shimadzu Corporation was built to jointly develop and validate multiple analytical methods for therapeutic drugs, using a fully automated platform. The purpose is the development and the validation of a unified methods set for LCMS, for 24/7 therapeutic drug analysis, with a single system configuration.

The evaluated analytical system was a fully automated platform, from Shimadzu Corporation, composed of CLAM-2040 automation module, coupled to Nexera(TM)X3 UHPLC and LCMS-8060NX(TM) triple quadrupole mass spectrometer. HL-7 interface standards were used for bidirectional communication between LIS (Dedalus, Germany) and the CLAM-LC/MS/MS. Target applications were Antibiotics, Direct Oral Anticoagulants, Antiepileptics, Neuroleptics, Antidepressant drugs (SSRI), Tricyclic Antidepressants, and Benzodiazepines as well as other compounds such as digoxin, methotrexate, and analytes related to Drugs of Abuse, totaling nearly 180 compounds. The methods are divided by drug groups, but the mobile phase and columns are unified. The fitness for purpose for 24/7 use was evaluated through the verification of random access functionality and calibration stability.

RESULTS:
All validations results were within the acceptance criteria. Individual methods validations include isobars resolution (above 1), calibration accuracy (85-115%), method repeatability (RSD below 15%), day to day intermediate precision (RSD below 15%), mobile phase stability (RT deviation below 2% after 2 weeks), LLOQ confirmation (S/N above 10 and RSD below 15%), absence of carryover confirmation (blank to LLOQ area ratio below 20%) and ring trial analysis (accuracy 85-115%). Repeated measurements for all methods in random alternation showed results within the acceptance criteria (accuracy: 85-115%, RSD: below 15%). Furthermore, calibration stability lasted for at least one week. These outcomes were made possible by standardizing the initial concentration of the mobile phase and the MS interface conditions.

CONCLUSION:
This strategy proved its fitness for purpose. The fast LCMS methods which can alternate smoothly, and the automated sample extraction enable robust therapeutic drug analysis with a high throughput and at low cost, without compromising the user comfort.

UPLC-MS/MS Method to Evaluate a Neurotransmitter Panel in Women Using Cannabis During Pregnancy Corentin Fracapane (Presenter) Université de Sherbrooke

>> POSTER (PDF)

Poster #9a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Cannabis use, especially during pregnancy, has increased during recent years. The lack of substantial clinical data on the prenatal effects of cannabis, coupled with its increased availability, and the tendency of pregnant individuals to self-medicate to relieve symptoms such as stress, nausea, and vomiting, contribute to this upward trend. Most studies on the consumption of cannabis during pregnancy concentrate on the social implications and the developmental outcomes of the child. Moreover, the intricate interplay among diverse neurotransmitter systems which govern multifaceted domains of human behavior, including emotion, learning, and memory renders them particularly susceptible to alterations. To our knowledge, there are no studies focusing on neurotransmitter levels in pregnant women consuming cannabis. There are thus unmet needs to evaluate the quantification of a neurotransmitter panel in biological fluids for these pregnant women. However, current methodological studies show limitations for neurotransmitter analysis. In fact, recent studies focus mainly on a singular neurotransmitter or a specific system like the GABAergic or cholinergic systems. Unfortunately, few multiplex analyses focusing on several neurotransmitters from various systems were reported. One of the challenges encountered was to find a chromatographic column that can effectively analyze all neurotransmitters of interest in a single analysis with a good separation for all molecules. Considering the lack of information evaluating the impact of cannabis use during pregnancy on the neurotransmitter systems, and the issues with the available methodologies, we have developed a robust tandem mass spectrometry multiplex method to analyse a neurotransmitter panel in pregnant women.

OBJECTIVES:
The objectives of this study are: 1) to develop and validate a mass spectrometry method for the measurement of a neurotransmitter panel comprising of norepinephrine (NE), gamma-aminobutyric acid (GABA), L-glutamate (Glu), serotonin (5-HT), epinephrine (E), dopamine (DA), and acetylcholine (ACh) as part of a multiplex analysis in plasma and urine specimens of pregnant cannabis users and controls (pregnant non-user); 2) to evaluate neurotransmitter levels during pregnancy in plasma and urine specimens of pregnant women exposed to cannabis compared to unexposed ones.

METHODS:
Samples will undergo processing via protein precipitation. The liquid chromatography method (column, gradient, mobile phases, flow rate, injection volume, etc.) and mass spectrometry parameters (gas voltages, flow rates, temperatures, collision energies and transitions used) have been developed and optimized according to each neurotransmitter. These neurotransmitters include NE, GABA, Glu, 5-HT, E, DA, Ach, and their respective internal deuterated standards. Following the optimization, a method was developed for the analysis of all neurotransmitters in a single multiplex assay. A full-comprehensive method validation is on going. Targeted analytes and their internal standards will be analyzed by an ultra performance liquid chromatography (UPLC) coupled to a triple quadrupole mass spectrometer (MS/MS) (Acquity I-Class Plus, Xevo TQ-XS, Waters Corp.) for its sensitivity and accuracy using a positive electrospray ionization mode. MRM transitions have been determined for the analytes and standards. Following the development and validation of the neurotransmitter method, specimens collected at 35 weeks of gestation from 40 participants comprised of 20 women exposed to cannabis and 20 control women (pregnant non-user) will be evaluated. Correlations will be established between neurotransmitter levels and clinical parameters for each participant. Subsequent statistical analyses will be conducted to elucidate the findings.

RESULTS:
For the final optimized conditions, chromatography was performed using a 2.5 μm Atlantis Premier BEH Z-HILIC column (100 × 2.1 mm i.d.) maintained at +30°C. A volume of 7.5 microliters was injected in the chromatographic system with a sample stack maintained at +6 ◦C. A gradient was optimized. The mobile phase A was a mixture of water/acetonitrile (90:10), 0.1% formic acid and 100 mM ammonium formate. The mobile phase B was water/acetonitrile (15:85), 0.1% formic acid and 20 mM ammonium formate. The best flow rate was 0.3 mL/min. The total run-time was 10.0 min. A good separation of all 7 neurotransmitters was obtained with a LOQ between 0.01 ng/mL and 2.5 ng/mL, and a linearity ranging from R2 = 0.9974 to 1.

CONCLUSION:
This project is crucial for the acquisition of preliminary findings for the study of the neurotransmitter physiology during pregnancy and its potential alterations subsequent to cannabis utilization. The mass spectrometry method developed and validated will be an asset for the assessment of neurotransmitters in other studies on neurodevelopmental disorders (ADHD, autism, schizophrenia, etc.). Future perspectives will involve the characterization of a neurotransmitter profile during infancy. This characterization will be conducted in correlations with the infant's neurological development and behavior.

Validation of a Clinical LC-MS/MS Assay for Thyroglobulin Measurement in Serum Magdalena Bolek (Presenter) NSW Health Pathology

Poster #10a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Serum thyroglobulin (Tg) determination is important in clinical management of patients with differentiated thyroid cancer (DTC), to monitor residual or recurrent disease. Tg is typically measured by automated immunoassays. However, thyroglobulin autoantibodies (TgAb), which are present in 20–30% of patients with DTC, can interfere with antibody-based analyses, giving rise to falsely low, or negative Tg measurements [1]. This interference varies between patients and Tg assays, and only loosely correlates with TgAb concentrations [1,2].

Mass spectrometry-based quantification of Tg [3-5] eliminates antibody interferences and has become the preferred method, despite technical challenges. This study describes the adaptation and validation of an LC-MS/MS method for Tg measurement in human serum.

METHODS:
The methodology was based on a previously published method [5], with some modifications. Calibration samples were prepared in Tg(-), TgAb(-) human serum spiked with human Tg (Tg II CalSet, Cobas, Roche). The method was compared to Elecsys Tg II assay (Cobas e601, Roche) using patient samples (n=70) and three levels of QCs prepared from pooled patient serum (0.3, 1.1, and 7 ng/mL). Analyte recovery was evaluated by spiking the internal standard before and after key extraction steps and compared to neat standard in the elution buffer.

RESULTS:
The LC-MS/MS assay achieved a lower limit of quantification of 0.05 ng/mL (CV <20%) and showed a good agreement with the immunoassay in TgAb(-) samples (n=41). The imprecision of low, mid, and high QC samples was 6.8, 7.3 and 5.4 % CV respectively (based on n=66 from at least 20 days). Linearity of the calibration curve between 0.05 and 12.5 ng/mL was excellent, with an average correlation co-efficient (r2) of 0.999, and average slope of 0.952 (intra-day %CV = 7.5). Improved recovery was achieved by decreasing denaturation time to 30 min and increasing the digestion time to 1 h.

CONCLUSION:
The adapted method has acceptable performance characteristics and suggests that accurate and sensitive mass-spectrometry-based quantification of serum thyroglobulin can be achieved in clinical laboratories with access to adequate LC-MS instrumentation and laboratory infrastructure. Future work will evaluate the usefulness of Tg-MS analyses for clinical decision making and the impact of automation on assay performance.

REFERENCES:
[1] C.A. Spencer, Clinical review: Clinical utility of thyroglobulin antibody (TgAb) measurements for patients with differentiated thyroid cancers (DTC), J. Clin. Endocrinol. Metab. 96 3615–3627 (2011).
[2] C. Evans, S. Tennant, P. Perros, Thyroglobulin in differentiated thyroid cancer, Clin. Chim. Acta 444:310-317 (2015).
[3] Clarke NJ, Zhang Y, Reitz RE. A novel mass spectrometry-based assay for the accurate measurement of thyroglobulin from patient samples containing antithyroglobulin autoantibodies. J Investig Med. 2012; 60:1157–1163.
[4] Netzel BC, Grebe SK, Algeciras-Schimnich A. Usefulness of a thyroglobulin liquid chromatography-tandem mass spectrometry assay for evaluation of suspected heterophile interference. Clin Chem (2014) 60:1016 –1018.
[5] Shi J, Phipps W, et al., A distributable LC-MS/MS method for the measurement of serum thyroglobulin. JMSACL, 2022, 26: 28-33.

The Use of Ambient Ionisation Mass Spectrometry Techniques to Address Fragmentation Across Colorectal Cancer Treatment in Healthcare Systems Lauren Ford (Presenter) Imperial College London

Poster #11a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Colorectal cancer (CRC) is the third most common cancer globally and a leading cause of cancer mortality. The patient pathway through the clinic is extremely fragmented, often resulting in delayed diagnosis and treatment. The initiation of cancer within the colon elicits changes to the glycolytic pathway and up regulation of de novo fatty acid synthesis. Thus, metabolomics analysis can be utilised to detect dysregulation of metabolites in tissues and surrounding biofluid and phenotype disease. Ambient Ionisation Mass Spectrometry can be used for high throughput feedback through each part of the patient journey through detection, diagnosis, surgery, post operative treatment and surveillance.

METHODS:
Faecal samples were collected from the colorectal cancer patients prior to surgery at Imperial College Healthcare Trust (ICHT) hospital sites (REC: 14/EE/0024). In total 10 CRC, 8 Adenoma and 15 healthy stool samples were collected. Surgical data was collected as a prospective, observational study from 8 patients undergoing colorectal or endoscopic resection. In-vivo analysis was performed by transporting the aerosol generated from electrocautery during the surgical/endoscopic procedure to a Xevo G2-XS QToF mass spectrometer (Waters), fitted with a REIMS source. Ex-vivo tissue analysis was carried out using the same conditions using during surgery (n=153). Remaining tissue was then sectioned to 10µm thickness and analysed using DESI-MSI and LD-REIMS-MSI (n=45). Multivariate statistical analysis was used to distinguish metabolites driving phenotypic changes.

RESULTS:
Faecal specimens were sampled using swabs coated in cationic polymer polyethyleneimine (PEI) and analysed using high-throughput DESI-MS. Each swab took 30 seconds to acquire data and analyse. Separation between the cancer and normal samples could be observed using principal component analysis (PCA) and support vector machine learning (SVM). Diagnostic accuracy for this test was 93% (93% sensitivity and 95% specificity).

During surgical procedures metabolomic signal was observed in 8 patients, spectra were collected from corresponding normal, tumour and polyp samples during resection. In-vivo data were patient specific, and rather distinct from the ex-vivo datasets but clear metabolomic signals could be observed. Ex-vivo tissue analysis was carried out on tissue biopsies using iKnife bulk tissue analysis and spatially resolved DESI-MSI and LD-REIMSI. iKnife data from ex-vivo analysis displays a diagnostic accuracy of 96% (Sensitivity 94% (polyp), 95% (Tumour) and specificity 99% (polyp), 97% (Tumour)) when classifying between tumour, polyp and normal tissues. Clustering was observed in mass spectrometry imaging between metabolites associated with cancer and normal cells.

Metabolites of interest could be observed in both faecal and tissue samples using novel ambient ionisation mass spectrometry (AIMS) techniques such as direct faecal analysis and iKnife. Biomarkers associated with cancer onset include very long chain fatty acids (VLCFA) and phosphatidylethanolamine (PE) lipids.

A subsection of samples (n=24) was also analysed using full exome sequencing to investigate the driver mutation oncogenes and their relationship with the metabolomic profile. KRAS mutations were classified with a 91% diagnostic accuracy and were stratified by lysolipid species, MSI displayed visually distinct localisation of lyso-phosphatidylserine species. This deeper characterisation of the tissues enables tailored treatment; for instance, KRAS mutation is associated with resistance to receptor tyrosine kinase inhibitors such as anti-EGFR 1.

CONCLUSION:
This work shows the potential for AIMS techniques in the clinical pathway to synergise patient diagnostics, surgical treatment, downstream aftercare and surveillance. Biomarkers relating to cancer can be mapped through the cancer treatment pathway, separate markers can be used to distinguish subtypes of cancer.

Reference:
1. Zhu, G., Pei, L., Xia, H., Tang, Q. & Bi, F. Role of oncogenic KRAS in the prognosis, diagnosis and treatment of colorectal cancer. Molecular Cancer vol. 20 Preprint at https://doi.org/10.1186/s12943-021-01441-4 (2021).

Do the Presence of Anti-Risankizumab Antibodies Interfere in LC-MS Risankizumab Quantitation? Alex Barbeln (Presenter) Mayo Clinic

Poster #12a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

BACKGROUND:
Risankizumab (RISA) is a humanized IgG1 kappa therapeutic monoclonal antibody (tmAb) that targets interleukin 23 (IL23). RISA is used in the treatment of moderate to severe plaque psoriasis, psoriatic arthritis, Crohn’s disease, and ulcerative colitis. Therapeutic drug monitoring (TDM) can be used to measure RISA serum concentrations, helping to identify thresholds associated with improved outcomes, determine target concentrations for patients or point to a loss of response. While loss of response is characterized by a return of symptoms, loss of response may be linked to either suboptimal RISA pharmacokinetics or the presence of anti-drug antibodies (ADA). Our goal is to show whether the presence of ADA in patient serum would interfere with RISA quantitation.

METHODS:
Anti-Risankizumab clones TZA058 and TZA059 (Bio-Rad; Hercules, CA), were spiked independently at concentrations of 39, 78, 156, 312, 625, and 1250 ng/mL into two pooled residual waste serum samples that were spiked with 5 and 75 µg/mL RISA. The ADA spiked samples were incubated overnight at 8°C. These ADA spikes were then run on the validated RISA quantitation assay and compared to the 5 and 75 µg/mL RISA spikes that were not spiked with any RISA ADA. Briefly, the RISA quantitation assay is a laboratory developed test employing immunopurification with magnetic beads coupled with IL23, followed by liquid chromatography and high-resolution mass spectrometry. Samples were prepared by adding IL23 coupled bead slurry, PBS, unknown/standard/QC, and stable isotopically labeled (SIL)-RISA (Promise Proteomics; Grenoble, France) to a 96-well filter plate (Sigma-Aldrich; St. Louis, MO). Samples were mixed at room temperature, then washed twice with PBS and twice with HPLC water. Samples were then eluted into a PCR plate using 5% acetic acid and reduced with 100 mM DTT in 1M ammonium bicarbonate. The enriched and reduced samples were analyzed using an Agilent 1290 Infinity II LC connected to an AB Sciex 7600 ZenoTOF mass spectrometer. A volume of 20 µL was injected onto the analytical column with separation being performed over a 4-minute gradient. The +11, +12, and +13 charge states were combined to give the XIC used for quantitation for both RISA and SIL-RISA.

RESULTS:
The ADA spikes made with clone TZA058 had an average percent difference of -13.4% (-6.2% to -29.9%) for the 5 µg/mL RISA pool and an average percent difference of -11.1% (-5.8% to -15.1%) for the 75 µg/mL RISA pool. The ADA spikes made from clone TZA059 had an average percent difference of -12.2% (-2.0 to -35.3%) for the 5 µg/mL RISA pool and an average percent difference of -12.6% (-6.4% to -16.9%) for the 75 µg/mL RISA pool. All ADA spikes quantitated within 20% difference from the spiked RISA pools, with exception to the 1250 ng/mL ADA in 5 µg/mL RISA for both the TZA058 and TZA059 clones, which quantitated at -29.9% and -35.3% difference respectively. Although the 1250 ng/mL ADA spikes saw a larger percentage difference from the 5 µg/mL RISA without ADA, the absolute difference was only 1.3 µg/mL for clone TZA058 and 1.5 µg/mL for clone TZA059, which is not a clinically significant drop in RISA concentration.

CONCLUSION:
We have shown that the presence of RISA ADA at high concentrations (1250 ng/mL) could potentially interfere with the drug quantitation assay by suppressing the RISA drug quantitation measurement specifically at low levels of RISA. In the example we’ve shown, this low bias may not be clinically significant. Although the development of RISA ADAs is considered rare (<1% of all patients), when RISA is below 10 µg/mL, it would be best practice to evaluate the presence of ADAs.

Development and Validation of a Sensitive UPLC-QToF Method for Vancomycin, Ceftazidime, and Ceftriaxone in Cerebrospinal Fluid and Serum Chao Sun (Presenter) University of Southern California, Children's Hospital of Los Angeles

>> POSTER (PDF)

Poster #13a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Vancomycin, ceftazidime, and ceftriaxone are antibiotics commonly used to treat central nervous system infections. We developed a sensitive liquid chromatography- quadrupole Time-of-Flight (UPLC-QToF) method that simultaneously measures these antibiotics in cerebrospinal fluid (CSF) and serum. This method will be used to analyze specimens to study the pharmacokinetics and blood-brain barrier penetration of these antibiotics. Additionally, it will aid in dosage optimization to minimize toxicity while maximizing therapeutic efficacy for patients with central nervous system infections.

METHODS:
CSF or serum was mixed with isotope-labeled internal standards and extracted using a 96-well Strata-X-C (Phenomenex) solid-phase extraction plate. The dried eluate was reconstituted with 75 µL of solvent and analyzed using a Xevo G3 QToF mass spectrometer (Waters) with an ACQUITY™ BEH C18 (1.7 µm, 2.1 × 50 mm) column (Waters). Accuracy was assessed using spike-recovery studies with CSF or synthetic serum. The analytical measurement range (AMR) was determined by spiking antibiotics into CSF or synthetic serum to generate seven specimens with increasing concentrations. Precision (repeatability and reproducibility) was assessed using quality control materials. Matrix effect was evaluated by comparing post-spiked extracted samples with neat antibiotic solutions (without solid-phase extraction).

RESULTS:
Linear regression was used to build the calibration curve. In both CSF and serum, the method demonstrated linearity for all antibiotics over a range of 0.1–10 µg/mL, with r² > 0.99. Repeatability and reproducibility were <10% CV. Minimum carryover was observed for both CSF and serum samples. Spike-recovery studies demonstrated recoveries of 89%–119% in serum samples and 91%–105% in CSF samples. Minimal matrix effect was observed in both serum and CSF samples.

CONCLUSION:
We have developed and validated a sensitive and rapid LC-QToF method for the simultaneous measurement of vancomycin, ceftazidime, and ceftriaxone in both serum and CSF.

Detection and Quantification of GLP-1 Analogs in Human Plasma: A Comprehensive Analytical Approach Belysia Walker (Presenter) Thermo Fisher Scientific

Poster #14a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION
Glucagon-like peptide-1 (GLP-1) analogs have emerged as pivotal therapeutic agents in the management of type 2 diabetes and obesity. These analogs mimic the physiological actions of endogenous GLP-1, including enhancing glucose-dependent insulin secretion, suppressing glucagon release, delaying gastric emptying, and promoting satiety. Given their significant clinical utility, accurate detection and quantification of GLP-1 analogs in human plasma are essential for pharmacokinetic studies, therapeutic monitoring, and ensuring patient safety. This study aims to develop a robust analytical method for the detection and quantification of semaglutide and liraglutide, 2 GLP-1 analogs in human plasma. The ability to precisely measure GLP-1 analog levels will facilitate better understanding of their pharmacodynamics and pharmacokinetics, ultimately contributing to optimized clinical outcomes in patients receiving GLP-1-based therapies.

