LC–MS/MS quantification of N-lactoyl-phenylalanine and N-lactoyl-tyrosine in human plasma: method validation and clinical application to metformin administration

Abstract

N-lactoyl amino acids, including N-lactoyl-phenylalanine (Lac-Phe) and N-lactoyl-tyrosine (Lac-Tyr), have recently attracted interest as endogenous metabolites of potential clinical and metabolic relevance. In particular, Lac-Phe has been implicated in appetite regulation and in the metabolic effects of metformin. However, validated analytical methods for the simultaneous quantification of these endogenous metabolites in human plasma remain limited. In this study, a sensitive liquid chromatography-tandem mass spectrometry method was developed and validated for the simultaneous quantification of Lac-Phe and Lac-Tyr in human plasma using a simple protein precipitation procedure with a 50-μL sample volume. To minimize interference from endogenous analytes, an eight-point calibration curve was prepared using an artificial surrogate matrix. The assay showed a calibration range of 0.5–100 ng/mL for both analytes, with a lower limit of quantification of 0.5 ng/mL and a total run time of 5.5 minutes. The method demonstrated acceptable analytical performance for both Lac-Phe and Lac-Tyr, with intra- and inter-day precision values of ≤ 8.8% and mean accuracy ranging from 88.0% to 101.4%. The validated assay was successfully applied to plasma samples from healthy volunteers collected before and after metformin administration. This method provides a reliable analytical tool for the simultaneous quantification of Lac-Phe and Lac-Tyr in human plasma and may facilitate future clinical and pharmacological investigations of lactoyl-amino acid metabolism.

Trial Registration

ClinicalTrials.gov Identifier: NCT04335526

INTRODUCTION

Lactoyl-amino acids are an emerging class of endogenous metabolites formed by the conjugation of lactate with amino acids through carnosine dipeptidase 2 (CNDP2)-mediated reverse proteolysis [1]. Within this metabolite class, N-lactoyl-phenylalanine (Lac-Phe) has recently attracted particular interest because of its potential role in metabolic regulation [2,3]. Recent studies have suggested that Lac-Phe may contribute to appetite regulation and may also be involved in the metabolic effects of metformin, which is widely used in the treatment of type 2 diabetes [4,5,6]. These findings have further expanded interest in the clinical relevance of Lac-Phe. As interest in the physiological and pharmacological significance of Lac-Phe continues to grow, the need for reliable analytical methods capable of accurately quantifying this metabolite in human biospecimens has also increased. N-lactoyl-tyrosine (Lac-Tyr) is another CNDP2-derived lactoyl-amino acid that has been reported to increase together with Lac-Phe following metformin administration [1,7]. Although its biological and clinical significance remains less well characterized, Lac-Tyr was included as an exploratory analyte to investigate coordinated changes in lactoyl-amino acid metabolism.

Accordingly, a robust and sensitive analytical method is required for the reliable measurement of these endogenous metabolites in human plasma. However, accurate quantification of endogenous compounds in biological matrices remains analytically challenging because of matrix effects, low circulating concentrations, and the lack of a true blank matrix. These issues are particularly important for endogenous analytes, for which the suitability of surrogate or alternative matrices must be carefully considered during method development and validation [8]. Despite growing biological interest in lactoyl-amino acids, validated analytical methods for the simultaneous quantification of Lac-Phe and Lac-Tyr in human plasma remain limited.

In this study, we developed and validated a sensitive and selective liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for the simultaneous quantification of Lac-Phe and Lac-Tyr in human plasma. To minimize interference from endogenous analytes, calibration standards were prepared using an artificial surrogate matrix. The method was validated in accordance with bioanalytical method validation principles, including assessments of selectivity, linearity, accuracy, precision, matrix effects, extraction recovery, carryover, and stability, and was subsequently applied to plasma samples collected from healthy volunteers before and after metformin administration. This validated assay provides a reliable analytical tool for future clinical and metabolic studies of lactoyl-amino acid metabolism.