METHODS
Synthetic and HPLC purified semaglutide and liraglutide were purchased from AdipoGen Life Sciences. All calibration and QC samples were prepared in human K2EDTA plasma samples. Protein precipitation was carried out for all samples prior to anionic exchange via the Oasis MAX μ-Elution plate. The eluents were then analyzed by LC-MS. A 10-min reverse-phase gradient was developed on a C4 column using a Vanquish™ Horizon UHPLC system. Data were acquired using Stellar™ MS operated in targeted MSn (tMSn, PRM equivalent) mode, and Chromeleon™ CDS software was used for data acquisition, processing and report generation.

PRELIMINARY DATA
Due to a high degree of chemical similarity between two GLP-1 receptor agonists (i.e., they are based on the same GLP-1 peptide sequence but differ in the attached fatty acid), one could be used as an internal standard for the analysis of the other. Samples were extracted using the Oasis MAX μ-Elution plate prior to LC-MS analysis. A 10-minute gradient was developed to selectively analyze both analogs without any significant matrix interferences. Data were acquired using tMSn mode, which enabled the collection of full MS2 scans for both precursor ions: m/z 1029.3 for semaglutide and m/z 938.8 for liraglutide. We explored 2 quantitative methods: a single precursor/fragment ion transition, and summing multiple fragment ions per precursor, for quantifying both analytes. With the single precursor/fragment ion transition method, we can easily quantify both analogs at 0.1 ng/mL with at least 3 orders of linear dynamic range. All QC samples (LLOQ, mid, and high level) were well within ±15% of the nominal concentration, with precision well within a 15% coefficient of variation. To further improve the sensitivity of the analysis, the peak area of multiple fragments from the same target was summed. By using this approach, we demonstrated at least a 2x improvement in the limit of detection without sacrificing selectivity. The improved sensitivity without additional method development may allow accurate monitoring of GLP-1 analog levels in vivo to detect adverse drug reactions and interactions earlier, improving overall drug safety profiles.

NOVEL ASPECT
A comprehensive analytical solution for quantifying semaglutide and liraglutide in human plasma.

Highly Sensitive Simultaneous Quantification of Human Serum 17β-Estradiol and Estrone Using MPDNP-F Derivatization and LC-MS/MS Koji Takahashi (Presenter) JEOL

Poster #16a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION
17β-estradiol (E2) is the most bioactive estrogen, with its secretion increasing during the onset of secondary sexual characteristics in women. After menopause, estrone (E1) becomes the predominant estrogen in the body. Postmenopausal estrogen concentrations are involved in estrogen-dependent diseases; therefore, a highly sensitive method for the simultaneous determination of E1 and E2 is important for diagnosing these conditions.
For the clinical diagnostics and estrogen related study, one-step derivatization method using 1-(2,4-dinitro- 5-fluorophenyl)-4,4-dimethylpiperazinium iodide (MPDNP-F) has been developed. Here we demonstrated highly sensitive simultaneous quantification of E1 and E2 using this method with LC-MS/MS.

METHODS
A sample of 180 µL (prepared standard solution or serum) was added to the Internal standard solution (E1-13C3 and E2-13C3) 180 µL and vortex-mixed for 1 min. Thereafter, the samples were centrifuged (8000×g, 5 min, 25 °C ), and the supernatant was diluted with 390 µL of water. Extraction was performed using an Oasis HLB elution plate (Waters). We equilibrated 200 µL each of ACN, MeOH, and water, and then loaded the sample onto the extraction plate. The sample was washed with 200 µL of water, followed by 200 µL of 50% MeOH solution (MeOH: water, 1:1, v/v), and eluted with 100 µL of ACN. The eluted analytes were evaporated under a nitrogen atmosphere.
Derivatization was performed 50 µL of MPDNP-F solution (0.4 mg/mL in ACN) and 40 µL of DMAP solution (0.5 mg/mL in ACN) and gently vortexed at 60 °C for 15 minutes. LC-MS/MS was performed using a QTRAP 4500TM (AB SCIEX) coupled to a 1290 Infinity I liquid chromatograph (Agilent) with a CAPCELL CORE C18 analytical column (2.7 µm) 2.1 mm i.d. × 100 mm (OSAKA SODA).

RESULTS
The accuracy of the E2 quantification was estimated using RCR576 and BCR577. Compared with certified standard values, quantified values were 101.8% and 104.2%, respectively. The precision of the assay was evaluated using prepared three different concentrations of E1 and E2 in human serum. The intra-assay RSD for E1 was below 5%, E2 was 2.5%, respectively. The inter-assay RSD for E1 was below 5%, E2 was 3.0%, respectively. The lower limit of quantification for E1 and E2 were 2.0 and 0.8 pg/mL, respectively

CONCLUSIONS & DISCUSSION
We established a method for quantifying E1 and E2 in serum using a one-step derivatization of MPDNP-F. This assay method achieved quantitative sensitivity of 2.0 pg/mL for E1 and 0.8 pg/mL for E2 from a small serum volume (180 µL). This result is one of the highest sensitivity analyses compared with previous reports. Although method validation using clinical samples is required in accordance with guidelines in future study, it is expected to be applied for the clinical research.

Vibrio metchnikovii Bacteremia in a Patient with Cholangitis and a History of Kidney Transplantation Taek Soo Kim (Presenter) Seoul National University Hospital

Poster #17a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

CASE DESCRIPTION:
We present a case of Vibrio metschnikovii bacteremia in a 63-year-old male patient with a history of end-stage renal disease, who underwent kidney transplantation six years earlier. The patient was scheduled for elective surgery for acute cholecystitis and had undergone percutaneous transhepatic gallbladder drainage (PTGBD) prior to admission. On the day of admission, he developed fever and chills unresponsive to acetaminophen in the early morning, followed by nausea and multiple episodes of vomiting in the afternoon, prompting a visit to the emergency department.
At presentation, the patient was alert but hypotensive (blood pressure 100/71 mmHg) and febrile (body temperature 38.9°C). Laboratory findings revealed leukopenia, thrombocytopenia, and an elevated hs-CRP level of 8.97 mg/dL. With a clinical suspicion of septic shock secondary to cholecystitis, empirical antimicrobial therapy with meropenem was initiated after obtaining blood cultures. On hospital day 1, Vibrio metschnikovii was identified via the Biotyper Sirius MALDI-TOF MS system (Software version 4.1.100, Bruker Daltonics GmbH, Bremen, Germany). Antimicrobial susceptibility test (AST) showed susceptibility to cefotaxime, ciprofloxacin, and tetracycline, but resistance to ceftazidime. Based on the AST, the antimicrobial regimen was adjusted to ciprofloxacin. On hospital day 2, the same organism was isolated from the PTGBD fluid. Follow-up blood cultures on days 2 and 4 were negative. The patient's fever and laboratory abnormalities gradually improved, and he was discharged in stable condition on hospital day 7.

BACKGROUND:
Vibrio species are oxidase-negative, nitrate reduction–negative, facultative anaerobic, curved Gram-negative bacilli. Over 100 Vibrio species have been identified, all of which are known to inhabit marine and estuarine environments. Among these, 15 species have been reported to be isolated from humans. The most commonly encountered human pathogens include Vibrio cholerae, V. parahaemolyticus, and V. vulnificus. Other species such as V. mimicus, V. fluvialis, V. furnissii, and V. alginolyticus are considered to have pathogenic potential, but are generally associated with milder clinical courses. Rare species such as V. metschnikovii, V. cincinnatiensis, V. metoecus, V. injenensis, and V. harveyi have been identified in sporadic case reports, but their clinical significance remains poorly understood.

MS METHOD and RESULT:
Identification was performed using the Bruker Biotyper Sirius MALDI-TOF MS system, which identified Vibrio metschnikovii with a score value of 2.13.

DISCUSSION and CONCLUSION:
Vibrio metschnikovii was named in honor of the renowned Russian microbiologist Elie Metchnikoff. Although the organism is widely distributed in seawater, freshwater, and sewage, human infections are exceedingly rare. The first reported human case was described in 1981 in Chicago, when the bacterium was isolated from the blood of a patient with cholecystitis. Since then, only 14 clinical cases have been reported worldwide, including three cases of bacteremia. Most involved elderly or immunocompromised individuals, and in many cases, the source of infection was unclear.
In Korea, the only previously reported case of V. metschnikovii infection was in 1999, involving a patient with peritonitis and septic shock. To our knowledge, this is the second reported case in Korea and the first in which V. metschnikovii was simultaneously isolated from both blood and PTGBD fluid in a patient with cholecystitis and septic shock.
Notably, MALDI-TOF analysis enabled rapid identification from short-term culture allowing for a reduction in reporting time by more than 8 hours and timely delivery of clinically relevant results.

Deciphering Spatial Variation in Lipid Profiles of X-Ray Irradiated Cells Using nLC and Orbitrap 240 MS Giovanni Pallante (Presenter) Thermo Fisher Scientific

Poster #18a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION
Lipidomics is of significant interest in cell biology because cells contain lipid droplets, which play an important role in managing cell stress in response to infection, irradiation and drug treatment. Herein, we describe an optimized workflow for single-cell lipidomics using Thermo Scientific Orbitrap Exploris™ 240 Mass Spectrometer, combined with nano-LC separation. The methodology is applied to investigate the lipidomic impact of X-ray radiation and its bystander effect on adjacent cells.

METHODS
PANC-1 cells were irradiated with X-rays, followed by live spatial sampling using the Single Cellome™ SS2000 (Yogokawa) system. NanoLC-MS was performed on a Thermo Scientific PepMap™ - column connected to a Thermo Scientific Vanquish™ Neo UHPLC system, with data acquisition in both positive and negative ionization modes. Thermo Scientific LipidSearch™ 5.0 and Compound Discoverer™ 3.3 software aided lipid annotation and data processing.

RESULTS
The nLC-MS method was optimized using lipid standards and bovine liver lipid extracts. The reproducibility, robustness, and sensitivity of the workflow were assessed using cell dilution series. The SS2000 system enabled accurate collection of single cells based on high-resolution confocal microscopy images. For cells irradiated with X-Ray, the system was used to collect single cells spatially, based on their distance from irradiated areas.

Single cells were collected and processed for nLC-MS analysis, with the SS2000 system preserving metabolic activity by isolating single adherent cells under incubation conditions. The lipid extracts were analyzed using polarity switching to cover a comprehensive range of lipid species.

DISCUSSION
Differences in cellular lipid profiles were observed based on the spatial location of the cells, indicating potential implications for cellular response to radiation and bystander effects. Major differences were noted between cells located at the periphery and those in the middle of the plate.

Development and Validation of a UPLC-MS/MS Method for Analyzing Drugs Commonly Used in Children with Attention-Deficit/Hyperactivity Disorder in Urine Ching-Mei Chen (Presenter) Chang Gung Medical Foundation

>> POSTER (PDF)

Poster #19a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Attention Deficit Hyperactivity Disorder (ADHD) is a prevalent neurobehavioral disorder, particularly among school-age children. This condition not only impacts children's daily lives and learning but may also persist into adolescence and adulthood. In Taiwan, the prevalence of ADHD is approximately 9%. The primary medications currently used to treat ADHD include central nervous system stimulants, such as methylphenidate, and norepinephrine reuptake inhibitors, such as atomoxetine. Additionally, because individuals with ADHD often experience manic symptoms, doctors may prescribe aripiprazole, a serotonin 5-HT2A receptor antagonist. While these medications are effective, they can also lead to various psychiatric side effects. Furthermore, individual differences in drug metabolism necessitate precise dosage adjustments to achieve optimal therapeutic effects while minimizing adverse reactions, particularly in the vulnerable populations of children and adolescents. These medications typically have a short half-life and are primarily excreted in the urine as metabolites.

OBJECTIVE(S):
The objective of this study is to develop a rapid and non-invasive method for determining the concentrations of methylphenidate, atomoxetine, aripiprazole, and their metabolites in urine. Furthermore, the study aims to compare the consistency between individual plasma drug concentrations and their corresponding metabolite concentrations with urine drug concentrations.

METHODS:
100 μL of urine was treated with glucuronidase and subjected to HLB solid-phase extraction. The urine was then directly analyzed using UPLC-MS/MS quantification. The three drugs and their metabolites were detected via electrospray ionization in positive mode, and the analytes were monitored in multiple reaction monitoring mode. The total analysis time for each sample was 3.5 minutes.

RESULTS:
We conducted method validation in accordance with the CLSI C62-A guidelines, establishing a detection limit of 1 ng/mL. The standard curve demonstrated linearity within a concentration range of 1-1000 ng/mL, with a linear correlation coefficient exceeding 0.999. Intra-batch and inter-batch imprecision were assessed using quality control samples, yielding values of less than 2.9% (n = 40) and 5.2% (n = 40), respectively. Accuracy was evaluated through recovery studies, with average recoveries for the six analytes ranging from 96.6% to 103.7%. This method exhibited no ion suppression or enhancement phenomena, and the presence of proteinuria and hematuria in samples did not impact the test results. Additionally, we analyzed the concentrations of three drugs and their metabolites in plasma and urine samples from 417 ADHD patients, confirming significant differences in drug concentrations among individuals receiving the same dosage.

CONCLUSION:
This study established an accurate, simple, and non-invasive UPLC-MS/MS method for the quantitative detection of ADHD medications in urine. It also demonstrated the significance of monitoring ADHD medication concentrations for personalized treatment and dosage adjustments, ensuring that each patient receives optimal treatment outcomes.

Leukocyte and CSF Arylsulfatase A Assay by LC-MS/MS Jie Chen (Presenter) Children's Hospital of Philadelphia

>> POSTER (PDF)

Poster #20a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Metachromatic leukodystrophy (MLD) is a rare, severe, neurodegenerative disorder caused by deficient activity of arylsulfatase A (ARSA), an enzyme responsible for cleaving the sulfate group from sulfatides (3-O-sulfogalactosylceramide) into galactosylceramide (GalCer). In MLD, sulfatide accumulation leads to progressive demyelination. A disease-modifying gene therapy is available for MLD but is currently only approved for the late-infantile and early juvenile forms, highlighting the need for disease stratification. Although genotype–phenotype correlations have been identified, they remain incomplete. Diagnostic biochemical approaches, including ARSA activity measurement and sulfatides quantitation, provide complementary information but remain limited for their prognostic utility. Santhanakumaran et al improved the conventional spectrophotometric ARSA activity assay and demonstrated that residual ARSA activity correlates with disease severity (1). However, spectrophotometric assays have several technical limitations, including poor sensitivity and selectivity due to the use of artificial substrate. We propose that an LC-MS/MS-based ARSA activity assay can overcome these limitations by enabling the use of a natural substrate, thereby providing more accurate and reliable measurements of ARSA activity (2). This will be crucial not only for diagnosis, but also for prognosis and treatment monitoring in MLD.

OBJECTIVES:
The study aims to develop a high-performance LC-MS/MS assay to measure ARSA activity in various biological matrices, including white blood cells (WBCs) and cerebrospinal fluid (CSF) for diagnostic, prognostic, and treatment monitoring purposes; and to conduct a comprehensive validation of the assay to support its implementation in a clinical diagnostic laboratory.

METHODS:
CSF and WBC lysates (isolated from EDTA-treated whole blood) were incubated with a reaction cocktail containing d₃-C18:0-sulfatide as the enzymatic substrate and d7-C18:0-GalCer as the internal standard. D₃-C18:0-sulfatide was selected as the substrate so that the enzymatic product would retain the three deuterium labels, allowing it to be distinguished from endogenous C18:0-GalCer by MS/MS. The reaction was carried out at room temperature with overnight incubation. Following incubation, the mixture was quenched and analyzed by UPLC-MS/MS, where the enzymatic product, d₃-C18:0- GalCer, was chromatographically separated from the undigested substrate, d₃-C18:0-sulfatide. An external calibration curve, combined with isotope-dilution using the internal standard, was used to quantify d₃-C18:0- GalCer generated through the enzymatic reaction. Enzyme activity was calculated by normalizing the amount of enzymatic product to the incubation time and volume (for CSF) or to the WBC cell count (for WBC lysates).

RESULTS:
ARSA activity was readily detectable in 10 µL of normal CSF and approximately 50,000 white blood cells, corresponding to 4.5 to 12.5 µL of whole blood from a healthy individual. The WBC isolation protocol we adopted minimized the loss of ARSA during the procedure, but it also resulted in RBC contamination in the final cell pellet, necessitating normalization of ARSA activity to WBC cell count rather than protein concentration. ARSA activity in RBCs was found to be at least 100-fold lower than in WBCs, supporting this approach. We obtained WBCs from a late-infantile MLD patient during initial evaluation and detected no residual ARSA activity, demonstrating that the assay is highly specific to the enzyme of interest.

The external calibration curve covered a range of 10-20,000 nM d₃-C18:0-GalCer. The accuracy and precision (A&P) and lower limit of quantitation (LLOQ) of the LC-MS/MS platform were assessed with neat d₃-C18:0-GalCer solution. A&P were evaluated at three levels (30, 1,200, 12,000 nM) with bias <10% and imprecision <10%. The LLOQ was determined at 10 nM with the bias <10% and imprecision <12.5%. The linearity and precision of the entire system, including enzymatic reaction and LC-MS/MS measurement, were evaluated using real CSF and WBC lysates. Linearity of ARSA activity in CSF was demonstrated across dilution levels ranging from 6.25% to 100%, with an R² value of 0.9998. For WBC ARSA activity, linearity was demonstrated over a 10% to 100% dilution range, with an R² value of 0.991. Total imprecision for CSF measurements was evaluated using pooled CSF samples mixed with heat-inactivated CSF at 2%, 20%, and 100% activity levels. Similarly, total imprecision for WBC measurements was assessed using pooled WBC lysates mixed with heat-inactivated WBC material at 2%, 10%, and 100% levels. Overall, these results underscore the strong analytical performance of this assay.

CONCLUSION:
Our isotope dilution LC-MS/MS assay for ARSA is highly selective, precise, and sensitive. It shows strong potential for use in presymptomatic diagnosis of MLD and in predicting disease onset.

REFERENCES:
(1) Santhanakumaran, V., Groeschel, S., Harzer, K., Kehrer, C., Elgün, S., Beck-Wödl, S., ... & Laugwitz, L. (2022). Predicting clinical phenotypes of metachromatic leukodystrophy based on the arylsulfatase A activity and the ARSA genotype?–Chances and challenges. Molecular Genetics and Metabolism, 137(3), 273-282.
(2) Hong, X., Kumar, A. B., Daiker, J., Yi, F., Sadilek, M., De Mattia, F., ... & Gelb, M. H. (2020). Leukocyte and dried blood spot arylsulfatase A assay by tandem mass spectrometry. Analytical chemistry, 92(9), 6341-6348.

To Free or Not to Free? An LC-MS/MS-Based Method Modification to Validate Clinical Performance of Bioavailable Testosterone Measurement Chelsea Swartchick (Presenter) Mayo Clinic

Poster #21a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION
Measurement of total testosterone is an invaluable method for assessing an individual’s androgen status. This assay is often performed in conjunction with free testosterone measurement, which should encompass approximately 1 to 2% of the total testosterone concentration. However, measurement of free testosterone is influenced by abnormalities in both the concentration and/or function of sex hormone binding globulin (SHBG), the primary transport and regulatory protein for testosterone, along with albumin, the most abundant plasma protein. Therefore, analysis of bioavailable testosterone, which accounts for both free and testosterone loosely bound to albumin, is a more reliable representation of an individual’s true testosterone status. Recently, our bioavailable testosterone assay underwent a method modification to enhance calibrator storage and quality, which required subsequent validation to confirm adequate clinical performance. It has been shown that a strong correlation exists between bioavailable and free testosterone. As such clinical performance was validated by investigating the concordance between the reference interval (RI) flags (low, normal, or high) obtained using the modified BAT method and the RI flags from validated LC-MS/MS total and free testosterone assays.

METHODS
Residual serum samples (n = 65) from patients with recent (< 2 weeks) total and free testosterone measurements were used, where 23 were from women and 42 were obtained from men. Reference interval flags were categorized as high, normal, or low as determined being above, within, or below established reference intervals, respectively. There were 14 specimens classified as high, 36 as normal, and 15 as low, by both their total and free testosterone measurements. Bioavailable testosterone sample preparation involved the differential precipitation of testosterone-bound SHBG with ammonium sulfate. All non-SHBG bound testosterone remained in the supernatant, which subsequently was spiked with internal standard, carbon-13 labeled testosterone. Liquid-liquid extraction was performed, and the resulting isolate was derivatized before being quantified by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). All 65 bioavailable testosterone results were then categorized as either high, normal, or low according to the RI dependent on the patient’s sex and age. Flags from the bioavailable testosterone assay and flags from the free/testosterone measurements were evaluated for agreement. Serum SHBG was quantified using a chemiluminescent sandwich immunoassay. Serum albumin concentration was determined using a colorimetric method using the absorbance at 570 nm.

RESULTS
59 of the 65 (90.7%) samples were correctly categorized with a RI flag as either low, normal, or high by both the free/total testosterone assays and the modified bioavailable testosterone method. The 6 samples that had discrepant flags required further investigation in which serum SHBG and albumin concentrations were evaluated. Of the discordant samples, 4 had elevated free and total testosterone, but normal bioavailable testosterone. Follow-up of this group demonstrated high SHBG concentrations. Elevated SHBG would expectedly bind more testosterone and decrease the bioavailable testosterone concentration, which provides a rationale of the normal categorization in this population. In a similar manner, one sample had a slightly elevated free and total testosterone, but normal bioavailable testosterone. The SHBG concentration of this sample was nearing the upper limit of normal, which follows the same principle as above where the bioavailable testosterone would be flagged as normal. The remaining discrepant sample had normal free and total testosterone, but elevated bioavailable testosterone. This specimen had an SHBG at the lower end of normal and an albumin at the higher end of normal. As there is less SHBG to bind testosterone and more albumin, an elevated bioavailable testosterone is appropriate.

CONCLUSION
Clinicians often prefer bioavailable testosterone as it more closely resembles the total bioactive testosterone in circulation when compared to free testosterone. Validation to ensure clinical accuracy of our bioavailable testosterone method modification involved a comparison of the reference interval flags between this method and verified free and total testosterone flags. In theory, samples that flag normal by both free and total testosterone assays should flag similarly with the bioavailable assay. We observed concordance in 90.7% of our samples, while the remaining 9.3% specimens had abnormal SHBG and/or albumin levels that were reflected appropriately in bioavailable testosterone RI flagging. In summary, our method modification performs clinically and can reflect the true testosterone status within a patient.

Simultaneous Quantitation and Discovery (Squad) of Fecal Bile Acids and Conjugates in Adults With Autism Spectrum Disorder Ema Ruzic (Presenter) Thermo Fisher Scientific

Poster #22a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Bile acids (BA) are synthesized from cholesterol in the liver and play a crucial role in lipid digestion and absorption. Fecal bile acids serve as biomarkers and signaling molecules due to their intricate interplay with gut microbiota. Disordered microbiomes can alter the composition and size of the bile acid pool, producing a variety of conjugated bile acids and structurally similar metabolites. These changes may indirectly confer disease states. Here, a metabolomics workflow was developed for the simultaneous quantitation and discovery (SQUAD) of fecal BA and BA conjugates on the Thermo Scientific™ Orbitrap Ascend™ Tribrid™ mass spectrometer. The workflow utilizes Real-Time Library Search (RTLS) for spectral similarity measures and heightened identification confidence for molecular species during method execution.

METHODS:
Human stool materials were obtained from adults with and without autism spectrum disorder (ASD). BA and microbially conjugated bile acid (MCBAs) standards were purchased from Bileomix Inc. Analytes were separated on a Thermo Scientific™ Hypersil GOLD™ column within a Thermo Scientific™ Vanquish™ Horizon system. Data were acquired on the Thermo Scientific™ Orbitrap Ascend™ Tribrid™ mass spectrometer, which facilitates sensitive and highly dynamic PRM-quantitation utilizing the linear ion trap. It also allows Orbitrap MS1 scanning for higher annotation rates. AcquireX™ and RTLS workflows were used to maximize the number of relevant compounds interrogated by MS2 and MS3 for confident annotation. Thermo Scientific™ TraceFinder™ 5.2 and Compound Discoverer™ 3.4 software were used for data processing, analyte quantitation, and unknown annotation.

RESULTS:
The BA and BA conjugate standards were used to evaluate a SQUAD metabolomics workflow for BA analysis in the collected fecal samples on an Ascend MS. The standards were used to generate calibration curves for absolute quantitation and to build an MS2 spectral library for the annotation of relevant unknown BA and BA conjugates via RTLS. This workflow facilitates reliable ion trap quantitation over a wide dynamic range (i.e., 5 – 6 orders of magnitude) of the targeted BA in feces. A lower limit of quantification (LLOQ) of 10 femtomoles and a lower limit of detection (LLOD) of 0.25 femtomoles was observed for most of the targets. Moreover, MSn-based quantitation of the mass analyzer enabled the selectivity required to detect and accurately quantify co-eluting isomers and isobars. It also leads to improved discrimination between signals derived from analytes and those resulting from matrix interferences.

Simultaneously, the HRAM Orbitrap data and the increased percentage of fragmented compounds using the advanced deep scan AcquireX workflow resulted in improved annotation capability compared to traditional DDA on a wider dynamic range of fecal compounds. Additionally, Real-Time Library Search provided spectral similarity measures and identification confidence scores for BA species upon which acquisition decision-making can be based during method execution.

DISCUSSION:
In our analysis we identified several microbially conjugated bile acids that were previously determined to be associated with gut conditions including IBD, in particular, we found in high abundance leucine/isoleucine, tyrosine, and phenylalanine conjugated bile acids. It’s important to note that neither chromatography nor MS2-level fragments were able to distinguish Isoleucine from leucine-conjugated bile acids. MS3 level on the Ascend system was required to detect specific fragments and quantify levels of Ile/leu conjugates.

Identification and Quantitation of Catechol Metabolites in Plasma after Intranasal Instillation Exposure of Smoldering Wood Smoke Extract Bruce Buchholz (Presenter) Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory

Poster #23a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Widespread exposure of wood smoke to the general population has increased dramatically in the past 20 years as wildfire intensity, frequency and geographical distribution have also grown. Inhalation of wood smoke has been linked to adverse cardiovascular, respiratory, and neurological effects. Phenolic compounds are prominent constituents of wood smoke, and catechol is especially abundant under smoldering conditions that produce dense smoke. Parallel Accelerator and Molecular Mass Spectrometry (PAMMS) was used to identify and measure metabolites of catechol in plasma after inhalation exposure to smoldering eucalyptus wood smoke extract in rats.

OBJECTIVE:
Identify and measure catechol metabolites in rat blood after inhalation exposure.

METHODS:
Smoldering eucalyptus wood smoke extract (WSE) was provided by U.S. EPA [1] and underwent solvent exchange to a saline solution with working concentration of 1 mg/mL. The WSE was spiked with 14C-catechol to attain a specific activity of 1.14kBq/umol. Male Sprague Dawley rats (n=5) were exposed by intranasal instillation to 50 ng/g of 14C-catechol in 1mg/mL of WSE. Blood was collected at 5 min and 2 hours post-dose, transferred to heparin-coated tubes, spun at 8,000g to separate plasma from blood, and stored at -20C. Analytes were extracted and purified from plasma using solid-phase extraction with C18 cartridges.

LC-MS analyses were conducted using a UHPLC and reverse-phase separation. Gradient elution was employed with mobile phase A consisting of 95:5 water:acetonitrile (v/v) with 10 mM ammonium formate and 0.125% formic acid, and mobile phase B consisting of 5:95 water:acetonitrile (v/v) with 0.125% formic acid. A flow splitter divided the eluent 50/50 between the Orbitrap Eclipse (Thermo Fisher) for molecular mass spectrometry and the moving wire interface for accelerator mass spectrometry (AMS).
Molecular mass spectrometry was performed using heated electrospray ionization (HESI) in negative ion mode. The spray voltage was set to 2,500 V, with the RF lens at 50%. Sheath gas, auxiliary gas, and sweep gas were set to 35, 7, and 0 arbitrary units, respectively. The ion transfer tube temperature was 300C, and the vaporizer temperature was 275C. Full-scan acquisition was conducted at a resolution of 120,000 m/z with a mass range of 80–800 m/z. Fluoranthene was used as an internal calibrant. Data was analyzed using FreeStyle software (Thermo Fisher).

AMS analysis was performed using LLNL’s custom-built moving wire interface. Briefly, the LC eluent was continuously deposited onto a moving nickel wire, which then passed through a drying oven maintained at 120C to remove solvent and other volatile compounds. The dried sample was combusted in an 800C oven, converting carbon-containing species to CO₂. The CO₂ was transferred via capillary to the AMS source. A 250 kV AMS (NEC; Middleton, WI, USA) was used to detect and quantify both stable and radiocarbon isotopes. Data was analyzed using in-house software.

RESULTS:
Four distinct elution peaks were detected in the 14C chromatograms of all samples, indicating that the 14C-catechol had been metabolized into at least four distinct products. To identify the radiolabeled ions, extracted ion chromatograms (EICs) from the orbitrap mass analyzer were overlaid on the 14C-chromatagram. The first peak (RT = 2.35 min) corresponded to the [M-H]⁻ ion of C₉H₁₇O₅N (m/z 218.1034), consistent with a catechol-glutamine conjugate. This assignment is supported by the accurate mass, elemental composition, and the biological context, in which catechol-derived o-quinones are known to undergo nucleophilic attack by amino acids such as glutamine, forming stable conjugates via Michael addition. The second peak (RT = 3.56 min) matched the [M-H]⁻ ion of benzene oxide (C₆H₆O, m/z 93.0346), a well-known reactive intermediate formed during the oxidative metabolism of aromatic compounds. The third peak (RT = 3.90 min) was identified as the [M-H]⁻ ion of C₈H₇7O₄NS (m/z 212.0023), consistent with a cysteine conjugate of catechol. This structure is supported by the presence of sulfur and nitrogen in the formula, characteristic of conjugates formed via the mercapturic acid pathway in response to catechol-derived electrophilic species. The fourth peak did not correspond to any detectable ion in the orbitrap data, suggesting it may represent a radiolabeled compound that was undetectable by mass spectrometry under the conditions used or present at levels below the detection limit. The parent compound was not detected.

To assess temporal changes in metabolite levels, relative peak intensities were compared between the 5-minute and 2-hour timepoints. The third peak (C₈H₇O₄NS) exhibited a statistically significant change in abundance over time, as determined by a paired t-test. The observed decrease in the third peak over time suggests that the cysteine conjugate is transient, potentially reflecting rapid clearance or further metabolism following its initial formation. This pattern is consistent with a short-lived intermediate in the mercapturic acid pathway, formed early after catechol exposure and subsequently processed or excreted.

CONCLUSION:
These results indicate that inhaled catechol from wood smoke quickly gets metabolized upon entry into systemic circulation. PAMMS enabled both identification and relative quantification of circulating catechol metabolites in plasma, demonstrating its potential for biomarker discovery and exposure assessment. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-ABS-2008234.

REFERENCE:
1. Kim, Y.H., et al., Mutagenicity and Lung Toxicity of Smoldering vs. Flaming Emissions from Various Biomass Fuels: Implications for Health Effects from Wildland Fires. Environ Health Perspect. 2018. 126(1):017011.

Developing a Lipidomics-Based Targeted Clinical Diagnostic Method From a Discovery Platform Charles Nichols (Presenter) AOA Dx

Poster #24a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Ovarian cancer (OC) is the 5th leading cause of cancer-related deaths in women, with >70% cases diagnosed at late stages where five-year survival ranges from 10-30%. In the U.S., the average time to diagnosis for OC is 9 months, largely due to the presentation of vague abdominal symptoms (VAS) which overlap with more common conditions. Additionally, the current standard of care diagnostic method (CA125 plus imaging) lacks sensitivity and specificity, especially for early stage. However, >80% of individuals experience VAS during early-stage disease, where five-year survival jumps to >90%. By developing a diagnostic tool targeting this symptomatic population, we can enable earlier OC detection and improve patient outcomes. AOA Dx is taking a novel discovery-based approach, ultimately harnessing the potential of lipidomics and metabolomics for the identification of biomarkers to improve early detection of OC.

OBJECTIVE(S):
Enable biomarker discovery by using untargeted methods on a HRMS platform and strategic cohort design. Characterize discovery biomarkers across cohorts. Translate significant, reproducible biomarkers to a targeted triple quadrupole (QqQ) method. Validate the targeted method for sample classification. Validate the identity of molecular features by acquiring and characterizing quality reference standards.

METHODS:
Two clinically annotated cohorts were designed to represent women with symptoms of OC (N1=544 and N2=423). Both cohorts included patients diagnosed with OC across subtypes and stages (N1=219 total: 80 early-stage (I/II), 139 late-stage (III/IV); N2 = 109 total: 52 early-stage (I/II), 57 late-stage (III/IV)) and non-cancerous controls including healthy donors (N1=82, N2=208), benign gynecological disorders (N1=168, N2=86), borderline tumors (N1=25, N2=20), and gastrointestinal disorders (N1=50). Each cohort was acquired independently using our discovery lipidomics platform.
Serum samples were prepared for LCMS analysis using methanol extraction and centrifugation without evaporation/reconstitution. Data were acquired using LC-HRMS on a Vanquish LC coupled to an Exploris 240 Orbitrap (Thermo Scientific). A 20-minute reverse-phase chromatographic separation was employed. The detector method was parsed into discrete experiments to acquire low-mass lipids and fatty acids [m/z 90-500] and high-mass lipids [m/z 500-1900] in efforts to minimize signal suppression of high molecular-weight ions by high-abundance low–molecular-weight species. ddMS2 methods were employed to improve the quality and number of feature identifications.
QqQ method development was executed on an I-Class HPLC coupled to a TQ-Absolute (Waters Corporation) using the equivalent chromatographic conditions. To facilitate MRM development, HRMS features of interest were subjected to targeted MS2 fragmentation at multiple collision energies. Our discovery feature-list was cross-referenced with Waters Corporation’s LipidQuan MRM database to facilitate method development.

RESULTS:
LC-HRMS features were deconvoluted and putatively identified using Compound Discover (3.3 SP3; Thermo Scientific). The molecular features were filtered by chromatographic quality, signal intensity, and significance. Peak Areas were normalized using the SERRF QC algorithm, and both Raw and Normalized Peak Areas were subject to significance ranking. Features that failed to report a library match were removed from further processing. Discovery LCMS data was also processed in Maven (2.10.20.1) using targeted libraries for fatty acids and gangliosides. After feature filtering, the two cohorts yielded 5,797 and 3,816 putative identifications, respectively. All features were individually characterized on their ability to correctly classify the cohort samples into non-cancer vs. cancer groups using machine learning AUCs. Top performing features were considered for their potential as clinical biomarkers and transfer to the QqQ. Our early targeted method consists of approximately 50 significant features and will be supplemented with additional compounds from the same class. Preliminary feature list includes lipophilic compounds of the classes cerebrosides, glycosphingolipids, phospholipids, steroids, and triglycerides.

DISCUSSION:
Early machine-learning investigations suggest that differences in biomarker abundance in both late-stage and early-stage OC serum compared with biologically complex controls can classify the presence or absence of ovarian cancer.
Translating tentative and putative identifications into a targeted method is challenging. While HRMS instruments yield high confidence molecular formulas, tentative identifications do not contain structural information. Further, the structural information within putative identifications vary in quality depending on the number of product ions matched to fragmentation database. Fortunately, lipid fragmentation is well characterized for many classes of lipids, with some classes yielding characteristic product ions, leaving minimal room for ambiguity for some putative identifications. However, nuances, such as double bond positions or fatty acid positioning, can be challenging to determine from MS2 information alone.
We are in the process of translating potential clinically significant features into a targeted QqQ method.
We are using multiple information sources to guide method development for our first targeted method. Our first effort is to employ ddMS2 or targeted MS2 data generated by the Orbitrap to build MRM transitions. We are also reviewing transitions from lipidomics methos published by LCMS vendors. Combining MRM transitions with retention time data reported by the Orbitrap allows early iterations of MRM methods to be quickly generated. Our ongoing work will repeat cohort samples on the triple quadrupole to verify feature significance across instruments. Concurrently, we are building a standards library as future work will focus on feature validation.

Selective and Sensitive Quantitation of 18 steroids in Human Serum using Thermo Scientific™ Stellar™ Mass Spectrometer Lauren Bishop (Presenter) UCSF

Poster #25a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Successful biomarker quantitation in complex matrices requires sensitivity and specificity to ensure the measured signal is attributed to the targeted analytes. To reduce background interferences and provide greater specificity, alternative fragmentation mechanisms and/or MS3 fragmentation are an efficient solution. The Thermo Scientific™ Stellar™ mass spectrometer (MS) is a hybrid quadrupole, dual-pressure linear ion trap MS that offers two types of orthogonal, yet complementary, collision-induced dissocation (CID) with rapid MS2 and MS3 scans. This poster is the first to demonstrate the advantages of utilizing multiple collision activations and MS3 scans provided by Stellar MS to the selective and sensitive quantification of 18 steroids in serum.

METHODS:
Commercially available 18 steroids standards, including 4 progestagens, 5 androgens, 2 estrogens, 3 mineralocorticoids, and 4 glucocorticoids, and their corresponding internal standards were serially diluted in 0.05% bovine serum albumin to generate calibrator samples over 4-orders of magnitude concentration range. Quality control human serum samples were purchased from Chromsystem. Steroids were extracted by protein precipitation, dried, and reconstituted in 50% methanol for the LC-MS analysis. Multiple fragmentation schemes were evaluated and optimized on the Stellar MS, and the data was analyzed in Thermo Scientific™ TraceFinder™ software for the quantitative performance, including detection limit, quantitation limit, and linearity.

RESULTS:
The detection of 18 steroids were optimized in Stellar MS under multiple fragmentation mechanisms (beam-type CID, resonance-type CID, and MS3 in different combinations) with polarity switching. Breakdown curves were generated to facilitate characteristic fragment ion selection to ensure high sensitivity and selectivity. Adaptive RT function was utilized to adjust the scheduled retention time window of the analyte in real-time to minimize any chromatographic changes such as aging column. Analytical performance of the calibration curves with R2 > 0.98 were achieved using a weighting factor of 1/x, and the limit of quantification (LOQ) values are established with % RSD < 20, │% Diff│ < 20, with relative ion ratio meeting the values specified by EU Council Directive 96/23/EC. Comparing to the MS2 beam-type CID commonly used in contemporary triple quadruple MS, steroids such as cortisone, estradiol, and dihydrotestosterone benefited from the alternative, resonance-type CID and MS3 acquisition modes, resulting in a 2-10 fold increase of the LOQ values.

CONCLUSION:
Stellar MS provides unique selectivity allied to fast acquisition for high-throughput and sensitive quantitation optimal for the routine analysis of endogenous steroids.

Adenine Phosphoribosyltransferase Deficiency: Metabolic Profiling of DHA Crystal-Induced Kidney Injury Vazha Dzhedzheia (Presenter) University of Iceland

Poster #26a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Adenine phosphoribosyltransferase (APRT) deficiency is a disorder of adenine metabolism characterized by abnormal urinary excretion of the poorly soluble 2,8-dihydroxyadenine (DHA), leading to kidney stone formation, crystal-induced kidney injury, and chronic kidney disease. Although treatment options exist, some patients experience adverse reactions or progress to end-stage kidney disease.

OBJECTIVES:
The purpose of this study is to understand the mechanisms of DHA crystal-induced kidney injury by identifying distinct metabolic and lipidomic changes linked to disease progression, characterizing inflammatory and renal injury biomarkers, and correlating metabolic disruptions with clinical outcomes and treatment efficacy.

METHODS:
This study employs an integrative omics approach using UPLC-MS-based metabolomics and lipidomics to analyze plasma and urine samples from APRT-deficient patients, collected as part of an open-label, randomized crossover clinical trial evaluating the effects of allopurinol and febuxostat on DHA plasma levels and urinary excretion. Metabolomic sample preparation will involve protein precipitation with organic solvents using an automated liquid handling system to ensure analytical reproducibility. This method will also be applied to plasma lipid extraction, while urine samples will undergo liquid-liquid extraction techniques for more effective lipid isolation. Data will be processed and interpreted using Lipostar2 and Progenesis QI software to identify significant metabolic and lipid alterations associated with the pathophysiology of APRT deficiency.

RESULTS:
To date, we have analyzed samples from 66 APRT-deficient patients and successfully annotated 313 distinct lipids across five major classes: Phospholipids, Glycolipids, Triglycerides, Diglycerides, and Cholesteryl Esters. Initial data inspection confirms high-quality spectral profiles, and preliminary statistical analysis suggests altered lipidomic profiles within the patient cohort. These lipidomic signatures have yet to be correlated with clinical biomarkers, treatment strategies, and disease outcomes. Next steps include metabolomics analysis of the same patient samples, comprehensive pathway analysis, correlation of lipidomic and metabolomic signatures with treatment responses, and integration of these multi-omics data to establish potential biomarkers of disease progression and treatment efficacy.

CONCLUSION:
The findings will provide insights into dysregulated metabolic pathways associated with DHA crystallization, identify prognostic biomarkers, and propose potential therapeutic targets. Ultimately, this research aims to improve the clinical management of APRT deficiency and related crystalline nephropathies.

Sensitivity Improvements in the Quantitation of Steroids on the SCIEX 7500+ System Ons Ousji (Presenter) SCIEX

Poster #28a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Steroid analysis is essential in clinical diagnostics and research due to the significant role steroids play in regulating metabolism, immune responses, and reproductive functions. Further, accurate quantification of steroid levels is vital for diagnosing and managing various health conditions.

Historically, steroids have been measured using immunoassays and GC-MS, but these methods have limitations. For example, immunoassays often suffer from cross-reactivity, leading to inaccurate results, while GC-MS requires extensive sample preparation. Therefore, LC-MS/MS has become the preferred method, offering high sensitivity and specificity with minimal sample preparation. LC-MS/MS can simultaneously measure multiple steroids in a single sample, making it ideal for clinical research and diagnostics.