METHODS

Chemicals and reagents

N-Lactoyl-phenylalanine (Lac-Phe, 95%) was purchased from Sigma-Aldrich (St. Louis, MO, USA), and N-lactoyl-tyrosine (Lac-Tyr, 95%) was obtained from TRC Reference Materials (Toronto, ON, Canada). The internal standard (IS), N-lactoyl-phenylalanine-d₅ (Lac-Phe-d₅, 97%), was purchased from Hello Bio (Bristol, UK). Phosphate-buffered saline (PBS, 1.0 M, pH 7), bovine serum albumin (BSA, 98%), and formic acid (95%) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Methanol (99.9%), acetonitrile (99.9%) and water of HPLC-grade were purchased from J.T. Baker (Phillipsburg, NJ, USA).

Preparation of solutions and surrogate matrix

Stock solutions of Lac-Phe, Lac-Tyr, and Lac-Phe-d₅ were prepared at a concentration of 1 mg/mL in 50% (v/v) methanol in water. The stock solution of Lac-Phe-d₅ was further diluted with 50% methanol to a concentration of 1,000 ng/mL and used as the IS working solution. Working solutions of Lac-Phe and Lac-Tyr were prepared by serial dilution of the respective stock solutions with 50% methanol. All stock solutions were stored at −20°C until use. Lac-Phe and Lac-Tyr are endogenous compounds present in human plasma. Therefore, BSA dissolved in PBS was used as a surrogate matrix. The surrogate matrix was prepared by dissolving BSA in PBS to a final concentration of approximately 40 mg/mL.

Preparation of calibration standards and quality control (QC) samples

Calibration standards were prepared by spiking Lac-Phe and Lac-Tyr into the surrogate matrix at final concentrations of 0.5, 1, 5, 10, 20, 50, 80, and 100 ng/mL by adding 10 μL of a mixed working solution of Lac-Phe and Lac-Tyr to 190 μL of the surrogate matrix. QC samples were prepared in the surrogate matrix at final concentrations of 0.5 ng/mL (lower limit of quantification, LLOQ), 1.5 ng/mL (low quality control, LQC), 30 ng/mL (medium quality control, MQC), and 75 ng/mL (high quality control, HQC) by adding 20 μL of the Lac-Phe/Lac-Tyr mixed working solution to 380 μL of the surrogate matrix. Five replicates were prepared at each QC concentration. Dilution quality control samples were prepared at 10- and 50-fold dilution levels by adding 20 μL of a Lac-Phe/Lac-Tyr mixed solution to 380 μL of the surrogate matrix. Five replicates were prepared for each dilution level within a single analytical batch.

Sample preparation

Sample preparation was performed using simple protein precipitation. For the preparation of calibration and QC samples in a surrogate matrix, 10 μL of the IS working solution was mixed with 50 μL of the surrogate matrix. For study samples, 10 μL of the IS working solution was mixed with 50 μL of human plasma.

Protein precipitation was achieved by adding 200 μL of acetonitrile:methanol (1:1, v/v) to each sample, followed by vortex mixing for 5 min at room temperature. The samples were centrifuged at 18,000 × g for 10 minutes at 4°C. An aliquot (200 μL) of the supernatant was transferred to a new tube and centrifuged again under the same conditions. Finally, 180 μL of the supernatant was transferred to an autosampler vial for LC–MS/MS analysis.

LC-MS/MS conditions

Quantitative analysis of Lac-Phe and Lac-Tyr was performed using an Agilent 1260 Infinity LC system coupled with an Agilent 6490 triple quadrupole mass spectrometer (Agilent Technologies, Santa Clara, CA, USA).

Chromatographic separation was achieved on an ACQUITY UPLC HSS T3 column (100 Å, 1.8 μm, 2.1 × 100 mm; Waters, Milford, MA, USA). The mobile phase consisted of solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in methanol). Gradient elution was performed at a flow rate of 0.3 mL/min. The gradient program started at 90% A (0.00 min), maintained until 0.50 minutes, decreased linearly to 10% A at 1.50 minutes, returned to the initial condition at 3.00 minutes, and was maintained until 5.50 minutes for column re-equilibration. The column temperature was maintained at 40°C, and the injection volume was 5 μL.

Mass spectrometric detection was performed using multiple reaction monitoring (MRM) in positive electrospray ionization mode. The capillary voltage was set to +3,000 V and the nozzle voltage to +1,500 V. Nitrogen was used as the nebulizer gas (30 psi), drying gas (14 L/min at 200°C), and sheath gas (10 L/min at 350°C).

The monitored MRM transitions (collision energy) were as follows: Lac-Phe, m/z 238.1 → 120.1 (15 V); Lac-Tyr, m/z 254.0 → 136.1 (15 V); and Lac-Phe-d₅ (IS), m/z 243.1 → 125.1 (15 V).

Method validation

The bioanalytical method for the quantification of Lac-Phe and Lac-Tyr was validated in accordance with the ICH M10 guideline on bioanalytical method validation [9]. An eight-point calibration curve was constructed by plotting the analyte-to-internal-standard peak area ratio against nominal concentrations using weighted (1/x²) least-squares linear regression. Parallelism between the surrogate matrix and pooled human plasma was further assessed by comparing eight-point calibration curves prepared over the same concentration range using the same regression model [8]. Sensitivity was evaluated by comparing the mean response of five LLOQ samples with that of zero samples within each inter-day batch. Carryover was assessed in five replicates by calculating the deviation in the responses of zero samples injected before and after the upper limit of quantification (ULOQ) sample.

Accuracy and precision were evaluated using QC samples at four concentration levels prepared in the surrogate matrix. Intra-day performance was assessed using five replicates within a single analytical run, and inter-day performance was evaluated across three independent analytical runs. Dilution integrity was assessed using dilution QC samples prepared above the ULOQ and diluted 10- and 50-fold with surrogate matrix before analysis. Matrix effects and extraction recovery were evaluated at low, medium, and high QC levels using the post-extraction spike approach [10]. Stability was assessed under bench-top, freeze–thaw, long-term storage, and autosampler conditions. Bench-top stability was evaluated after 4 hours at room temperature, freeze–thaw stability after three cycles, long-term stability after storage at −80°C for 3 days, and autosampler stability after 36 hours at 4°C. Acceptance criteria were defined as ±15% for accuracy (±20% at LLOQ) and ≤ 15% coefficient of variation for precision (≤ 20% at LLOQ).

Clinical application

A total of 30 human plasma samples obtained from 15 healthy male adult volunteers were analyzed using the developed LC–MS/MS method. The samples were collected as part of a phase I clinical trial conducted at Seoul National University Bundang Hospital (Seongnam, Korea). The study protocol was approved by the Korean Ministry of Food and Drug Safety and the Institutional Review Board of Seoul National University Bundang Hospital (IRB No. B-2004-606-003), and the trial was registered at ClinicalTrials.gov (identifier: NCT04335526). During the metformin-alone treatment period, participants received oral metformin tablets at a total dose of 2,000 mg. On Day 1, 500 mg was administered in the afternoon and another 500 mg in the evening. On Day 2, 1,000 mg was administered at 09:00 under fasting conditions. Paired plasma samples collected before the first metformin dose on Day 1 and 2 hours after the final dose on Day 2 were analyzed. Blood samples were collected into EDTA-containing tubes, centrifuged to separate plasma, and stored at −80°C until analysis.

Statistical analyses

The concentrations of Lac-Phe and Lac-Tyr were quantified using Agilent MassHunter software (Agilent Technologies Inc., Santa Clara, CA, USA). Statistical analyses were performed using R statistical computing software (version 4.4.1; R Foundation for Statistical Computing, Vienna, Austria). Differences in plasma Lac-Phe and Lac-Tyr concentrations before and after metformin administration were evaluated using a paired t-test. A two-sided p value of < 0.05 was considered statistically significant.

RESULTS

Selectivity, sensitivity, and linearity

Selectivity was evaluated using blank, zero, and LLOQ samples prepared in the processed surrogate matrix. No significant interfering peaks were observed at the retention times of Lac-Phe, Lac-Tyr, or the IS in blank and zero samples, demonstrating adequate selectivity of the developed assay (Fig. 1). The validated LLOQ was 0.5 ng/mL for both Lac-Phe and Lac-Tyr, which met the predefined criteria for accuracy and precision. Sensitivity was further confirmed by comparing the responses of LLOQ samples with those of zero samples in each validation batch. Calibration curves for Lac-Phe and Lac-Tyr were then constructed over the concentration range of 0.5–100 ng/mL using weighted (1/x²) linear regression. Both analytes showed excellent linearity, with coefficients of determination (r² ) of 0.9988 for Lac-Phe and 0.9989 for Lac-Tyr (Fig. 2). Parallelism between the surrogate matrix and pooled human plasma was further assessed by comparing eight-point calibration curves prepared over the same concentration range. The relative slope differences were 8.8% for Lac-Phe and 15.4% for Lac-Tyr. Based on the approximately 15% criterion previously used in the bioanalytical literature, the Lac-Phe curves showed comparable slopes, whereas the Lac-Tyr result was marginally above this value (Supplementary Fig. 1).