High sensitivity is crucial in steroid analysis because many steroids exist in very low concentrations within biological samples. Enhanced sensitivity can also enable reduced injection volumes, which is beneficial when sample volumes are limited, particularly in clinical settings or research involving rare specimens.

OBJECTIVES:
The objective of this work was to develop a method for the analysis of 18 steroids in solvent using a next generation mass spectrometer, the SCIEX 7500+ system. The advantages of this high sensitivity were explored by comparing the results with the previous generation of triple quadrupole mass spectrometer, the QTRAP 6500+ system.

METHODS:
Analytical grade standards for the target analytes were purchased from commercial sources, and a mixed stock solution was prepared in methanol. Calibration standards were prepared in 30:70 (v/v) methanol/water covering greater than 5 orders of magnitude.

The calibration standards were injected on a SCIEX ExionAE system using a Phenomenex Kinetex C18 column for chromatographic separation. The gradient run was 16 min with mobile phases of 0.05mM ammonium fluoride in water (“A”) and 0.05mM ammonium fluoride in methanol (“B”). The injection volume was 10 µL. The samples were analyzed using the SCIEX 7500+ system with an OptiFlow Pro source operated in electrospray ionization mode. Samples were also run on a QTRAP 6500+ system for sensitivity comparison.

RESULTS:
The SCIEX 7500+ system showed high levels of sensitivity, allowing for the low-level quantitation of almost all steroids at sub-pg levels, reaching as low as 2.5 fg on column for some compounds. Linearity (1/x weighting, r2 > 0.999) was achieved for all steroids across the calibration range. Average accuracy and precision were assessed for 3 replicates and was within acceptable validation requirements with average accuracy of 85-115% and precision of < 15% for three replicate injections at all concentrations.

The sensitivity differences between the 7500+ and 6500+ systems were investigated by calculating the relative gains in peak area and S/N (n=3) for each MRM transition at the LLOQ concentration determined on the QTRAP 6500+ system. Across all transitions, the average area gain was 14-fold and the average increase in S/N was 4.7-fold demonstrating the significant sensitivity increase of the 7500+ system. In addition to sensitivity, the assay reproducibility was assessed by comparing the RSD% at the same concentrations for 3 replicate injections. Overall, the 7500+ system showed better precision (i.e., lower RSD% values). Specifically, the CV% was 2.1% for the 7500+ system as compared to 5. 9% for the QTRAP 6500+ system.

DISCUSSION:
This technical note demonstrates a method for the analysis of 18 steroids in solvent, achieving lower limits of quantitation (LLOQ) as low as 2.5 fg on column on the SCIEX 7500+ system. The advantage of this high sensitivity was explored by comparing the results with the previous generation of MS, a QTRAP 6500+ instrument. The results showed an increase in sensitivity for the SCIEX 7500+ system, shown in significant gains for both peak area, and signal-to-noise (S/N), utilizing optimization of the Q0 dissociation (Q0D) parameter, a feature unique to the improved front-end technology of the SCIEX 7500+ system. This increase in sensitivity would potentially allow for simplified sample preparation, reduced injection volume or a decrease in the detection levels previously achieved in similar workflows.

High-Throughput Vitamin A and E Analysis in Serum to Detect Failing Physiological Functions Quickly Using LDTD-MS/MS Sarah Demers (Presenter) Phytronix Technologies

>> POSTER (PDF)

Poster #29a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Vitamin A and E are essential for maintaining key physiological functions critical to overall health. Monitoring their levels is crucial for clinical diagnostics, identifying nutritional deficiencies, and managing chronic diseases and various health conditions. As fat-soluble vitamins stored in the liver, their serum levels can also serve as indicators of liver health1. High-throughput techniques, such as LDTD-MS/MS, provide a rapid, accurate, and sensitive method for quantifying these vital nutrients, allowing for efficient analysis of large sample sets. For large-scale epidemiological studies, reliable and high-throughput analytical methods are essential to ensure precise data collection.

OBJECTIVES:
In this project, an optimized extraction method is developed and integrated into an automated system for the analysis of vitamins A and E in serum. The extracted vitamins are quantified using Laser Diode Thermal Desorption coupled with tandem mass spectrometry (LDTD-MS/MS), selected for its rapid and efficient analytical capabilities.

METHODS:
Human serum contains varying endogenous levels of vitamins, and several calibration preparations were evaluated, including spiking with bovine serum albumin/NaCl, charcoal-stripped serum matrices, or direct addition to serum. The extraction process involves mixing 50 µL of the matrix with 50 µL of an internal standard solution diluted in ethanol. Next, 200 µL of hexane is added and mixed. A 6 µL aliquot of the upper layer is then spotted onto LazWell96 plates, which are evaporated to dryness before analysis by LDTD-MS/MS.

The mass spectrometer is operated in positive ionization mode, while the LDTD system uses a flow rate of 6 L/min with air as the carrier gas and a laser power ramp from 0 to 45% over 3 seconds.

RESULTS:
All-trans retinol (Vitamin A) and α-tocopherol (Vitamin E) were selected as the target compounds for this analysis. Given that Vitamin A is light-sensitive, the addition of BHT (butylhydroxytoluene) was necessary to prevent degradation of the samples. Various concentrations of BHT, ranging from 0.01% to 1% w/v, were tested in both the stock solution and extraction solvent preparations. A concentration of 0.1% BHT provided optimal protection in both cases. The extraction process was carried out in a light-protected environment.

Calibration curves for α-tocopherol (0.5 to 20 µg/mL) and retinol (0.1 to 4 µg/mL) were prepared in both a bovine serum albumin (BSA) solution (30 mg/mL BSA with 0.9% w/v NaCl) and in charcoal-stripped serum. These matrices, differing significantly from regular serum, produced distinct peak desorption patterns on the LazWell plate surface. To address this, a dosed addition of α-tocopherol and retinol directly in serum was evaluated. The peak area-to-internal standard (IS) ratio was used for signal normalization. The efficiency of these calibration preparations was validated through a cross-validation study.

Preliminary data showed excellent regression correlation coefficients (r > 0.999). Between-run accuracy ranged from 90.6% to 113.0%, and precision was less than 13% CV.

For LDTD-MS/MS analysis, both wet stability (extracted solutions stored at 4°C and -20°C for 12 hours) and dry stability (extracts on the LazWell plate for 1 hour at room temperature) were evaluated, rather than assessing the autosampler's stability. After the specified stability periods, calibration curves were analyzed, with precision for the standards ranging from 3.9% to 14.1% CV and accuracy between 85.8% and 114.5% of the nominal values.

CONCLUSION:
In conclusion, an optimized extraction procedure combined with LDTD-MS/MS analysis was successfully developed for rapid quantification of Vitamin A and E in serum. The method, requiring just 8 seconds per sample, showed high accuracy and precision, with excellent precision and accuracy results. This efficient approach is well-suited for large-scale studies and clinical diagnostics.

Detection of Phosphatidylethanol in Human Whole Blood Leonard Chay (Presenter) SCIEX

Poster #30a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Phosphatidylethanols (PEth) are phospholipids that form when phosphatidylcholine reacts with ethanol in the presence of the enzyme Phospholipase D. These phospholipids have an average circulation half-life of 4 days, making PEth detectable in the blood for weeks after alcohol consumption. This makes PEth a longer-term marker for alcohol intake compared to traditional metabolites such as ethyl sulfate and ethyl glucuronide. PEth 16:0/18:1 is the most abundant form among the homologues and will be the focus of this work.

OBJECTIVE:
The objective of this study is to develop a liquid chromatography tandem mass spectrometric (LC-MS/MS) research assay to evaluate and quantify PEth 16:0/18:1 in human whole blood, with an extraction procedure containing an internal standard.

METHODS:
An aliquot of 100 µl of human whole blood was spiked with 400 µl of isopropyl alcohol (IPA) containing deuterated PEth 16:0/18:1 (PEth-d5) as an internal standard. The sample was thoroughly mixed and centrifuged at 13,500 rpm for 10 minutes. A 200 µl aliquot of the resulting supernatant was transferred to an LC-MS vial for analysis.

Chromatographic separation was performed using a Phenomenex Luna Omega column (30 x 2.1mm, 1.6µm, 00A-4748-AN). Mobile phase A was ammonium formate in water/isopropyl alcohol/acetonitrile (30:10:60, v/v), and mobile phase B was ammonium formate in water/isopropyl alcohol/acetonitrile (1:79:20, v/v). Total analysis time was 3.5 minutes at a flow rate of 450 µl/min. The injection volume used was 10 µl.

Mass spectrometric analysis was performed using a SCIEX Triple Quadrupole 4500 tandem mass spectrometer operating in negative electrospray ionization mode. The compound-dependent parameters were optimized via direct infusion.


RESULTS:
The developed assay demonstrates a quantifiable linear dynamic range from 20-2000 ng/ml in human whole blood. Calibration curves for both the quantifier and qualifier ions of PEth 16:0/18:1 show good linearity using 4 replicates, with a linear regression value (R2) greater than 0.99. The correlation of variation (%CV) among the replicates was below 4%.

CONCLUSION:
The developed assay offers a simple and robust method to detect and quantify PEth 16:0/18:1 in human whole blood across a wide linear dynamic range. The assay provided separation of matrix interference peaks and demonstrated excellent reproducibility.

Tips and Tricks for Developing Tests for the Quantitation of Therapeutic Monoclonal Antibodies Paula Ladwig (Presenter) Mayo Clinic

Poster #31a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Therapeutic monoclonal antibodies have drastically improved the treatment of autoimmune disorders, cancers and other rare conditions. As more therapeutic monoclonal antibodies and their biosimilars have become available and also widely advertised, there has also been an increase in the clinical need for clinicians to determine the effectiveness of the current regimen specifically for example for the therapeutic drug testing for inflammatory bowel disease and ulcerative colitis. Historically, clinical laboratories were struggling to develop quantitative methods for this class of large proteins as traditional methods for large or small molecules were not directly applicable. Over the last 10 years, mass spectrometry has become a sensitive and specific method for the quantitation of these large protein therapeutics.

As mass spectrometry methods have been successful over the last 10 years for therapeutic monoclonal antibodies and laboratories are even moving into the next generation of this testing, clinical laboratory developers have gained insightful experience working through peptide methods along with bringing forth large protein mass spec quantitation to the routine clinical laboratory. As with every new class of analytes and technologies there is a learning curve and growing pains. Sharing of complications, headaches along with best practices can be insightful to others working in the field of proteins and mass spectrometry.

OBJECTIVE(S):
To provide tips, tricks and best practices for successful development, validation and implementation of mass spectrometry methods for the quantitation of therapeutic monoclonal antibodies.

METHODS:
Multiple methods have been utilized for the enrichment of therapeutic monoclonal antibodies from serum to include enrichment methods to enrich for either all immunoglobulins, a class or subclass or for a specific antibody. Detection targets have included either peptide or the light chain from the therapeutic monoclonal antibody of interest. Instrumentation has spanned single or multiplex HPLC systems connected to either a triple quad, orbitrap or time of flight mass spectrometer.

RESULTS:
Discussion points for development include matching the antibody type, class and even subclass along with needed sensitivity and specificity to the enrichment method and possibly even detection method. For instance, infliximab as a chimeric therapeutic monoclonal antibody, has a unique peptide sequence that allows for a less specific enrichment method if using tryptic peptide detection. A method for an IgG4 subclass can take advantage of a subclass specific enrichment method. A humanized or fully human IgG1 therapeutic monoclonal antibody may need an antibody capture to truly allow for the needed sensitivity and specificity.

Other critical discussion points include examples where inaccuracy and imprecision can be lowered using normal human serum as the matrix for standards and controls, the stable isotopic full length internal standard over the use of a surrogate when available or using the pharmacy drug for spiking standards and controls over using another commercially available product.

There is also discussion around the type of instrumentation used for detection. A quadrupole is useful for the detection of tryptic peptides, while high resolution accurate mass instrumentation is needed for detection of intact light chains. An orbitrap can be utilized for a test for a single analyte of interest, but a TOF can allow for detection of panels.

CONCLUSION:
The goal of the session is not to provide method specific details from our published methods, but instead to share the lessons learned as we streamlined our development, validation and implementation pipeline for clinical asks for therapeutic drug monitoring for new therapeutic monoclonal antibodies.

REFERENCES:
1.Willrich MA, Murray DL, Barnidge DR, Ladwig PM, Snyder MR. &ldquo;Quantitation of infliximab using clonotypic peptides and selective reaction monitoring by LC-MS/MS.&rdquo; Int Immunopharmacol. 2015 Sep;28(1):513-20.
2.Ladwig PM, Barnidge DR, Willrich MA. &ldquo;Quantification of the IgG2/4 kappa Monoclonal Therapeutic Eculizumab from Serum Using Isotype Specific Affinity Purification and Microflow LC-ESI-Q-TOF Mass Spectrometry.&rdquo; J Am Soc Mass Spectrom. 2017 May;28(5):811-817.
3.Ladwig PM, Barnidge DR, Willrich MAV. &ldquo;Mass Spectrometry Approaches for Identification and Quantitation of Therapeutic Monoclonal Antibodies in the Clinical Laboratory.&rdquo; Clin Vaccine Immunol. 2017 May 5;24(5)
4.Cradic KW, Ladwig PM, Rivard AL, Katrangi W, Wintgens KF, Willrich MAV. Vedolizumab quantitation using high-resolution accurate mass-mass spectrometry middle-up protein subunit: method validation. Clin Chem Lab Med. 2020 June 25;58(6):864-872.
5.Ladwig PM, Willrich MAV. Ravulizumab, characterization and quantitation of a new C5 inhibitor using isotype specific affinity purification and high-resolution mass spectrometry. J Mass Spectrom Adv Clin Lab. 2021 Aug 12;21:10-18.
6.Ladwig PM, Rivard AL, Barbeln A, Maus A, Murray DL, Snyder MR, Willrich MAV. Infliximab Therapeutic monitoring by tryptic peptide LC-MS/MS method improvements lead to improved accuracy with decreased imprecision and turnaround time. J Mass Spectrom Adv Clin Lab. 2024 Feb 3;32:24-30.

Sensitive and Selective Quantitation of Bile Acids Using Targeted MS2/MS3 on the Stellar MS Kerry Hassell (Presenter) ThermoFisher Scientific

Poster #32a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Bile acids (BA), synthesized from cholesterol in the liver, are crucial for lipid digestion and absorption. In the gastrointestinal tract, primary bile acids are converted by host microbiota into various secondary and conjugated bile acids. Disordered microbiomes can alter the composition and concentration of the bile acid pool, impacting human health. Normally, blood levels of bile acids are low, but they can increase due to disorders such as liver disease. Consequently, identifying and quantifying these compounds has become a significant research focus. We present a highly selective targeted MS2 and MS3 approach for the quantitation of bile acids and their conjugates in serum using the Thermo Scientific™ Stellar™ mass spectrometer.

METHODS:
Unlabeled and stable-isotope-labeled bile acid standards were purchased from Cambridge Isotope Laboratories. These standards were used to prepare dilution series solutions in 50% methanol and generate calibration curves. Bile acids were extracted from the serum of adults with and without autism spectrum disorder (ASD) using 80% methanol, separated on a reverse-phase C18 column, and analyzed using the Thermo Scientific™ Stellar™ mass spectrometer. A scheduled method combining targeted-MS2 and targeted-MS3 scan modes with two activation types, CID and HCD, was employed to quantify these molecules. Data was processed using Thermo Scientific™ TraceFinder™ 5.2 software and Thermo Scientific™ Chromeleon Software.

PRELIMINARY DATA:
The BA standards were utilized to generate calibration curves for absolute quantitation, facilitating reliable ion trap quantitation over a wide dynamic range of 5 to 6 orders of magnitude for the targeted BA. The MSn-based quantitation of the mass analyzer provided the selectivity needed to detect and accurately quantify co-eluting isomers and isobars, while also improving discrimination between analyte signals and matrix interferences.

The new Stellar mass spectrometer enhances quantitative methods by combining different activation types and integrating targeted MS2 and MS3 acquisition modes in a single run. Bile acid standard curves (0.1 to 1000 nM) demonstrated a dynamic range over 5 orders of magnitude and excellent sensitivity in the femtomole range. Enhanced selectivity was achieved with MS3 acquisition, allowing accurate quantitation of co-eluting isomers that MS2 could not separate. This platform also provides valuable structural insights, potentially eliminating the need for extended chromatographic separation. The developed method was successfully employed to quantify metabolites in fecal samples.

CONCLUSION:
High specificity and sensitivity with combined MS2 and MS3 levels for the quantitation of bile acids in blood.

Profiling Cysteine Phosphorylation via Mass Spectrometry Luba Mahbub (Presenter) McGill University

>> POSTER (PDF)

Poster #33a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Cysteine phosphorylation in proteins plays a regulatory role in signaling but remains underrepresented in phosphoproteomics due to its transient and chemically labile nature. Phosphatases of regenerating liver (PRLs) are cystine-based phosphatases that form a stable phosphocysteine intermediate during catalysis, which can accumulate endogenously in cells. All three human PRL isoforms are involved in cell growth and magnesium homeostasis and are frequently associated with cancer metastasis. Although both their catalytic and non-catalytic functions have been implicated in promoting metastasis, the underlying mechanism remains poorly understood. Quantifying phosphocysteine levels in cells and tissues may provide insight into their role in tumor progression.

OBJECTIVES:
The absence of PRL isoform-specific antibodies, combined with the acid- and heat-labile nature of phosphocysteine, has limited its characterization in biological systems. This study aims to establish a derivatization-based mass spectrometry workflow for the quantitative detection of PRL cysteine phosphorylation. A broader goal is to determine whether other cellular proteins also harbor stable cysteine phosphorylation.

METHODS:
Cell lysates or purified protein samples were first denatured, and free cysteine residues were reduced and alkylated to prevent nonspecific labeling. Phosphocysteine was then hydrolyzed by controlled heating, converting it to reactive thiols. The newly exposed thiols were selectively labeled with maleimide-based reagents. Labeled proteins were enriched using affinity purification, followed by tryptic digestion. Phosphocysteine-derived peptides were identified and quantified by mass spectrometry.

RESULTS:
A robust workflow was developed for detecting cysteine phosphorylation in recombinant PRLs under denaturing conditions. Denaturation in guanidine hydrochloride rendered the phosphocysteine intermediate more resistant to premature hydrolysis, enabling its preservation throughout the initial steps of the workflow. Following thermal hydrolysis, selective labeling of phosphocysteine-derived cysteines was achieved using biotin-maleimide. Labeled peptides from recombinant PRL2 expressed in E. coli were reliably detected by both MALDI-TOF and LC-MS/MS. NeutrAvidin-based enrichment of biotinylated proteins significantly enhanced detection sensitivity, allowing confident identification of modified peptides over background. Systematic optimization of reduction, alkylation, and labeling steps minimized nonspecific reactivity and improved recovery of phosphocysteine-derived peptides.

CONCLUSION:
This workflow offers a reliable, MS-compatible approach for the selective detection of cysteine phosphorylation in PRLs and is adaptable to other phosphocysteine-containing proteins. Future work will apply this strategy to mammalian cells and tissues to evaluate phosphocysteine levels in PRLs and explore their contribution to metastatic cancer.

Breath VOC Profiling for Pre-Diabetes and Type 2 Diabetes Using TD-GC-TOF-MS: A South Asian Cross-Sectional Study Trenton Stewart (Presenter) Warwick University

>> POSTER (PDF)

Poster #34a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Type 2 diabetes mellitus (T2DM) is an escalating global health concern, with over 463 million people affected worldwide as of 2019 — a number projected to reach 700 million by 2045. In Nepal, the burden is rising rapidly, with recent estimates suggesting that approximately 10% of the adult population now lives with T2DM. Factors such as urbanization, changing diets, sedentary lifestyles, and limited access to preventive care are contributing to the steep rise in prevalence across South Asia. Current diagnostic strategies—such as fasting blood glucose and HbA1c tests—require blood sampling and are often underutilized in low-resource settings due to infrastructure, cost, and patient discomfort. These methods also fail to detect early metabolic changes before clinical thresholds are crossed.

Exhaled breath contains volatile organic compounds (VOCs) that reflect underlying metabolic processes, and several studies have shown altered VOC profiles in individuals with diabetes. Acetone, for example, was first identified in diabetic breath as early as the 19th century and remains one of the most studied VOCs linked to glucose metabolism. Breath analysis represents a non-invasive, painless, and potentially real-time method of screening for metabolic disease. This study aims to evaluate the utility of breath VOC profiling for detecting pre-diabetes and T2DM in a South Asian cohort, using thermal desorption–gas chromatography–time-of-flight mass spectrometry (TD-GC-TOF-MS). The goal is to support the development of a portable, clinically relevant, and cost-effective diagnostic tool suitable for widespread screening in resource-constrained settings.

METHODS:
This cross-sectional validation study is set to recruit 240 participants from clinical sites in Nepal, including 80 healthy controls, 80 individuals with pre-diabetes, and 80 individuals with T2DM. This study is ensuring statistical power through a sample size calculation with a 95% confidence and accounting for 5% type 1 errors alongside the 50% prevalence, a total of 146 participants would be required. At least 73 diabetic participants and 73 controls would need to be recruited to ensure the power of the sensitivity and specificity of the study. Each participant will provide two exhaled breath samples, one fasting and one postprandial, collected using a BIO-VOC2 breath sampler. For each sampling, five full exhalations will be overlaid onto a single thermal desorption (TD) tube to enhance compound detection. After collection, TD tubes will be sealed and stored in a cooler at 4 °C, then transported under refrigerated conditions to the Biomedical Sensors Laboratory at the University of Warwick (UK) for analysis.

Breath VOC samples were analysed using a TD-GC-TOF-MS system. Chromatographic peaks were deconvoluted and integrated using Chrome Compare™ software. The resulting VOC data were analysed using multivariate statistical analysis and machine learning (ML) techniques to identify compound features capable of differentiating between glycaemic and healthy classifications. ML models included Random Forest, XGboost, and Logistic Regression. Feature selection was based on the Mann–Whitney U test, with Benjamini–Hochberg false discovery rate (FDR-BH) correction applied to control for multiple comparisons.

Ethical approval for this study was granted by the Nepal Health Research Council (NHRC), reference number 3833, and the Manmohan Memorial Medical College Research Committee (MMMC-RC), reference number 1471079180. Written informed consent was obtained from all participants prior to enrolment.

RESULTS:
Preliminary multivariate analysis was conducted on a subset of 132 participants: 51 healthy controls, 35 individuals with pre-diabetes, and 46 individuals with T2DM. These preliminary analyses revealed several VOCs with statistically significant differences between glycaemic classifications. In the Healthy vs pre-diabetic group, Longifolene achieved significance after multiple testing correction (adjusted p = 0.028). In the pre-diabetic vs T2DM group, Nonadecane (adjusted p = 0.017), Octadecane (p = 0.019), and Silanediol, dimethyl- (p = 0.025) also remained significant following FDR correction. While no VOCs reached statistical significance in the Healthy vs Diabetic group after FDR correction, several compounds showed strong combined p-values in other comparisons. Notably, Cyclohexane (corrected p = 0.061) in the Healthy vs pre-diabetic group and 2-Pentanone (corrected p = 0.073) in the pre-diabetic vs T2DM group demonstrated significant raw and combined p-values. As these are preliminary results, it is expected further discriminatory markers may be identified as the further data is collected.

CONCLUSION:
Preliminary results from this validation study indicate that breath VOC profiling using TD-GC-TOF-MS can differentiate between glycaemic states in a South Asian population. Several compounds, including Longifolene, Nonadecane, and Octadecane, showed statistically significant differentiation between classification groups. These preliminary findings support the potential of breath analysis as a non-invasive, low-burden tool for early identification of individuals at risk of pre-diabetes and type 2 diabetes, particularly in resource-limited settings. As data collection continues, it is anticipated that further discriminatory markers will be identified, strengthening the clinical utility of this approach.

Characterizing the Structural Heterogeneity of Polysorbates Using High Resolution Ion Mobility-Mass Spectrometry (HRIM-MS) Strategies Kyle Lira (Presenter) Vanderbilt University

Poster #35a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Polysorbates (PS) are synthetic surfactants commonly utilized commercially to increase the solubility and bioavailability of molecules of interest in cosmetics, foods, and pharmaceuticals. Commercial PS products (e.g., Tween) consist of heterogeneous mixtures of numerous species with similar structures; consequently, PS are analytically challenging to characterize. Typically, MS techniques are utilized in PS analysis, yet these cannot resolve isomeric species which are prevalent in PS mixtures. Using ion mobility (IM)—a gas phase separation technique that discriminates ions by size, shape and charge—structurally-similar compounds, including isomers, can be resolved. From IM experiments, a reproducible chemical descriptor known as a collision cross section (CCS) can be determined for individual species. Here, we evaluate the capability of various high resolution ion mobility (HRIM) techniques for the characterization of two common polysorbates (PS-20 and PS-80). Additionally, IM profiles and CCSs are compared to measurements obtained on a drift tube ion mobility (DTIMS) platform which is capable of measuring CCS values from first principles.

OBJECTIVES:
The goal of this work is to integrate analytical data from various HRIM techniques towards a unified ion mobility resource that will support pharmaceutical and life sciences research. To this end, each IM technique is leveraged for specific analytical strengths.

METHODS:
Data for PS-20 and PS-80 (USP) were acquired in positive mode via direct infusion (10 µg/mL) on a DTIMS (6560 IM-QTOF, Agilent Technologies), a traveling wave structures for lossless ion manipulation IM (TWSLIM, MOBIE, MOBILion Systems) coupled to MS (6546 QTOF, Agilent), a trapped IM-MS (timsTOF Pro 2, Bruker Corporation), and a cyclic IM-MS (cIM, SELECT SERIES, Waters Corporation). DTIMS measurements were obtained in 4-bit multiplexing mode to enable high resolution demultiplexing (HRdm, Agilent). Arrival times were determined using vendor software (HRIM Data Processor, MOBILion; IM-MS Browser, Agilent; Compass DataAnalysis, Bruker; and MassLynx, Waters). CCS values were determined using previously-documented calibration procedures incorporating a MS tuning mixture (ESI-L, Agilent), with CCS values from HRIM techniques validated against drift tube measurements (single-field calibration).

RESULTS:
A target list obtained from DTIMS-MS analysis was used to support PS identifications from HRIM analyses. First, the MS distributions for each PS structural series were extracted from the IM-MS datasets, and both the weight- and number-averaged mass distributions (Mw, Mn) were determined for each oligomer series. It was found that Mw and Mn were comparable, providing a basis for measurement alignment and standardization. Next, IM-MS conformational space correlations—which provide information on structural series, number of fatty acids, type of adduct, and charge state—were found to be generally consistent between DTIMS-MS and HRIM-MS analysis once aligned in CCS space. While most IM profiles observed using DTIMS (Rp ~60) do not exhibit multiple peak features, HRIM analysis revealed that some PS species were candidates for isomeric separation. For further validation, TIMS and cIM were utilized in a targeted mode to optimize IM resolution and confirm the presence of isomers. Mass accuracy values for PS identifications typically are better than ±5.0 ppm and are strongly correlated to ion abundance.

CONCLUSION OR DISCUSSION:
Preliminary results suggest that the previous DTIMS-MS measurements are reproducible across multiple IM platforms, with some features containing multiple isomers under a single IM profile. For most PS species, IM profiles are consistently comprised of single peaks even under high IM resolution, indicating that CCS values obtained from drift tube IM are also applicable for HRIM, which allows direct comparison of various IM techniques via CCS alignment.

Method Development for Rapid Evaporative Ionization Mass Spectrometry Guided Tissue Investigation During Intraoperative Head and Neck Surgery Gabriel Stefan Horkovics-Kováts (Presenter) University of Regensburg

Poster #36a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Head and neck squamous cell carcinoma (HNSCC) is the sixth most common malignancy with 600,000 new cases each year. Patients with HNSCC have a poor prognosis, with five-year overall survival rates of 40-50%. As with most tumour surgeries, the key objective is to remove the tumour completely, while preserving as much functional tissue as possible to maintain the patient's quality of life. In the field of HNSCC surgeries, this is particularly important for tumours of the tongue, where extensive reconstruction using other donor parts of the body and additional operations could be avoided. The ambient Rapid Evaporative Ionization Mass Spectrometry (REIMS) technique has the potential to provide real-time, tissue-specific feedback by analysing the ablated fume during monopolar or CO2 laser surgery, which can enable a more precise tumour resection. Therefore, we present the method development for intraoperative HNSCC surgeries using REIMS, to achieve best possible mass spectrometry data for multivariate statistical model building and potential real-time tissue classification.

METHODS:
The tissue-specific fumes produced by the ablation of electrosurgical monopolar or surgical CO2 lasers are collected by an evacuation tube connected to the REIMS source of a Xevo G2-XS QTof mass spectrometer (Waters), which was specially designed for the operating theatre. Here, the sampling position, tube length, tube thickness, suction strength and dilution gas flow are tested regarding best possible signal-to-noise ratio in the mass spectrometry data acquisition during HNSCC surgery. Especially a robust and reproducible gathering of the surgical fumes in deep regions of the human throat can be challenging. Therefore, appropriate sampling options must be tested or developed that collect the fume as effectively as possible without obstructing the surgeon's view and without affecting the surgeon's narrow operating space through the patient's mouth. For this purpose, both a set of commercially available electrosurgical instruments with integrated suction tubes and conventional electrosurgical instruments with attached suction tubes of different lengths and thicknesses are tested. After this hardware-focused optimization phase, tissues surrounding the tumour, such as mucosa and muscle tissue, are sampled in vivo and the resected tumours are sampled ex vivo for later creation of multivariate statistical models under supervised annotation by the surgeons to enable real-time classification possibilities.

RESULTS:
Our preliminary results show that a short suction tube length with narrow tube thickness, harmonized suction strength and dilution gas flow, and an immediate suction point above the ablation site leads to the best possible signal-to-noise ratios in the obtained mass spectra. In vivo sampling depends on many environmental factors of the tissue, such as active blood flow, possible bleeding or varying energy distribution of the surgical instruments through the patient. These factors can often induce poor-quality signal or mixed spectra leading to the necessity of extensive data postprocessing and sample annotation challenges. In contrast, ex vivo sampling of tumour resections produce spectra with a high signal-to-noise ratio, which are perfectly suited for the creation of multivariate statistical models.

DISCUSSION:
HNSCC operations present many challenges for obtaining sufficient fume for REIMS analysis due to the surgical environment, showing the need of an optimized hardware setup to reach high quality signal-to-noise ratios from sampled tissues. After adequate model building from in and ex vivo data, the real-time information provided by REIMS during electrosurgical or CO2 laser dissection has the potential to improve intra-operative decision making, leading to more precise tumour removal for a variety of HNSCC tumours.

Development/Validation of a Quantitative Ultra-Performance Liquid Chromatography Quadrupole Time-of-Flight (UPLC-QTof) Method for Urine Organic Acid Analysis Yi Xiao (Presenter) Children’s Hospital Los Angeles

Poster #37a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Urine organic acid (UOA) analysis is essential for diagnosing inborn errors of metabolism (IEMs). Traditionally, UOA analysis is performed with gas chromatography-mass spectrometry (GC-MS) and requires time-consuming sample preparation procedures including liquid-liquid extraction and derivatization. There is an opportunity to perform UOA analysis with liquid chromatography-mass spectrometry (LC-MS) due to the rapid development in the past few years. We describe the development and validation of a quantitative ultra-performance liquid chromatography quadrupole time-of-flight (UPLC-QTof) method for UOA analysis with a “dilute-and-shoot” approach.

METHODS:
Urine specimens were diluted to normalize creatinine concentrations to 1 mmol/L. 20 µL of each urine specimen (diluted), calibrator, or quality control (QC) material was mixed with 400 µL of mobile phase A (0.05% formic acid in water) and a mixture of isotope-labeled internal standards. After centrifugation, 10 µL of the supernatant was analyzed using a Xevo G3 QTof mass spectrometer (Waters) with a ACQUITYTM Premier HSS T3 1.8 µm VanGuardTM FIT 2.1 x 150 mm column (Waters). Data collection was performed with negative electrospray ionization (ESI) mode using the MSE method to produce fragment ions when applicable. Repeatability, reproducibility, and carryover were assessed using the QC materials. The analytical measuring range (AMR) was assessed using synthetic urine spiked with increasing concentrations of each organic acid. Accuracy was assessed by method comparison with the UOA test performed by the traditional GC-MS method and by spike-recovery study using a pooled urine specimen. Matrix effect was evaluated with matrix dilution study. Strategies to obtain linear rather than quadratic calibration curves were also explored.

RESULTS:
An optimized UPLC method was used to enable high-resolution separation of selected UOAs (N = 29) and isomers. Total analytical time was 20 min per injection. Both linear and quadratic regressions were used to build the calibration curves. AMR and correlation coefficients of a few representative UOAs were: orotic acid (3.4 to 214.2 mmol/mol creatinine, R^2 = 0.99, linear regression); 2-methylcitric acid (4 to 189 mmol/mol creatinine, R^2 = 0.99, linear regression); 3-methylcrotonylglycine (0.3 to 18.0 mmol/mol creatinine, R^2 = 0.99, linear regression). Repeatability and reproducibility were mostly <=10% CV and no carryover was observed. Spike-recovery study demonstrated recoveries between 80% and 120%, and method comparison study demonstrated no discrepancies.

CONCLUSION:
We have developed and validated a novel UPLC-QTof method for UOA analysis to support the diagnosis of IEMs with acceptable analytical and clinical performances. Compared with the traditional GC-MS method, the UPLC-QTof method requires a very small specimen volume and does not require laborious and time-consuming sample preparation steps. Continued optimization of the method will be pursued to measure more UOAs to support the diagnosis of a broader range of IEMs.

Development of an LC-DMS-MS/MS Assay for the Quantification of Cortisol and Cortisone in Segmental Human Hair Kayla Moehnke (Presenter) Mayo Clinic

Poster #38a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Cushing’s syndrome (CS) is typically diagnosed using point-in-time assessments of cortisol levels, including salivary cortisol, serum cortisol following dexamethasone suppression, and 24-hour urinary free cortisol tests. While these methods are effective for many patients, they can be less reliable in a subset of individuals due to variability in cortisol secretion. In these cases, cortisol levels may fluctuate significantly, which can lead to misdiagnosis if testing occurs during a period of normal or suppressed cortisol output. This is particularly relevant in patients with cyclical Cushing’s syndrome (cCS), a form of the disease characterized by alternating periods of hypercortisolism and normal cortisol levels. Additionally, up to 25% of patients with Cushing’s disease (CD)—a pituitary-driven form of CS—may show normal urinary cortisol levels at certain times, further complicating diagnosis. These fluctuations, both clinical and biochemical, underscore the importance of timing in diagnostic testing and highlight the need for repeated or longitudinal assessments in suspected cases of cyclical CS.

Hair analysis offers a unique retrospective matrix for assessing long-term cortisol production, with the potential to detect hypercortisolism for many months, depending on length of hair. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) has emerged as a gold standard for steroid quantification due to its high specificity and sensitivity. This study aimed to develop and validate a robust LC-MS/MS assay for simultaneous quantification of cortisol and cortisone in segmented human scalp hair, enabling retrospective profiling of glucocorticoid exposure in patients suspected of cCS.

METHODS:
Analytical development focused on optimizing sample extraction and chromatographic performance. The established sample preparation protocol began with collecting a pencil-width bundle of hair strands, cut as close to the scalp as possible from the posterior vertex region. Hair strands were washed sequentially with 100% isopropanol and water, then dried for 1 hour at 70 °C. Once dried, the hair was cut into 1 cm segments, with each segment representing a distinct period of hair growth. These segments were maintained and processed separately to preserve the temporal resolution of the sample.

Mechanical lysing was performed using stainless steel beads under wet conditions (1 mL methanol + 1mcg/mL estriol), with parameters such as number of beads, shaking speed, and duration systematically optimized to ensure complete disruption of the hair matrix. Comparative studies of dry versus wet lysing were conducted to assess their impact on analyte release while reducing sample loss caused by electrostatic dispersion of hair strands. Methanol volumes ranging from 1 to 5 mL and sample weights ranging from 10 to 50 mg were tested to determine optimal extraction efficiency. The final step, prior to sample purification, included an incubation at 50°C of minced sample in methanol, which was assessed from no incubation up to 48 hours.

Solid-phase extraction (SPE) was optimized in a 96-well plate format, with sorbent chemistries and elution conditions evaluated to maximize analyte recovery and minimize matrix effects. The adopted LC method employed a C18 analytical column (4.6 × 50 mm, 3.5 µm) and an HTL C18 XL loading column (0.5 × 50 mm). The mobile phases included: eluting mobile phase A (100% water), loading mobile phase B (5% methanol), mobile phase B (100% methanol), mobile phase C (45:45:10 acetonitrile:isopropanol:acetone), and mobile phase D (0.1% ammonium hydroxide in water). Final extracts were analyzed using LC-MS/MS with and without differential ion mobility (DMS) device installed.

RESULTS:
The finalized protocol utilized wet lysing of ~10 mg hair samples in 1 mL methanol, incubation of 18 hours, subsequent centrifugation to aspirate 500 mcL of supernatant, followed by SPE cleanup in a 96-well plate format. This configuration supported high-throughput processing with consistent analyte recovery. The LC-DMS-MS/MS method demonstrated excellent chromatographic resolution, low limits of detection, and acceptable intra- and inter-assay precision. The assay was suitable for applications requiring sensitive and specific quantification of hair glucocorticoids.

CONCLUSION:
This study presents an optimized LC-DMS-MS/MS assay for the simultaneous quantification of cortisol and cortisone in human hair, offering a powerful tool for the retrospective assessment of glucocorticoid exposure. Key technical challenges addressed included losses of hair sample due to electrostatic dispersion during sample handling, the need for precise weighing of low-mass samples, the impact of lysing conditions on extraction efficiency, and optimal sample purification conditions. Additional considerations such as cosmetic hair treatments, shampoo frequency, and environmental exposures must be accounted for in study design, as they may influence analyte levels and clinical interpretation.

The method demonstrates high sensitivity, specificity, and reproducibility, with minimal matrix interference. Segmental hair analysis shows potential for detecting cyclical hormone patterns, which may offer valuable diagnostic insight in patients with suspected cyclical Cushing’s syndrome. While preliminary findings are encouraging, further clinical validation is needed to confirm the assay’s diagnostic performance and utility in broader patient populations. This method represents a meaningful advancement in the non-invasive monitoring of hypothalamic-pituitary-adrenal axis disorders and could contribute to improved diagnostic accuracy and patient outcomes in CCS with continued study.

Global Metabolomic and Lipidomic Remodeling Induced by Ketogenic Diet Sang-Guk Lee (Presenter) Yonsei University College of Medicine

Poster #39a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
The ketogenic diet (KD) has been used as a treatment for drug-resistant epilepsy and studied for its therapeutic potential in metabolic disorders, such as obesity, type 2 diabetes, and inflammation, and neurodegenerative diseases. However, its restrictive nature is challenging for long-term adherence. A comprehensive understanding of KD-induced metabolic effects is essential for developing pharmacological alternatives that mimic its therapeutic effects.

OBJECTIVES:
To elucidate the KD-induced metabolic and lipidomic remodeling and their biological significance.

METHODS:
Fifteen healthy adults (aged 24–38 years) underwent a 3-day KD intervention, with a macronutrient composition of 75% fat, 20% protein, and 5% carbohydrate, while maintaining isocaloric intake. Serum samples were collected before and after the KD, and analyzed using untargeted metabolomics and lipidomics approaches.

Serum samples were prepared for ultra-performance liquid chromatography-Orbitrap mass spectrometry (UPLC-Orbitrap-MS) analysis. The prepared samples were analyzed using an Orbitrap Exploris 120 Mass spectrometer coupled with a Vanquish Flex UHPLC system (Thermo Fisher Scientific, San Jose, CA, USA). Chromatographic separations for both metabolomics and lipidomics were performed using an ACQUITY UPLC BEH C18 column (2.1 × 100 mm, 1.7 µm, Waters, Milford, MA, USA). MS analyses for both metabolomics and lipidomics were conducted in full-scan mode (m/z 80-1200) with a resolution of 120,000, and data-dependent MS/MS acquisition in positive and negative ionization modes.

RESULTS:
KD increased acylcarnitine species, acetyl-L-carnitine, and β-hydroxybutyrate, and decreased lactate levels, indicating enhanced fatty acid oxidation and ketogenesis. The gamma-aminobutyric acid/glutamate ratio also increased, supporting the KD role in neuroprotection. Lipidomic analysis showed that KD reduced total triglyceride, and induced long carbon chains and high degrees of unsaturation, suggesting a selective metabolic adaptation favoring sustained energy supply. Phospholipid remodeling was observed, including increased plasmalogens, an elevated phosphatidylcholine/phosphatidylethanolamine ratio, while lysophosphatidylcholine/phosphatidylcholine and lysophosphatidylethanolamine/ phosphatidylethanolamine ratios were downregulated, reflecting improved neuronal membrane integrity and reduced inflammatory potential. Sphingolipid metabolism was altered, with increased sphingomyelin and hexosylceramide, indicating neuroprotection-associated metabolic changes. Phosphatidylethanolamine 36:5 (16:0/20:5) was closely correlated with the quantitative insulin sensitivity check index, insulin levels, and HOMA-IR, highlighting its potential as a metabolic marker for improved insulin sensitivity.

CONCLUSION:
Key metabolic pathways underlying energy metabolism remodeling in response to KD were elucidated and novel insights into lipid metabolism were established, revealing neuroprotective and anti-inflammatory effects. The findings can be used for the development of alternative therapeutic strategies that mimic the KD beneficial effects.

How the Concentration of Bovine Serum Albumin in the Standard Matrix Affects the Bioavailable Testosterone Assay Jennifer Kemp (Presenter) Mayo Clinic

Poster #40a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Our historical bioavailable testosterone assay involved differential precipitation of sex hormone-binding globulin (SHBG) followed by scintillation counting, which required outdated equipment and radioactive-labeled testosterone. To address these limitations, we transitioned to liquid chromatography-tandem mass spectrometry (LC-MS/MS) and eliminated scintillation counting. However, finding a suitable standard matrix was challenging.