Figure 1. Representative multiple reaction monitoring chromatograms of Lac-Phe and Lac-Tyr in processed surrogate matrix. Chromatograms are shown for (A) blank, (B) zero, and (C) LLOQ samples. Blank samples were prepared without analyte or internal standard spiking; zero samples contained only the internal standard (Lac-Phe-d₅); and LLOQ samples contained Lac-Phe and Lac-Tyr at the lower limit of quantification together with the internal standard (Lac-Phe-d₅). Retention time and peak area are annotated in each chromatogram.

Lac-Phe, N-lactoyl-phenylalanine; RT, retention time; Lac-Tyr, N-lactoyl-tyrosine; LLOQ, lower limit of quantification.

Figure 2. Calibration curves and QC samples for (A) Lac-Phe and (B) Lac-Tyr. Black dots represent calibration standards prepared in the surrogate matrix at concentrations of 0.5, 1, 5, 10, 20, 50, 80, and 100 ng/mL. Red triangles represent QC samples prepared in the surrogate matrix at concentrations of 0.5, 1.5, 30, and 75 ng/mL.

STD, calibration standard; Lac-Phe, N-lactoyl-phenylalanine; Lac-Tyr, N-lactoyl-tyrosine; QC, quality control.

Accuracy, precision, and dilution integrity

The intra- and inter-day accuracy and precision results for Lac-Phe and Lac-Tyr are summarized in Table 1. Intra-day precision ranged from 1.62% to 8.79%, whereas inter-day precision ranged from 0.34% to 3.22% across all QC levels. Accuracy ranged from 88.02% to 101.42%, satisfying the predefined validation criteria. These findings indicate that the method provides reproducible and reliable quantification over the validated concentration range. Dilution integrity was also assessed using samples prepared above the ULOQ and diluted 10- and 50-fold with surrogate matrix before analysis. The diluted samples met the acceptance criteria for both accuracy and precision, confirming that samples exceeding the calibration range could be accurately quantified after appropriate dilution.

Table 1. Intra-day and inter-day precision and accuracy of Lac-Phe and Lac-Tyr in surrogate matrix.

Compound name	Levels	Nominal concentration (ng/mL)	Intra-day (n = 5)			Inter-day (n = 3)
			Observed concentration (ng/mL)	Precision (%)	Mean accuracy (%)	Observed concentration (ng/mL)	Precision (%)	Mean accuracy (%)
Lac-Phe	LLOQ	0.5	0.47 ± 0.02	4.75	94.64	0.49 ± 0.02	3.22	97.68
	LQC	1.5	1.36 ± 0.09	6.56	90.59	1.37 ± 0.02	1.18	91.11
	MQC	30	28.83 ± 0.57	1.99	96.11	29.38 ± 0.53	1.81	97.92
	HQC	75	74.55 ± 2.70	3.62	99.39	76.07 ± 1.61	2.12	101.42
Lac-Tyr	LLOQ	0.5	0.46 ± 0.04	8.79	91.30	0.47 ± 0.01	2.73	93.08
	LQC	1.5	1.32 ± 0.04	2.88	88.02	1.33 ± 0.03	1.91	88.92
	MQC	30	27.71 ± 0.45	1.62	92.36	28.41 ± 0.62	2.18	94.70
	HQC	75	74.29 ± 1.51	2.03	99.06	74.57 ± 0.26	0.34	99.42

Values are presented as mean ± standard deviation unless otherwise indicated.

Lac-Phe, N-lactoyl-phenylalanine; Lac-Tyr, N-lactoyl-tyrosine; LLOQ, lower limit of quantification; LQC, low quality control; MQC, medium quality control; HQC, high quality control.

Matrix effect, recovery, and stability

Matrix effects and extraction recoveries were evaluated at the LQC, MQC, and HQC levels, and the results are summarized in Table 2. For Lac-Phe, matrix effects ranged from 101.5% to 103.5%, indicating minimal ion suppression or enhancement. In contrast, Lac-Tyr showed moderate ion suppression, with matrix effects ranging from 85.4% to 89.3%. Extraction recoveries ranged from 21.0% to 23.8% for Lac-Phe and from 18.2% to 21.4% for Lac-Tyr. Although the absolute recoveries were relatively low, the values were consistent across QC levels. Stability studies further showed that both analytes remained stable under the tested freeze–thaw, bench-top, autosampler, and re-injection conditions, with all results meeting the predefined acceptance criteria (Table 3).

Table 2. Matrix effects and extraction recoveries of Lac-Phe and Lac-Tyr at three quality control levels.

Compound	Parameter	LQC	MQC	HQC
Lac-Phe	Matrix effect (%)	103.5 ± 4.2	101.5 ± 3.1	102.7 ± 2.8
	Recovery (%)	21.0 ± 1.3	21.4 ± 0.9	23.8 ± 1.1
Lac-Tyr	Matrix effect (%)	85.4 ± 3.8	85.7 ± 2.5	89.3 ± 2.2
	Recovery (%)	18.2 ± 1.2	18.2 ± 0.7	21.4 ± 0.8

Values are presented as mean ± standard deviation.

Lac-Phe, N-lactoyl-phenylalanine; Lac-Tyr, N-lactoyl-tyrosine; LQC, low quality control; MQC, medium quality control; HQC, high quality control.

Table 3. Stability of Lac-Phe and Lac-Tyr under various storage and handling conditions.

Stability condition	Levels	Lac-Phe			Lac-Tyr
		Stability (%)	CV (%)	Bias (%)	Stability (%)	CV (%)	Bias (%)
Freeze–thaw (3 cycles, n = 4)	LQC	107.44	1.36	7.44	106.09	2.37	6.09
	HQC	100.53	1.37	0.53	98.78	2.39	−1.22
Bench-top (RT, 4 h, n = 6)	LQC	104.00	4.11	4.00	102.51	6.18	2.51
	HQC	98.93	1.85	−1.07	103.60	2.83	3.60
Autosampler (4℃, 36 h, n = 4)	LQC	110.48	7.06	10.48	108.73	3.94	8.73
	HQC	106.57	3.01	6.57	103.95	2.29	3.95
Re-injection (after 24 h, n = 4)	LQC	99.35	10.86	−0.65	97.96	2.51	−2.04
	HQC	103.00	3.71	3.00	97.39	2.87	−2.61

Lac-Phe, N-lactoyl-phenylalanine; Lac-Tyr, N-lactoyl-tyrosine; CV, coefficient of variation; LQC, low quality control; HQC, high quality control; RT, room temperature.

Clinical application

The validated LC–MS/MS method was applied to plasma samples obtained from healthy volunteers (n = 15) collected before and after metformin administration. Both Lac-Phe and Lac-Tyr were successfully detected and quantified in all analyzed samples within the validated calibration range. As shown in Fig. 3, the plasma concentrations of both analytes increased significantly after metformin administration compared with pre-dose levels. These results demonstrate that the developed assay is sufficiently sensitive and robust for the analysis of endogenous lactoyl-amino acids in clinical plasma samples. The observed changes also support the potential utility of this method for investigating pharmacologically induced alterations in lactoyl-amino acid metabolism in future clinical and metabolic studies.

Figure 3. Clinical application of the validated liquid chromatography-tandem mass spectrometry method for the quantification of (A) Lac-Phe and (B) Lac-Tyr in plasma samples collected from healthy volunteers before and after metformin administration. Pre: before metformin administration; Post: after metformin administration. Each point represents an individual subject, and gray lines connect paired samples from the same subject. Box plots indicate the median and interquartile range, with whiskers representing the minimum and maximum values. Statistical significance was evaluated using a paired t-test.

Lac-Phe, N-lactoyl-phenylalanine; Lac-Tyr, N-lactoyl-tyrosine.

****p < 0.0001.

DISCUSSION

In this study, we developed and validated a sensitive and selective LC-MS/MS method for the simultaneous quantification of Lac-Phe and Lac-Tyr in human plasma. The method showed satisfactory analytical performance over the validated concentration range, with acceptable linearity, sensitivity, accuracy, precision, dilution integrity, and stability. In addition, the assay was successfully applied to plasma samples obtained from healthy volunteers before and after metformin administration, supporting its applicability to clinical studies of lactoyl-amino acid metabolism.