OBJECTIVE:
Due to the extraction method, binding proteins, and non-specific adsorption, it was essential to have a standard curve in a matrix that could be extracted and remain stable over time. After testing various matrices, we found bovine serum albumin (BSA) to be the most consistent and stable. The concentration of BSA required careful optimization due to non-specific adsorption of testosterone to various vessels and the presence of small amounts of testosterone present in BSA. We tested 0.1%, 0.25%, 0.5%, and 1% BSA in phosphate-buffered saline to determine the optimal concentration for the standard matrix.

METHODS:
The method is based on the differential precipitation of SHBG by ammonium sulfate. SHBG bound testosterone is precipitated in patient sample using saturated ammonium sulfate, leaving the bioavailable testosterone in the supernatant. After addition of 13C labelled testosterone internal standard (IS), the bioavailable testosterone and IS are extracted from the supernatant using isopropanol. The bioavailable testosterone and IS are then derivatized and analyzed by LC-MS/MS on a Sciex 7500 coupled to a ThermoFisher Scientific TLX-4 equipped with Vanquish pumps. Samples were first loaded onto a Phenomenex Cartridge Security Guard (C12, 4x2mm) and then separated using a Waters XBridge (C18, 2.1x50mm) analytical column.

RESULTS:
The bioavailable testosterone concentration varied depending on the BSA concentration in the standard matrix. Standard curves were prepared utilizing the following testosterone concentrations: 0.5, 1, 2, 10, 50, 200, 500 ng/dL, while varying the BSA concentration. The results for each calibrator at each BSA concentration were as follows when ran as unknowns against the chosen 0.25% BSA matrix: 0.1% BSA: 0.37, 0.7, 1.7, 8.4, 42.2, 183, 474 ng/dL; 0.25% BSA: 0.51, 1.0, 1.9, 9.7, 50.2, 200, 532 ng/dL; 0.5% BSA: 0.58, 1.14, 2.12, 9.4, 46.5, 197, 487 ng/dL; 1.0% BSA: 0.84, 1.3, 2.2, 10.6, 53.5, 204, 478 ng/dL. As the % BSA increased, the peak intensity in the blank for each BSA concentration also rose. The resulting blank peak intensities at each BSA concentration were as follows: 0.1% BSA: 598; 0.25% BSA: 644; 0.5 % BSA: 2090; 1.0% BSA: 3020. Based on these results, 0.25% BSA was chosen as the optimal level for minimizing testosterone contribution while preventing non-specific adsorption. The slopes obtained when comparing patient results to the optimally chosen level (0.25% BSA) were: 0.1% BSA: 1.074; 0.5% BSA: 1.028; 1.0% BSA: 0.959.

CONCLUSION:
The concentration of BSA in the standard curve matrix significantly affects the bioavailable testosterone concentration in patient samples. We found that 0.25% BSA was sufficient to prevent non-specific adsorption without contributing to the testosterone concentration.

Leveraging Machine Intelligence for Rapid, Broad Coverage, Absolute Quantification of Metabolites in Biological Samples Jack Howland (Presenter) Matterworks

Poster #41a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Liquid chromatography–mass spectrometry remains a foundational technology across the life sciences, pharma R&D, and environmental research. Rapid acquisition and conversion of mass spectra into actionable data is highly desirable in both discovery and in routine workflows. Traditionally, researchers have had to choose between onerous targeted analyses—measuring the absolute concentration of a small list of known compounds— and untargeted, broad coverage analyses to identify, but not quantify, as many known and unknown compounds as possible. In 2022 we released PyxisTM, a deep learning-based tool capable of inferring the analyte concentrations directly from raw mass spectra with a standardized HILIC method. In this work, we expand Pyxis’ capabilities to a broader spectrum of small molecules and metabolic pathways, partially facilitated by the extension of the model to incorporate data acquired using a newly developed Reversed Phase liquid chromatography method.

To enable direct machine ingestion of raw instrument data, we developed Large Spectral Models (LSMs) via self-supervised training on massive quantities of unlabeled raw data aggregated across many downstream applications, instrument platforms, and configurations. We adapt our LSMs to specific prediction problems by fine-tuning on relatively small high-quality application-specific datasets.

METHODS:
To benchmark Pyxis’ latest model, we selected the NIST Standard Reference Material (SRM) 1950 (Metabolites in Frozen Human Plasma), and the Reference Material (RM) 8231 (Frozen Human Plasma Suite for Metabolomics), representing a total of four different phenotypes. Metabolites were extracted from these biological samples by precipitating proteins with an 80% organic solution. For each sample type, multiple different sample to solvent ratios were used to create a dilution series. The initial analysis of SRM 1950 specifically was prepared by two different operators, thus generating two separate batches. The extracts were mixed with StandardCandles™, Pyxis’ analyte and matrix independent calibrators. Four microliters from each sample extract were analyzed using a Transcend LX-2 multichannel system coupled to an Orbitrap Exploris 120 mass spectrometer (both Thermo Fisher Scientific™). Metabolites were extracted from biological matrices by precipitating proteins with an organic solution with StandardCandlesTM, Pyxis’ analyte and matrix independent calibrators. HILIC-based chromatographic separation was performed with the Waters™ Atlantis Premier BEH Z-HILIC column (2.5 mm, 2.1 x 50 mm), and a mobile phase consisting of 20 mM ammonium carbonate with 0.25% (v/v) ammonium hydroxide (pH=9.6) (solvent A), and acetonitrile (solvent B). High resolution spectra were acquired for 6.7 minutes in full scan and polarity switching mode. For RP analyses, we recorded MS1 data for 7 minutes in polarity switching mode and used the ACQUITY HSS T3 UPLC analytical (1.8 mm, 2.1 x 50 mm) and guard columns (Waters). The RP mobile phase consisted of 0.2% formic acid in water (solvent A) and 0.1% formic acid in methanol (solvent B). The raw instrument files were analyzed using Pyxis to identify and infer the concentration of metabolites in the biological samples.

RESULTS:
Sample extracts from distinct phenotypes were analyzed using Pyxis. Briefly, the SRM 1950 consists of pooled frozen human plasma from healthy donors, and the RM 8231 is a suite of pooled plasma from three distinct phenotypes, diabetic, hypertriglyceridemic, and young African Americans (20-25 years of age).

The predicted concentrations in the widely used NIST Human Plasma Reference Material (SRM 1950), were benchmarked against recently published data.The reference dataset generated by the Wishart lab provides an extensive and quantitative characterization of the human reference plasma, with over a thousand high-confidence values for the scientific community to easily benchmark their targeted and untargeted metabolomics and lipidomics analyses. The median R-squared of the 50 most abundant analytes was 0.95 and the slope was 1.24, which demonstrates Pyxis’ ability to infer analyte concentrations with accuracy. Importantly, Pyxis does not require the use of stable isotope labeled internal standards, calibration curves, and the associated extensive sample preparation steps.

Upon validating Pyxis-inferred concentrations, we sought to make a comparison between different sub-population cohorts in the sample set: healthy population, young African-American, diabetic, and hyperglyceridemic. Using both HILIC and RP column chemistries, we quantified over 1000 different analytes across multiple classes—namely, endogenous metabolites, natural products, products of the gut microbiome metabolism, and drugs—to identify phenotype-specific markers. The metabolic profiles inferred by our model enabled the accurate segregation of plasma samples from the different phenotypes.

CONCLUSION:

This work showcases the expanded applicability of Pyxis to biological samples using both RP and HILIC LC-MS methods. We demonstrate Pyxis’ ability to discern changes in biological samples and were able to replicate consensus concentrations in standard reference materials. Finally, we achieved quantitative separation of phenotypes with, not just analytical accuracy, but also with significant efficiency when compared to traditional workflows.

Identification of New Urine Glycan Biomarkers for Lupus Nephritis Using Spatial N-Glycomic Profiling of Biopsy Tissues Linked With Patient Matched Biofluids Richard Drake (Presenter) Medical University of South Carolina

Poster #42a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Systemic Lupus Erythematosus (SLE) is a chronic autoimmune disease primarily affecting women of childbearing age. Approximately 50% of SLE patients develop Lupus Nephritis (LN), a severe kidney complication linked to increased morbidity and mortality. Current urine biomarkers, like the urine protein-to-creatinine ratio (UPr:Ucr), lack specificity for glomerular damage, often requiring repeated biopsies to monitor disease progression, highlighting the need for sensitive, non-invasive alternatives. Our previous studies in urine identified elevated glycosphingolipids and an altered N-glycome as potential LN markers.

OBJECTIVES:
We hypothesize that glycosylation changes precede irreversible kidney damage detectable by current urine markers. Using N-glycomic profiling of LN patient matched tissues, urine and serum by matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI), the goal was to identify LN-specific urinary glycan biomarkers, providing insights into renal pathology and potentially be predictive of treatment response.

METHODS:
We analyzed patient-matched urine, serum, and FFPE kidney biopsy samples from fourteen LN patients to characterize N-glycan composition using established MALDI-MSI workflows1-4. Analysis of healthy control samples for renal biopsies, urine and serum were available for reference. FFPE kidney biopsies were dewaxed, antigen-retrieved (citraconic buffer [pH 3.0], 95°C, 30 min), and prepared for analysis. Urine (1 µL) and serum (1µL) were suspended in 2µL sodium bicarbonate (pH 8.0), and 1 µL was spotted in duplicate onto amine-reactive slides. Samples were treated with a molecular coating of PNGaseF PRIME (0.1 µg/µL) via an HTX M5 Sprayer, followed by α-cyano-4-hydroxycinnamic acid matrix coating using an HTX Sublimate apparatus. N-glycan imaging was performed using timsTOF fleX MALDI-QTOF at 10 micron spatial resolution for tissue, with spectra processed in SCiLS Lab and annotated using an in-house N-glycan database based on accurate mass determinations.

RESULTS:
A key for obtaining high resolution MALDI-MSI spatial data for the LN kidney biopsies was use of matrix sublimation via an HTX Sublimate, which minimized N-glycan delocalization. This approach enabled the identification of distinct glycosylation signatures associated with key renal structures, including the glomerulus and distal/proximal convoluted tubules. In healthy control tissues, glomeruli have distinct tri- and tetra-antennary N-glycans that are also multiply sialylated. Proximal and distal convoluted tubules have abundant bisecting bi-,tri- and tetra-antennary structures with 1-4 fucoses, and the antennae number progressively increases along the length of the tubule. Analysis of the fourteen patient-matched urine, serum, and biopsy samples resulted in a peak list of 78 N-glycans present in each sample. Overall, there was a strong correlation between the urinary glycomic signatures and those associated with glomeruli histology. Thirteen N-glycans were elevated in both biopsy and urine samples and minimally detected in serum, This suggests that the associated glomerular damage for LN can be detected in the patient urines as reflective of glomerular specific N-glycans.

DISCUSSION:
Overall, having determined the uniqueness of the spatial N-glycome in normal healthy kidney glomeruli and distal/proximal tubules allows direct assessment of the LN-associated glycome and underlying histopathology changes that could provide unique biomarker targets. A urine-based glycomic assay in the LN setting could provide a companion biomarker assay for monitoring disease progression and treatment responses.

REFERENCES:
1. McDowell CT, et al. (2021) Applications and continued evolution of glycan imaging mass spectrometry. Mass Spectrom Rev. 42(2):674-705.
2. Drake RR, et al. (2018) In Situ Imaging of N-Glycans by MALDI Imaging Mass Spectrometry of Fresh or Formalin-Fixed Paraffin-Embedded Tissue. Curr Protoc Protein Sci. 94(1):e68.
3. Blaschke CRK, et al. (2020) Rapid N-Glycan Profiling of Serum and Plasma by a Novel Slide-Based Imaging Mass Spectrometry Workflow. J Am Soc Mass Spectrom. 31(12):2511-2520.
4. Blaschke CRK, et al. (2021) Direct N-Glycosylation Profiling of Urine and Prostatic Fluid Glycoproteins and Extracellular Vesicles. Front Chem. 9:734280.

Reducing False-Positive Phosphatidylethanol (PEth) Quantification in Alcohol Testing via Isobaric Lipidome Analysis and Interference Correction Ching-Hua Lee (Presenter) National Taiwan University

>> POSTER (PDF)

Poster #43a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Phosphatidylethanol (PEth) is a widely used biomarker for alcohol consumption in clinical and forensic settings. A consensus establishes that a blood PEth(16:0/18:1) concentration >20 ng/mL strongly indicates alcohol use. However, studies show that single drinking events with blood alcohol concentrations (BAC) of 0.3–0.5 g/kg or daily consumption of ~16 g ethanol for three months typically do not result in median PEth levels exceeding 20 ng/mL in collected post-drinking. For stricter monitoring of alcohol consumption in patients with alcohol-related liver disease (ALD) or those awaiting liver transplantation – where stringent alcohol control is critical - lowering the cutoff for PEth using LC-MS/MS is necessary. However, reducing the cutoff often increases false-positive results, potentially causing patient conflicts or delaying surgeries. Given that PEth is biosynthesized only in the presence of ethanol, we hypothesize that co-eluting phospholipid isobars in commonly used LC-MS/MS configurations contribute to false positives and inaccurate quantification. In this study, we apply a lipidomics approach to identify interference sources and optimize PEth analysis strategy to lower the cutoff concentration without compromising specificity.

METHODS:
Fifteen participants were enrolled: six with no alcohol consumption history in recent months and nine who self-reported past drinking habits. Three pooled samples were analyzed: (1) pool blank (n=6), (2) pool drinker (n=9), and (3) pool-QC (1:1 mix of pools 1 and 2). Lipidomics analysis of whole blood was performed using an Agilent 6545XT QTOF with iterative data-dependent analysis (DDA). Agilent 6460 and 6495 QQQ instruments, operating in different scan modes, identified interference sources to complement QTOF data. Multiple reaction monitoring (MRM) was used to develop quantitative methods for PEth homologues and interfering lipids. Whole blood samples (50 μL) were extracted using a modified Folch method. Lipid separation was performed using an Agilent Eclipse Plus C18 RRHD and a Phenomenex Kinetex C8 column. Mobile phase and gradient settings were evaluated to investigate relative retention time (RT) shifts between PEth and lipid isobars. PEth identification criteria included: (1) mass accuracy <20 ppm by HRMS and (2) retention time alignment with PEth(16:0/18:1) synthetic standard and equivalent carbon number (ECN) estimation.

RESULTS:
We identified 11 PEth homologues and their interfering lipid isobars, which follow specific co-elution patterns: PEth(X:Y) co-elutes with PS(X+2:Y) and PA(X+2:Y+1), where X and Y represent the total carbon number and double bonds on the acyl chain, respectively. For instance, PEth(16:0/18:1) (m/z 701) is affected by in-source fragmentation (ISF) of PS(18:0_18:1) (m/z 788→701) and PA(36:2) isomers (m/z 699), including PA(18:1_18:1), PA(18:0_18:2), and PA(16:0_20:3). These interferences persist across various LC settings and are influenced by skimmer voltage and elution conditions.

To evaluate the impact of ISF from PS(18:0_18:1) on PEth(16:0/18:1) quantification, we used an the optimal skimmer voltage (170–190 V), determined with a synthetic standard. This resulted in ~3% ISF, leading to overestimation of MRM 701→281 to 40 ppb in pool blank samples under LC conditions with adequate separation (RT 5.3 min), demonstrating that ISF causes significant false-positive quantification.

Further evaluate the interferences other than ISF, we used an Agilent 6495 QQQ with an ion funnel to minimum ISF (< 0.5%) from PS(18:0_18:1). Pool blank spiked with 20 ppb PEth(16:0/18:1) yielded quantification result of 22 ppb (MRM 701→255) and 25 ppb (MRM 701→281) with adequate separation. Based on co-elution patterns and quantification results, we selected MRM 701→255 as the final quantifier to minimize interferences. However, high-throughput LC settings (RT 1.8 min) increased the quantification result to 24.7 ppb for MRM 701→255 due to compressed lipid elution times, causing co-elution of FA(16:0) fragments from multiple sources.

Further investigating the interferences suggested that lipids generating MRM 700→255 contribute approximately 24% of the [M+1] isotope to MRM 701→255. By mathematically subtracting isotopic interference yielded a final concentration with > 90 % accuracy. The optimized PEth quantification strategy - MRM 701→255 transition with isotope correction - enabled cutoff level to be lowered to 5 ppb with high specificity.

CONCLUSION:
Using a QTOF iterative-DDA lipidomics approach combined with QQQ instruments and various scan modes, we identified interference sources for PEth(16:0/18:1) analysis and developed an analytical strategy to minimize interference and improve quantification accuracy. These adjustments enable a lower PEth cutoff without compromising specificity, enhancing the reliability of alcohol consumption monitoring for ALD and liver transplant patients.

Urinary Metabolomics to Identify Diagnostic Biomarkers of Delirium: A Pilot Study Zaineb Hamoodi (Presenter) McMaster University

>> POSTER (PDF)

Poster #44a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Delirium is a condition characterized by fluctuating changes in cognitive function, attention, memory, and consciousness. Delirium is not known to be caused by one specific factor; rather, it is the balance between predisposing factors such as age, cognitive impairment, and comorbidities, as well as precipitating factors such as polypharmacy, infection, and surgery. Therefore, the rates of delirium in older adults who undergo anesthesia and surgery are especially high, with rates varying depending on the type of surgery that is performed. For example, patients who undergo hip fracture repair surgery develop post-operative delirium up to 53% of the time(1). Despite its frequency, delirium is often missed and underdiagnosed in clinical settings, leading to greater risk for injuries, mortality and prolonged hospitalization with persistent symptoms after hospital discharge. Current strategies for detecting delirium rely on subjective measures, where nurses and clinicians must observe changes in the cognitive function of their patients, with delirium assessed via the Confusion Assessment Method (CAM). Alternatively, urinary biomarkers for early detection of delirium may offer a more objective and less biased approach for risk assessment and/or diagnosis of vulnerable patients.

OBJECTIVE:
This project aims to apply metabolomics to identify a panel of urinary biomarkers that enable the prognosis and/or diagnosis of delirium onset in older patients following hip surgery using capillary electrophoresis-mass spectrometry (CE-MS).

METHODS:
In this pilot study, 56 urine samples from 9 patients (8 females, 1 male; mean age: 89; 2 total CAM+ patients) who underwent hip fracture repair surgery were collected over a period of 1 month along with the administration of a CAM assessment. The urine samples were analyzed using multisegment injection-CE-MS (MSI-CE-MS) to measure polar/ionic metabolites in both positive and negative mode. A targeted metabolomic approach was used using an authenticated list of known urinary metabolites, where their integrated peaks were normalized to an internal standard and hydration status adjusted for creatinine. After removing redundant, unreliable, and spurious signals, the data was analyzed using a partial least-squares-discriminant analysis (PLS-DA) model to rank urinary metabolites associated with CAM-positive delirium patients. Also, a receiver operating characteristic (ROC) curve was applied to identify a urinary metabolite ratio that differentiated delirium from delirium-free patients after hip surgery.

RESULTS:
The PLS-DA model showed that there was good separation between CAM-positive and CAM-negative patients. Two high ranking urinary metabolites (VIP score > 1.8) in this model were putatively identified as uric acid and phenylalanine. A ratiometric ROC curve using uric acid and phenylalanine was determined to provide the optimal discrimination between CAM-positive delirium patients to CAM-negative controls with an area under the curve (AUC) of 0.947 (p < 0.001).

DISCUSSION:
Changes in large neutral amino acids has often been implicated in the pathophysiology of delirium. For example, phenylalanine has been shown to be increased in serum in patients who developed post-operative delirium(2). In our urine samples, we observed a similar trend, with phenylalanine increased in CAM-positive patients. Additionally, lower uric acid in serum has been shown to be a risk factor for post-operative delirium in older patients(3). Similarly, our results also showed that lower uric acid correlated with a CAM-positive assessment. To further support these results, a ROC curve highlighted that a ratio of these two metabolites showed good discrimination between patients who developed delirium and patients who did not develop delirium.
In conclusion, the results from this pilot study indicate that a panel of urinary metabolites may serve as non-invasive biomarkers for the diagnosis of delirium in older patients following hip surgery. To further validate these preliminary findings, a multi-centre clinical study is currently underway to recruit a larger patient population with repeat urine samples collected during hospitalization to allow for greater study power.

REFERENCES:
(1) Cerejeira, J.; Mukaetova-Ladinska, E. B. A Clinical Update on Delirium: From Early Recognition to Effective Management. Nurs. Res. Pract. 2011, 2011, 1–12. https://doi.org/10.1155/2011/875196.
(2) Guo, Y.; Zhang, Y.; Jia, P.; Wang, W.; Zhou, Q.; Sun, L.; Zhao, A.; Zhang, X.; Wang, X.; Li, Y.; Zhang, J.; Jiang, W. Preoperative Serum Metabolites Are Associated With Postoperative Delirium in Elderly Hip-Fracture Patients. J. Gerontol. Ser. A 2017, 72 (12), 1689–1696. https://doi.org/10.1093/gerona/glx001.
(3) Xu, L.; Lyu, W.; Wei, P.; Zheng, Q.; Li, C.; Zhang, Z.; Li, J. Lower Preoperative Serum Uric Acid Level May Be a Risk Factor for Postoperative Delirium in Older Patients Undergoing Hip Fracture Surgery: A Matched Retrospective Case-Control Study. BMC Anesthesiol. 2022, 22 (1). https://doi.org/10.1186/s12871-022-01824-0.

E2 Infinity and Beyond: Advancing Estradiol Testing with the Cobas i 601 Analyzer Matthew Crawford (Presenter) Labcorp

Poster #45a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
The measurement of estradiol (E2) is used to evaluate a variety of developmental and biological processes. Estradiol is one of the most potent steroid hormones, requiring sensitive and selective measurement down to <2 pg/mL (pediatrics, male patients, post-menopausal females). Increased levels of E2, up to 4000 pg/mL, are also assessed in conjunction with fertility evaluations to monitor ovulatory status and in preparation for in vitro fertilization. The wide dynamic range and improved analytical selectivity/sensitivity provided by LC-MS/MS is beneficial for both measurements of excess and diminished estradiol levels. Complex extraction procedures and 2D LC systems are challenges to scalability for clinical laboratories. The Roche Cobas® i 601 analyzer represents a fully integrated solution that offers parallel sample processing and LC-MS/MS analysis.

METHODS:
The integrated, fully automated i 601 analyzer is designed as a sample-to-result solution mirroring clinical autoanalyzers that are ubiquitous in clinical diagnostic laboratories. The i 601 analyzer incorporates automated parallel sample processing (incubation with isotopically labeled E2 and pretreatment to release binding proteins) prior to antibody capture using paramagnetic particles. Following washing and elution, samples are partially evaporated on-board to concentrate extracts prior to liquid chromatography separation and analysis using an MRM mode triple quadrupole mass spectrometer. Chromatograms are automatically checked for peak quality prior to integration. Measurement results are calculated based on a leading calibration adjusted by an instrument-specific 2-point calibration. Estradiol was used to establish the performance characteristics of the i 601 analyzer.

RESULTS:
Preliminary validation experiments have been conducted for the analysis of estradiol in serum samples. Following assay and instrument calibration, studies were performed over 28 days without recalibration. Imprecision was determined according to the CLSI EP5 guideline with two manufacturer quality controls and two serum matrix pools. Total variability based on 5 replicates per day over 5 days was 3.2%-4.6% CV (Coefficient of Variation). Intermediate precision based on 2 replicates, 2 times per day over 20 days was 3.3%-4.5% CV. Method comparison (45 samples, spanning the i 601analyzer measurement range), to the Labcorp CDC certified LLE with 2D-LC-MS/MS assay resulted in a Deming slope of 1.00, correlation coefficient of 0.998 and mean bias of 2.3%. Assay and instrument robustness testing, representing anticipated laboratory operation, was performed with 40 serum pools spanning the measurement range. Comparing results analyzed in batch mode (only assaying E2) versus random access testing (multiple assays in parallel) resulted in a Deming slope of 1.01, correlation coefficient of 0.99817 and a mean bias of 1.0%. Replacement of all reagent materials (lot-to-lot variability assessment) resulted in a Deming slope of 1.01, correlation coefficient of 0.99813 and a mean bias of -3.2%. Assessment of longitudinal 28-day calibration stability resulted in a Deming slope of 0.98, a correlation coefficient of 0.99848 and a mean bias of -1.4%. Assessment of elevated blood matrix interferences, such as hemolysis, lipemia and icterus, resulted in biases ranging from -1.3% to 3.1%. Matrix equivalency was performed using a 5-point admixing scheme using the high calibrator (3000 pg/mL) and 3 different serum pools, resulting in a mean recovery of 100.8%. A gravimetric spike and recovery experiment was performed at 3 different levels using 3 unique serum pools, resulting in a mean recovery of 103.3%.
Recovery and imprecision results for samples spanning the ULMI (4000 pg/mL) were assessed using four serum pools gravimetrically prepared to within 80-120% of the ULMI. Samples >4000 pg/mL initially resulted as >Test and underwent automatic on-board dilution (1:4) prior to reanalysis. Results showed a mean recovery of 101.5%. Additional experimental data will follow in accordance with CLSI-C62.

CONCLUSIONS:
Linearity, matrix equivalency, and recovery studies demonstrated excellent precision and accurate quantitation across the measurement range. Results from method comparison were highly correlated to a CDC-certified laboratory developed test (LDT). Robust stability of calibration over 28 days and multiple reagent packs was shown. Improved process efficiency streamlines complex extraction and analysis techniques, resulting in 25-fold improvement in workflow efficiency (tech time). The enhanced process efficiency of the i 601 analyzer delivers an innovative, integrated solution that simplifies the complexity of estradiol testing.

Disclaimer: Cobas® i 601 analyzer and Ionify® Steroids 1 reagent packs are CE-marked. Further Ionify reagents are still under development. These products are not cleared or available for use in the US. COBAS and IONIFY are trademarks of Roche. This study was funded by Roche Diagnostics GmbH.

Impact of Collection Tube Type on Amino Acid Concentration: Plasma and Serum Maren Jensen (Presenter) ARUP

Poster #46a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

BACKGROUND:
Inherited disorders of amino acid metabolism are commonly diagnosed and monitored in patients through measurement of amino acids in biological fluids. Patterns of amino acids are indicative of specific disorders, the patient’s metabolic status, and they can also indicate dietary artifacts and/or medications. It is therefore important to understand the differences in amino acid concentration that could be associated with the specimen type used for the analysis. Here, we are comparing plasma derived from different anticoagulants and serum.

METHODS:
This study was conducted according to a protocol approved by the University of Utah IRB. Whole blood was collected from three self-reported healthy donors using the following four tube types: sodium heparin (NaHep), lithium heparin (LiHep), ethylene-diamine-tetra-acetic acid (EDTA), and serum separator (SST). The blood collection from each donor occurred consecutively and the samples were kept at ambient temperature until after the last donor’s collection was complete (16-30 minutes depending on donor order). Samples were refrigerated for ~45 minutes, then centrifuged at 1600 rcf for 10 minutes. The resulting plasma or serum layer was aliquotted to microtubes and stored at -20°C and -80°C for ~24 hours.

For the extraction, 20 µL of plasma or serum and 50 µL of deionized water were added to wells of a 96-well, 1-mL deep plate. A mixture of 5-sulfosalicylic acid and stable isotope-labeled internal standards (Chromsystems’ Amino Acid Analysis Internal Standard Set [plasma/serum] supplemented with L-alanine-13C3, L-glutamine-13C5, L-proline-13C5,15N, and L-valine-13C515N) was added to the wells (90 µL), then the plate was covered and vortexed for 3 minutes. After centrifugation for 5 minutes, 60 µL of supernatant were transferred to a second 96-well plate, followed by the addition of 120 µL of acetonitrile with 0.5% formic acid. The plate was covered, vortexed for 1 minute and centrifuged for 2 minutes. Each aliquot was analyzed in triplicate within a single batch.

Chromatographic separation and data acquisition were accomplished using an Intrada Amino Acids column (3 x 100 mm, 3 µm; Imtakt USA, Beaverton, OR) and Acquity I-class pump system paired to a Xevo TQ-XS triple-quad mass spectrometer with electrospray ionization (Waters, Milford, MA), using multiple reaction monitoring in positive ion mode. Amino acids were quantified using TargetLynx software (Waters); the mean and CV% of the triplicate measurements were calculated. The thirty amino acids with quantifiable results for at least one donor were evaluated for the comparison. The NaHep plasma was used as the reference collection tube; the percent bias of the mean concentration between NaHep and the remaining tubes (LiHep, EDTA, SST) was calculated by storage temperature (MS Excel).

RESULTS:
(1) LiHep plasma: all amino acid results were within ±15% of the values obtained in the NaHep plasma. (2) EDTA plasma: all donor samples showed a decrease in cystine (~20% at -80°C; ~35% at -20°C), as well as the appearance of a peak at the retention time of sarcosine with a concentration near the lower limit of quantitation (6-7 µM; LLOQ: 5 µM). (3) Serum (SST): all donor samples showed a pattern of increased arginine (25-119%), taurine (41-177%), glutamic acid (30-43%), serine (12-31%) and the appearance of aspartic acid (13-20 µM; LLOQ: 5 µM). This is most likely due to clotting-triggered leakage from platelets, which have a fairly high concentration of the above amino acids. The increase in concentration appears to correlate with the length of time before sample processing.

All aliquots stored at -20°C showed a drop in cystine compared to their counterparts at -80°C for each collection tube type, due to binding to protein. No other differences were noted when comparing the two storage temperatures.

CONCLUSION:
There was no significant difference in amino acid concentration between plasma derived from NaHep and LiHep stored at the same temperature. EDTA samples showed mild differences from NaHep for cystine and sarcosine, which would have minimal impact on clinical assessment. Serum (SST) showed increases for arginine, aspartic acid, glutamic acid, serine and taurine, which are likely caused by clotting and could become significant depending on clotting time. Plasma derived from collection tubes with heparin or EDTA anticoagulants had comparable amino acid profiles; however, caution should be exercised in interpreting amino acid results from serum specimens. Additionally, storage at -20°C impacted cystine concentration even within 24 hours.

Evaluation of a Fast Dart-MS Oral Fluid Method as a Potential Toxicology Screening Workflow Francois Espourteille (Presenter) Bruker Daltonics

Poster #47a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
DART-MSMS, Direct Analysis in Real Time, was introduced in the recent past as a technology capable of providing seconds analysis time, MS level data and semi to fully quantitative assays. Currently, most toxicology screens use immunoassay which suffer from a range of well-known shortcomings; false positive, false negatives, cross-reactivity, limited coverage across the toxicology target range, and long development time for new assays.

OBJECTIVE(S):
Developed a quick, semi-quantitative screening toxicology assay for oral fluids and in the process, demonstrate a workflow capable of replacing immunoassay as a screening technique.

METHODS:
Multi-drugs kits were provided by Pinpoint Testing, LLC. Kits contain regression curves and quality control samples for all drugs, in oral fluid matrix. The experimental work was performed on a Bruker DART-TQ+ platform. All drugs were tuned and optimized. A sample preparation method was optimized and transferred to a robotic platform (Opentrons), on which samples were extracted, subsequently evaporated (Biotage SpeedDry), and finally extracts were spotted on a Bruker HTS96 screen using a Rainin MicroPro20 system. Analysis was performed with DART on a Bruker TQ+ MS detector. The screened samples were subsequently confirmed on a Bruker LC-TQ+ MS.

RESULTS:
DART and TQ+ parameters were optimized for selectivity, sensitivity, precision, and run time. Unique MS/MS transitions and collision energies were identified for all target compounds followed by optimization of scan time (20 ms), cone temperature (300°C), and collision cell pressure (1.5 mTorr). Optimized DART temperature ranged from 250-350, depending on drug target. Grid voltages were optimized to 50 V for all drugs. Concentration range varied depending on specific drug cutoff. Data were processed via standard MS quantitation software, using internal standards.
Initial data shows the validity of the technique for toxicology screening. Three standards were used to generate a regression curve; the resulting R2 was typically 0.985 or better. Three sets of QCs were used, one below the cutoff, one slightly above the cutoff and a higher QC. Both the QC below cutoff and the QC above cutoff showed that on repeat analysis, they remained true to their original level. Blank QC were also run and were always negative.
Cross-validation of the samples demonstrated good correlation with LC-MS data, in that all samples that screened positive by DART were confirmed positive and all that screened negative were confirmed negative.

CONCLUSION:
This rapid chromatography-free workflow is adequate in determining the status of the sample at or below the common cutoff values with improved accuracy compared to IA based screening approaches. There was no evidence of significant false positives or false negatives. This workflow shows the DART-MS screening is capable of providing high quality data as a toxicology screening assay.

In-Depth Quantitative Characterization of Important Regulators and Predictive Biomarkers for CAR T-Cell Therapy-Induced Toxicities in a Clinical Cohort Jun Qu (Presenter) SUNY,Pharmaceutical Sciences Department

Poster #48a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
CAR T cells targeting B cell maturation antigen (BCMA) and CD19 achieve high response rates in B cell malignancies. However, toxicity remains a significant clinical challenge, with adverse events such as immune effector cell-associated neurotoxicity syndrome (ICANS) frequently observed following CAR T-cell infusion. To better address this toxicity and improve patient management, we aimed to investigate the mechanisms of ICANS by studying the dynamic changes of proteins in plasma and immune cells. In addition, we sought to identify candidate predictive biomarkers for CAR T-induced toxicities by comparing the pre-treatment protein profiles of patients who later developed ICANS with those who did not.

METHODS:
We analyzed a longitudinal cohort of 38 patients with relapsed/refractory multiple myeloma (n=21), non-Hodgkin lymphoma (n=16), and plasma cell leukemia (n=1) who underwent standard of care CAR-T therapy. We collected over 100 blood samples across 16 time points from Day 0 to Day 30 post-treatment and plasma and immune cells, including neutrophils, monocytes, and T cells, were isolated for proteomic profiling. The samples were prepared using a reproducible protocol and analyzed with a highly sensitive and robust LC-MS pipeline, featuring a trapping nano-LC system and Orbitrap Astral mass spectrometer. We quantified 4,000–6,000 proteins using nanogram-level inputs with high precision (e.g., median protein CV = 12.7%). Time-course bioinformatics analyses were generated for over 2,500 plasma proteins and more than 6,000 immune cell proteins per patient, with protein expression normalized by subtracting log2 intensity at Day 0. Patients with no toxicity were compared to those who developed ICANS to identify proteins associated with onset and progression. Additionally, baseline (day 0; pre-treatment) profiles were examined to identify potential predictive biomarkers for ICANS. Proteins that are significantly increased at baseline (fold change >1.4-fold, p < 0.05) in patients who later developed ICANS are candidate predictive biomarkers for ICANS.

RESULTS:
Baseline comparisons of plasma from control patients (no ICANS) and those who went on to develop ICANS (grades 1-5) revealed candidate biomarkers associated with ICANS development. These proteins are enriched in pathways related to glutathione metabolism and hydrogen peroxide catabolism (e.g., GLRX, GSR, GSHR), and MHC class I antigen presentation (e.g., HLA-A), suggesting a role for oxidative stress, redox regulation, and adaptive immunity in ICANS susceptibility and identifying potential avenues for early intervention. Baseline comparisons of neutrophils showed that proteins involved in complement activation (e.g., C4B, CFB, CLU, C9, C3, C7), acute-phase response (e.g., SERPINA1, SERPINF2), and platelet activation (e.g., VCL, FGA) were increased at baseline in patients who later developed ICANS. Longitudinal analysis of plasma proteins revealed distinct expression clusters, with upregulation of proteins involved in leukocyte and platelet aggregation as well as complement proteins over time in patients with ICANS. In these patients, neutrophils exhibited dynamic regulation of protein complex assembly and wound healing, highlighting their role in inflammation. T cells showed temporal changes in RNA splicing-related proteins consistent with T cell activation. Monocytes showed time-dependent regulation of vesicle organization and transport, suggesting involvement in antigen presentation and cytokine secretion. Collectively, these data suggest that complex immune crosstalk underlies the pathophysiology of ICANS.

CONCLUSION:
This is the first in-depth, time-resolved proteomic profiling of plasma and immune cells during CAR-T therapy, offering new insights into ICANS mechanisms that can guide risk stratification and management strategies for patients with multiple myeloma and lymphoma.

Enhancement of Traceability, Throughput and Ecologic Footprint of a Vitamin B1/B6 LC-MS Workflow via Fluent Automation Pauline Marini (Presenter) Eurofins Biomnis

>> POSTER (PDF)

Poster #49a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

BACKGROUND:
Vitamins B1 (thiamine) and B6 (pyridoxine) are essential micronutrients involved in numerous metabolic processes, including energy production, neurotransmitter synthesis, and amino acid metabolism. Deficiencies in these vitamins can lead to a range of neurological and hematological disorders, making accurate assessment of their levels crucial for both clinical diagnosis and nutritional research. Quantification of vitamin B1 and B6 typically requires analysis of plasma or whole blood samples, which can be challenging due to low endogenous concentrations and the instability of these analytes [1,2]. Automated sample preparation and analysis workflows have become increasingly important [3], enabling high-throughput, standardized, and reproducible processing of whole blood specimens. These advances help ensure reliable monitoring the Vitamin B1 and B6 status in clinical practice and research settings.

OBJECTIVE:
This study aimed to evaluate the sample preparation and analysis of two vitamins, namely Vitamin B1 or Thiamin, and Vitamin B6 or Pyridoxalphosphate, by automation with a Tecan Fluent and a Tecan LC-MS kit compared to the previously established lab developed test with a Tecan Freedom Evo automation. The focus was on evaluating improvements in sample tracking, processing efficiency, and resource consumption compared to conventional methods, and determining the suitability of this novel automated workflow for routine laboratory applications in vitamin status monitoring.

METHODS:
Sample preparation followed the distributed by Tecan LC-MS kit Ref # 30261557 / 30261558. Hereby, whole blood samples were being homogenized as well as extracted a deproteinization solution and addition of isotopically labeled internal standards of each of the analytes. Six calibrators and quality control samples at three concentration levels, within the expected physiological range, were reconstituted from a lyophilized whole blood matrix and treated as samples thereafter. Subsequently, the supernatant of these extracts was injected into an LC-MS/MS system consisting of a Thermo Scientific Vanquish UHPLC and a TSQ Quantis mass spectrometer operated by Chromeleon. For the automated procedure, a Tecan Fluent 780 liquid handler equipped with a robotic gripper arm as well as an 8-channel pipetting arm with piercing tips was utilized. The liquid handler furthermore was equipped with a 4-slot plate centrifuge, four orbital shakers as well as two tube rotators with integrated barcode scanners. Total durations of both workflows were taken and their analytical performance was evaluated.

RESULTS:
The average total duration per plate with the automated workflow in an eight-plate-scenario was approximately 33.00 minutes. The approximation included a sample scanning duration of 2.00 minutes, a plate transfer and shaking duration of 6.00 minutes, reagent distribution durations of 4.00 minutes, sample distribution durations of 12.00 minutes, centrifugation durations of 4.00 minutes and a supernatant transfer of 5.00 minutes. The average total duration per plate with the established workflow in a four-plate-scenario was 35.00 minutes. This included a sample scanning duration of 2.50 minutes, reagent distribution durations of 5 minutes, shaking times of 6.25 minutes, sample distribution times of 7.50 minutes, a centrifugation of 3.75 minutes a supernatant transfer of 5.00 minutes as well as a de- and recapping of sample tubes of 5.00 minutes.

CONCLUSION:
The Tecan Fluent automation proved to be a robust and valuable enhancement of vitamin B1/B6 LC-MS worklflows. Sample prep time and most importantly staff occupation has been shown to be reduced significantly compared to established workflow. While pipetting steps themselves did not show significant differences in duration, the automated cap piercing had an impact on the overall workflow. With its analytical performance and its low environmental impact, the novel automated workflow should be the preferred fit for a sustainable and state-of-the-art laboratory environment.

REFERENCES:
1. R.J.A.C. Roelofsen-de Beer, B.D. van Zelst, R. Wardle, P.G. Kooij, Y.B. de Rijke,
Simultaneous measurement of whole blood vitamin B1 and vitamin B6 using LC-ESI–MS/MS,
Journal of Chromatography B, Volume 1063, 2017, Pages 67-73
2. Thomas M Annesley, Larry Clayton, Simple Extraction Protocol for Analysis of Immunosuppressant Drugs in Whole Blood, Clinical Chemistry, Volume 50, Issue 10, 1 October 2004, Pages 1845–1848, https://doi.org/10.1373/clinchem.2004.037416
3. More, D., Khan, N., Tekade, R. K., & Sengupta, P. (2024). An Update on Current Trend in Sample Preparation Automation in Bioanalysis: strategies, Challenges and Future Direction. Critical Reviews in Analytical Chemistry, 1–25. https://doi.org/10.1080/10408347.2024.2362707.

Easy Strategy for the Addition of New NPS Compound: Differentiation of Methylfentanyl Isomers and Isobaric Compound Catherine Daigle (Presenter) Laboratoire de Sciences Judiciaires et de Médecine Légale

>> POSTER (PDF)

Poster #50a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION
New Psychoactive Substance (NPS) are a diverse group of compounds synthesized in clandestine laboratories to mimic drugs of abuse that are illegal by law. In an effort to stay ahead of regulatory frameworks, the chemical structures of these substances – often derivatives of opioids or benzodiazepines – are frequently modified. These structural changes often result in increased potency, making the substances significantly more dangerous than their parent drug. For instance, protonitazene is approximately three times more potent than fentanyl. To effectively monitor the emergence of these substances on the illicit market, a flexible analytical method is required – one that allows for the rapid and straightforward inclusion of new compounds without the need for chromatographic modifications.