The quantification of endogenous metabolites in biological matrices is analytically challenging because true blank matrices are not readily available and endogenous background levels may interfere with calibration and validation. To address this issue, an artificial surrogate matrix was used for the preparation of calibration standards and QC samples in the present study [11]. This approach enabled reliable assessment of assay performance while minimizing interference from endogenous Lac-Phe and Lac-Tyr. Given the increasing interest in lactoyl-amino acids in metabolic and clinical research, the availability of a validated analytical method suitable for human plasma may facilitate further investigation of these compounds under both physiological and pharmacological conditions.

In the present method, Lac-Phe showed minimal matrix effects, whereas Lac-Tyr exhibited moderate ion suppression. Although the extraction recoveries of both analytes were relatively low, they were reproducible across the tested QC levels. In bioanalytical assays, consistent and reproducible recovery is generally considered more important than high absolute recovery, provided that the method demonstrates acceptable accuracy and precision. Therefore, the observed recovery values were considered acceptable for the intended purpose of quantitative analysis. Nevertheless, the relatively low extraction recoveries indicate room for further optimization. Although the method met the predefined sensitivity, accuracy, and precision criteria at the validated LLOQ of 0.5 ng/mL, improved extraction efficiency may be beneficial if quantification below the current LLOQ is required. Together with the favorable precision and accuracy results, these findings indicate that the current sample preparation procedure was suitable for the simultaneous analysis of Lac-Phe and Lac-Tyr in plasma.

The validated assay was further applied to clinical plasma samples collected before and after metformin administration. Both Lac-Phe and Lac-Tyr concentrations increased significantly after metformin administration. The increase in Lac-Phe is consistent with recent evidence suggesting that Lac-Phe may be involved in the metabolic effects of metformin [4,5]. More recently, untargeted metabolomics in patients with type 2 diabetes identified multiple N-lactoyl-amino acids, including Lac-Phe and Lac-Tyr, as dose-responsive plasma biomarkers of metformin exposure and adherence [7]. In contrast, the biological significance of the increase in Lac-Tyr remains less well characterized. Although the present study was not designed to investigate the mechanistic basis of these changes, the observed results demonstrate that the developed method is sufficiently sensitive and robust to detect pharmacologically induced alterations in circulating lactoyl-amino acid levels.

This study has several limitations. First, the clinical applicability of the method was evaluated in a relatively small number of healthy volunteers. Second, the biological interpretation and detailed pharmacokinetic relationships of the metformin-associated changes in Lac-Phe and Lac-Tyr were beyond the primary scope of this analytical validation study. Serial blood sampling for metformin pharmacokinetic assessment was not performed, and only one pre-dose and one post-dose sample were available; therefore, concentration–time profiles, pharmacokinetic parameters, and exposure–response relationships could not be evaluated. Third, the additional comparison between the surrogate matrix and plasma was based on one pooled plasma lot and did not include independently prepared plasma QCs quantified against the surrogate-matrix calibration curve. Further evaluation using multiple individual plasma lots and plasma QCs would strengthen the validation of the surrogate-matrix approach. Fourth, Lac-Phe-d₅ was used as the IS for both analytes, which may not fully compensate for matrix-dependent effects on Lac-Tyr. Future studies using an analyte-specific IS, such as Lac-Tyr-d₄, may further improve the robustness of Lac-Tyr quantification. Despite these limitations, the present method provides a reliable analytical tool for the simultaneous quantification of Lac-Phe and Lac-Tyr in human plasma and may be useful for future clinical pharmacology and metabolic studies.

SUPPLEMENTARY MATERIAL

Supplementary Figure 1

Comparison of calibration responses between the surrogate matrix and human plasma. Eight-point calibration curves for (A) Lac-Phe and (B) Lac-Tyr were prepared in 40 mg/mL bovine serum albumin in phosphate-buffered saline and [a single-donor human plasma lot/a pooled human plasma lot] over the concentration range of 0.5–100 ng/mL. The curves were fitted using linear regression with 1/x² weighting. The absolute slope difference was calculated as |slope_plasma − slope_surrogate| / slope_surrogate × 100.