OBJECTIVES
Development of a simple and rapid approach to incorporate ortho-methylfentanyl into the NPS screening method, considering the presence of isobaric and isomers compounds, without altering chromatographic conditions.

METHODS
Urine and blood samples spiked with a solution containing over 69 NPS compounds were extracted by protein precipitation using a mixture of acetonitrile and acetone. Chromatographic separation was performed using a Zorbax C18 column with a gradient of ammonium formate (pH 3.0) with acetonitrile. Compound optimization and detection were carried out on a SCIEX API 6500 mass spectrometer operating in positive ion mode. Analyte identification was based on retention time and ion ratio using two mass transitions. For unknown samples, a substance was considered present if the retention time was within ±0.05 min and the ion ratio within ±30% of the average reference samples values.

RESULTS
In the 8th version of the NPS determination method, 15 new psychoactive substances were successfully incorporated into the existing method. For the majority of these compounds, the addition was achieved without any interference in quantification. However, selectivity was an issue for the addition of ortho-methylfentanyl. Specifically, the presence of the cis-3-methylfentanyl isomer and isobaric compounds - butyryl fentanyl and isobutyryl fentanyl - in the reference sample, caused interference with the mass transition selected for ortho-methylfentanyl (m/z: 351, trans:351/146 351/188).

O-methylfentanyl (RT: 7,00 min, ion ratio¹: 26.3) chromatographic resolution and selectivity was achieved with isobutyryl fentanyl (RT: 7.13 min, ion ratio¹: 5.1) and butyryl fentanyl (RT: 7.27 min, ion ratio¹ 5.1) and partially achieved with cis-3-methylfentanyl (RT: 7.07, ion ratio¹: 9126.7). Moreover, partial chromatographic separation was also observed between isobutyryl fentanyl (RT: 7.13 min) and cis-3-methylfentanyl (RT: 7.07). The partial chromatographic separation between o-methylfentanyl, cis-3-methylfentanyl and isobutyryl fentanyl compromised the analytical method’s selectivity. The contribution of each other at the selected transitions led to altered ion ratios in the reference solution, thus affecting the ability to confidently identify individual substances in unknown samples. For example, the ion ratio of cis-3-methylfentanyl alone was 2941.6, compared to 965.9 when measured in the reference solution containing all interfering compounds.

In this context, modifying the chromatographic conditions would have required extensive development time and was therefore not considered a viable option. During compound optimization, all detectable fragments for each substance were evaluated and recorded in a table. This summary table was then used to assess whether an alternative set of transitions could offer improved selectivity for each isomer, without compromising sensitivity.

Among the 11 transitions tested for cis-3-methylfentanyl and ortho-methylfentanyl, only the fragment at m/z 150 Da was selective for cis-3-methylfentanyl. However, ion ration selectivity requires two selective transitions, which was not achievable for this compound. In contrast, for ortho-methylfentanyl, the transitions at m/z 188 Da and 146 Da demonstrated sufficient selectivity, minimizing the impact of cis-3-methylfentanyl on the ion ratio. Using the transitions identified from the summary table, the ion ratio for ortho-methylfentanyl alone was 26.3, compared to 21.4 in the reference solution containing all isobaric compounds.

In this case, selectivity for cis-3-methylfentanyl could not be achieved in the presence of ortho-methylfentanyl within the reference solution. After evaluating the prevalence of cis-3-methylfentanyl in our cases between 2019 and 2025, it appeared that cis-3-methylfentanyl had never been detected in any sample. Consequently, this compound was removed from the method. Method selectivity is ensured through retention time and ion ratio criteria. If an interfering peak is observed at the transitions of o-methylfentanyl but fails to meet ion ratio acceptance thresholds, the presence of cis-3-mehylfentanyl can be suspected and confirmed using an alternative method. Similarly, the ion ratio can be used to differentiate between cis-3-methylfentanyl and isobutyryl fentanyl.

This evaluation was also extended to potential interferences from other isomer, such as trans-3-methylfentanyl and p-methylfentanyl, and demonstrated acceptable selectivity for o-methylfentanyl.

CONCLUSION
Our NPS analytical method is continuously evolving to ensure the detection of emerging compounds on the illicit market. Ion ratio is a useful tool to differentiate isobaric compounds in addition to chromatographic resolution. The development of a summary table containing optimization data for all NPS compounds has proven to be an effective resource for quickly resolving selectivity issues and will be instrumental in guiding future updates of the NPS method.

¹ Ion ratio at o-methylfentanyl selected transition.

Paper Spray Mass Spectrometry Utilizing On-Paper Enzyme Hydrolysis of Glucuronide Metabolites for Rapid Drug Quantitation in Urine Taelor Zarkovic (Presenter) Vancouver Island University / University of Victoria

Poster #51a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Urine drug screening is a routine measurement in clinical laboratories for ensuring substance use disorder patient compliance and therapeutic drug monitoring, as well as for forensic toxicology and employment drug testing. During metabolism, opioids and other drugs of abuse form hydrophilic glucuronide conjugates for urinary excretion. Typically, glucuronides are either acid or enzymatically hydrolysed from the parent drug prior to measurement by liquid chromatography mass spectrometry (LC-MS) to provide total patient drug levels. While effective, these methods require sample preparation and clean-up steps, complicating measurement and increasing sample turn-around time. We present a rapid (1.2 minute) and direct paper spray mass spectrometry (PS-MS) method that utilizes on-paper enzyme hydrolysis of glucuronide drug metabolites in urine.

OBJECTIVE(S):
The primary objective of this research is to evaluate the feasibility of an on-paper enzymatic hydrolysis of glucuronide metabolites prior to urine drug screening using paper spray mass spectrometry.

METHODS:
Measurements were performed in positive ion mode using a triple quadrupole mass spectrometer (Thermo Fisher Scientific™ TSQ Altis™) equipped with a Verispray™ PaperSpray ion source (Thermo Fisher Scientific™). For quantitation and confirmation, selected reaction monitoring with three transitions for target analytes and two transitions for isotopically labelled internal standards was utilized. A 90/10/0.1 acetonitrile/water/formic acid (v/v/v%) spray solvent was used for PS-MS analysis, with a single PS-MS measurement complete in 1.2 minutes.

For each sample measurement, 5 uL of beta-glucuronidase enzyme (B-One Kura Biotech) was deposited on VeriSpray™ PS-MS sample strips (Thermo Fisher Scientific™), followed by 5 uL of drug free human urine fortified with a cocktail of glucuronide metabolites (codeine-6-beta-D-glucuronide, morphine-3-beta-D-glucuronide, hydromorphone-3-beta-D-glucuronide, lorazepam glucuronide, oxazepam glucuronide, and temazepam glucuronide) and internal standards (codeine-d6, morphine-d6, and hydromorphone-d3, lorazepam-d4, oxazepam-d5, and temazepam-d5). Calibration curves were obtained using standards prepared in drug free human urine fortified with parent drug standards (codeine, morphine, and hydromorphone, lorazepam, oxazepam, and temazepam) from 10 – 1500 ng/mL.

RESULTS:
Calibrations for the parent drugs in blank human urine display excellent linearity over three orders of magnitude with R2 > 0.99. Limits of detection range from 7 to 52 ng/mL for the target parent drugs and lower limits of quantitation are between 10 and 100 ng/mL. The presence of enzyme on the paper does not affect parent drug quantitation.

Deposited enzyme volumes (2 – 10 uL) were evaluated for on-paper hydrolysis of a 5 uL urine sample. The optimal volume of enzyme that provided efficient glucuronide hydrolysis and reproducibility was found to be 5 uL (1:1 enzyme:urine ratio).

Effective enzymatic cleavage of glucuronide metabolites to yield parent drugs was observed for target molecules tested over the relevant clinical range of 10 - 1500 ng/mL. Hydrolysis reproducibility was assessed at two levels (500 and 1000 ng/mL) with an average % CV of 16% and % bias of 10% for opioid glucuronides. Benzodiazepine glucuronides display what appear to be lower hydrolysis efficiencies for on-paper hydrolysis. This is being explored further and preliminary data suggests the enzyme may have some activity towards the benzodiazepine parent drugs and ISTDs, in addition to the glucuronides.

CONCLUSION:
On-paper hydrolysis of glucuronide drug metabolites in urine, followed by paper spray mass spectrometry measurement has been demonstrated. Coupling the enzymatic hydrolysis directly to PS-MS measurement enables fast sample turnaround times, significantly lower reagent consumption, and reduced cost of analysis.

Defining, Measuring, and Evaluating Operational Excellence for Three DIA-MS Informatics Solutions Gautam Saxena (Presenter) GoldenHaystack Lab

Poster #52a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
For nearly 30 years, MS-based discovery proteomics has been promising to outshine genomics. However, at least four major challenges have historically prevented our community from realizing the expected potential benefits: (a) the dynamic range of the LC-MS instruments was not sufficiently large; (b) the throughput of the LC-MS instruments often prevented larger-scale studies (i.e., 100s or 1000s of samples); (c) >99% of labs used informatics solutions that only accessed < ~10% of the data in the MS files (despite the much higher biological value of the peptides present in the remaining ~90%); and (d) the informatics was too slow, disjointed, difficult to use, and of questionable robustness to realistically analyze multiple large-scale projects in any reasonable timeframe and showcase the results to scientists primarily outside of our community.

However, in the last several years, technical advantages have largely addressed the first three concerns: the LC-MS vendors have released order-of-magnitude improvements in both sensitivity and dynamic range for their latest instrument platforms while simultaneously increasing throughput by at least 3x; about half-a-dozen small companies have been founded with the ability to enrich the lower abundant proteins for plasma and CSF samples; and, in early 2025, we introduced an algorithm called GoldenHaystack (GH) that can quantify not just (a) the ~10% of peptides in DIA-MS datasets that are in protein library (i.e., FASTA) search spaces (like traditional algorithms) but also (b) the biologically invaluable remaining ~90% of peptides in DIA-MS files that may not have an initially known sequence or PTMs (unlike traditional algorithms).

OBJECTIVE(S):
None of those above independent technical developments to characterize the plasma proteome by MS matter however if the informatics solution is almost unusably slow, overly complicated to use, lacking in scalable visualization tools (to ensure auditability/transparency of informatics conclusions), and/or of questionable robustness. In other words, the informatics now needs to achieve operational excellence that is measurable. We therefore set out to first define and then measure DIA-MS informatics operational excellence (a) for our GH algorithm and, to the extent possible, (b) for two popular DIA-MS informatics solutions.

METHODS:
For a 2583 cerebrospinal fluid (CSF) sample of a case control design, where case were individuals with Parkinson’s Disease (PD) and control were individual with similar clinical features but with no PD diagnosis, CSF samples were digested with trypsin and were analyzed on the Orbitrap Exploris 480 (ThermoFisher) using DIA-MS. We measure for GH, DIA-NN and MSFragger: (a) wall clock time for end-to-end informatics processing (i.e., from raw file to final AI/ML results); (b) wall clock time when a pipeline is rerun with either a single LC-MS file added/removed or a single search parameter changed; (c) wall-clock time for re-running the informatics pipeline after a failure occurs in the middle or towards the end of the pipeline; (d) reproducibility of results (i.e., degree to which running the same pipeline twice produces identical results); and (e) wall-clock time to visualize XICs and spectra for both expected MS2 fragments as well as user-queried MS2 fragments.

RESULTS:
GH processed end-to-end the 2583 raw file PD project in less than 7 hours on a single large computer of 180 CPU cores (360 virtual threads when hyperthreading was turned on); for rerunning of the pipeline for a change in search parameters, it took less than 1 hour; for failures midway or towards the end of the pipeline, the net additional time was negligible (< 5 minutes) to “continue from point of initial failure” onwards; for reproducibility, running the pipeline twice produced 100% identical results; and, for visualization of XICs, it took seconds to produce the graphs, even on projects of 100s or 1000s of samples. For MS-Fragger and DIA-NN, on the identical 180 CPU core computer, processing times were typically either orders-of-magnitude larger or it was not possible to run the pipeline in any reasonable timeframe; DIA-NN showed variation in results when running the exact same pipeline twice with the same input; and neither DIA-NN nor MS-Fragger could support visualization of the XICs through Skyline.

CONCLUSION:
In this work, we have defined and measured operational excellence for DIA-MS informatics and show phenomenal measures of operational excellence for our recently introduced solution named GoldenHaystack.

Identifying Genetic Drivers of Pediatric Chronic Pancreatitis using Urinary Proteomic Profiles Katelyn Brusach (Presenter) The Mayo Clinic

Poster #53a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
Understanding how genetic mutations contribute to pancreatitis is essential for advancing precision medicine and developing targeted interventions. Mutations in CFTR, PRSS1, and SPINK1 are well established genetic drivers of chronic pancreatitis (CP). While there have been advancements in understanding how these genes promote pancreatitis, the underlying biological mechanisms remain incompletely understood. Since CP has been linked to smoking and alcohol consumption, children with pancreatitis provide a unique cohort for studying hereditary contributors to disease onset and severity. Identifying the proteomic profiles of pancreatic drivers would allow for disease tracking, personalized and precision medicine, and ultimately advance our understanding of disease onset and progression. Biofluids provide pathophysiological information without requiring tissue biopsy. Urine is an underexplored, noninvasive biofluid that may capture pancreatic pathophysiology without requiring a blood draw or complex processing. To evaluate the causality of genetic factors involved in CP, this study aims to evaluate the impact of these genetic mutations using urinary proteomic profiles to better understand pathways contributing to disease prognosis.

METHODS:
A total of 46 CP patients with genetic testing of CFTR, PRSS1, and SPINK1 were included in this study. All patients were below the legal drinking age (range = 3-20 years, median = 14 years). To assess the relationship between urinary proteins and gene positive or negative results, urine samples were also collected from all 46 patients. Urine proteomics samples were quantified using data-independent acquisition mass spectrometry.

RESULTS:
Of 46 genetically tested samples, 37% were positive for CFTR (17/46), 26% were positive for PRSS1 (12/46), and 26% were positive for SPINK1 (12/46). A total of 1761 urinary proteins were detected and quantified. We identified 241 significant pathways associated with urinary proteins in gene positive patients. CFTR had 28 significantly increased pathways and 40 significantly decreased pathways (68 total). Comparatively, PRSS1 had 8 significantly increased pathways and 42 significantly decreased pathways (50 total). Finally, SPINK1 had 76 significantly upregulated pathways and 46 significantly downregulated pathways (122 total). Of the significantly increased pathways, 50% of the CFTR pathways were associated with cell death (64% were related to p75NTR). While PRSS1 did not have many significantly increased pathways, 38% of them were associated with metabolism. Within the 76 significantly increased SPINK1 pathways, 26% were related to cell proliferation and 20% were related to cell signal transduction (of which 27% were associated with ALK). Pathways included were filtered for significance (p < 0.05) and FDR corrected (< 0.05). Significant pathways with only one gene hit were excluded.

DISCUSSION:
Understanding how genetic mutations contribute to the onset and progression of CP will help identify opportunities for early diagnosis, risk prediction, and ultimately precision medicine. In this study, we identified urinary proteomic profiles which give insight into how pathways linked with CFTR, PRSS1, and SPINK1 positive patients may promote disease. We found that CFTR-related pathways were connected to cell death, with 50% of cell death pathways being related to p75NTR, which is believed to be a driver of metabolic and fibrotic diseases. Alternatively, PRSS1 pathways were related to metabolic dysfunction. SPINK1 pathways were largely associated with the regulation of cell proliferation, cell signal transduction, and metabolism of RNA, which may ultimately lead to neoplasia. SPINK1 cell signal transduction pathways were 27% ALK-related, which has been linked to cancer. CP is a known risk factor for pancreatic cancer, so understanding which patients are likely to progress may be important in monitoring and prognosis. The distinct pathways associated with these three genes suggest that genetic mutations may produce unique molecular phenotypes indicating different mechanisms of disease progression. Identifying these different mechanistic drivers of disease represented in urinary proteomic profiles will improve personalized patient care plans and ultimately improve patient outcomes. We propose that pathophysiological insights associated with genetic modifications are reflected in urinary protein profiles, which would provide novel insights into the mechanisms driving disease progression.

Spot on: A Green(er) Approach to Dried Matrix Spots Using Liquid Microjunction – Surface Sampling Probe – Mass Spectrometry and Computer Vision Daniel Reddy (Presenter) Queen's University at Kingston

Poster #15a View Map

This poster will be attended on Wednesday at 09:15 for 1 hour 15 minutes in the Exhibit Hall.

INTRODUCTION:
In 2023, the global medical devices market was estimated at $518 billion USD; the 2023 global paper diagnostics market, which includes volumetric absorptive microsampling devices [Within which dried matrix spots (DMS) are situated], was valued at $15.5 billion USD, which equates to ~3% of the overall medical devices market. With DMS, a fixed volume of sample, generally ~10 microliters, is volumetrically absorbed into a substrate; DMS methods have been noted as clinically-relevant, cost-effective, simple, and reproducible. Though blood microsampling is perhaps the most commonly-recognized DMS application, other samples include saliva and urine for a range of analyte scenarios. Regardless of the sample, DMS often rely on at least tens-of-microliter volumes, and DMS preparative/sampling methods usually involve direct transfer of the biofluid onto a sample storage medium, which is oftentimes filter paper. This transfer step has several inherent pitfalls, including variance in sample volume being applied and/or uneven lateral distribution of the sample. Furthermore, most analytical techniques cannot be directly integrated with the storage medium, i.e., the biofluid spot must be extracted offline from the substrate for complex and solvent-intensive analytical methods. Additionally, DMS methods have not yet been thoroughly supplemented by artificial intelligence/computer vision.

OBJECTIVES:
Given the market interests in paper-based devices, we have combined a simple hydrophobic surface treatment with laser-micromachining to modify a paper substrate with surface energy traps (SETs) to precisely confine, dry-down, and directly analyze DMS with liquid microjunction - surface sampling probe - mass spectrometry (LMJ-SSP-MS) [Note that the LMJ-SSP enables a combined (automatable) workflow of sample extraction and subsequent introduction of analyte(s) to the MS]. Additionally, we are creating and optimizing an object-based detection program to leverage computer vision to aid the sampling process. Taken together, we anticipate that this work will impact: 1) The manner by which DMS are prepared and then 2) the analytical process(es) through which DMS are interrogated.

METHODS:
The modified (paper) substrates are rendered hydrophobic by dip-coating with a commercially available coating formulation, Aculon (San Diego, California USA). After the surface hydrophobicity treatment, the substrate(s) is/are micromachined with an Oxford Lasers A Series picosecond compact laser-micromachining system. Note that the laser-micromachining system is used to both excise the (paper) substrates from the parent Whatman filter paper sheet and to machine the circular SETs, typically less than 1 mm diameter, and the reticules for visual targeting. SETs are visually assessed with various aqueous dye solutions and a combination of optical and scanning electron microscopies, as well as goniometry. Lastly, SETs are qualitatively/quantitatively assessed using various aqueous and artificial biofluid solutions and LMJ-SSP-MS, in particular a model analyte recovery study. In-house, Python-based computer programs have been developed to control the 3D-printer chassis's xyz movement, as well as the camera's ability to recognize the targeting reticules via computer vision.

RESULTS:
Previously, we have observed analyte signal reductions (Up to 32-fold) when sampling from a porous substrate. Here, laser-micromachining enables the creation of superficial-yet-functional SETs, meaning that by tuning the laser power and speed, the SETs are prepared such that the liquid is confined within and dries-down on the SET without penetrating deeply into the substrate; this superficiality maximizes sample recovery/introduction via LMJ-SSP-MS. After the surface hydrophobicity treatment, the modified paper substrate achieves a nearly superhydrophobic water contact angle, roughly 145 degrees, given the surface roughness afforded by the cellulose fiber network in conjunction with the Aculon coating. After laser-micromachining, a one microliter droplet is neatly retained within the SET and dries-down within the laser-micromachined area with minimal lateral diffusion (< 5%) and practically imperceptible transverse diffusion. Preliminary results using a bilirubin standard in artificial urine show promise that near-quantitative recoveries could potentially be achieved, though in this study we are using the absence/presence of bilirubin in an artificial urine standard solution as a binary "Yes/No" proof-of-principle example, i.e. a rudimentary model (hyper)bilirubinemia system. Moreover, under certain dimensional parameters, Version 1 of the computer vision program can detect a model SET's center within an average of 0.023 mm (2.3% error) (n = 3).

DISCUSSION/CONCLUSION:
Especially on a paper substrate, SETs can be difficult to visualize for sampling, so the targeting indicators, or reticules, that we have prepared around the SETs for visualization are poised to be a marked improvement to the design and use of (paper) substrates for DMS preparation and sampling. Moreover, the introduction of novel substrates like polymeric materials for DMS preparation could further increase the appeal of DMS as a diagnostic tool. As this project progresses, we plan to include additional demonstrations using analyte(s) of interest in artificial blood and saliva solutions. Lastly, in keeping with trends towards laboratory automation and high throughput assay(s)/screening(s), we are working to ensure that our early-stage object detection program that recognizes the reticules on the various substrates is accurate, precise, and robust, i.e., integrable with other xyz locomotion systems.