Development and Validation of a UPLC-MS/MS Method for the Quantification of Amantadine in Rat Plasma: Application to a Pharmacokinetic Study Under High-Altitude Hypoxia and Mechanistic Insights

Abstract

Background/Objectives: This study aimed to develop an ultra-performance liquid chromatography–tandem mass spectrometry (UPLC-MS/MS) method for quantifying amantadine (AMA) in rat plasma and to investigate its pharmacokinetics under simulated high-altitude hypoxia, contrasting its behavior with that of its structural analog memantine (MEM). Methods: The method entailed using memantine (MEM) as an internal standard. Sample preparation involved protein precipitation, followed by gradient elution with detection via positive electrospray ionization and selective reaction monitoring (SRM). The method validation complied with the International Conference on Harmonization (ICH) M10 guidelines. Pharmacokinetic studies were conducted in rats exposed to either low altitude (1500 m) or simulated high altitude (6500 m) after a single oral dose of AMA (10 mg/kg). Results: The assay demonstrated linearity from 5 to 1000 µg/L, with accuracy, precision, recovery, and stability all meeting the respective acceptance criteria. Hypoxia did not significantly alter systemic exposure to AMA, as measured by parameters such as the area under the concentration–time curve (AUC), maximum concentration (Cmax), and apparent clearance (CLz/F). However, hypoxia prolonged the elimination half-life by 55% and increased the variance in the mean residence time. This finding contrasts sharply with our previous results on MEM under identical hypoxic conditions, which showed a 72.15% increase in AUC and a 41.99% decrease in CLz/F. Conclusions: A robust UPLC-MS/MS method for quantifying AMA was successfully established. AMA exhibits unique pharmacokinetic resilience to acute hypoxia, characterized by increased variability in elimination without changes in overall exposure. This profile starkly differs from the heightened exposure and reduced clearance observed for drugs like MEM, which are predominantly cleared by hepatic metabolism (under the studied conditions). These findings are consistent with the concept that a drug’s primary elimination pathway (renal excretion vs. hepatic metabolism) critically determines its pharmacokinetic susceptibility to hypoxic stress.

1. Introduction

Amantadine (AMA) is an antiviral and anti-Parkinsonian drug with a well-established pharmacokinetic (PK) profile, characterized by minimal hepatic metabolism and predominant renal excretion of the unchanged drug [1,2]. Since its introduction, numerous analytical methods, including several liquid chromatography–tandem mass spectrometry (LC-MS/MS) assays, have been developed for quantification in biological matrices, primarily human plasma, offering high sensitivity and selectivity [3,4,5]. These methods typically achieve lower limits of quantitation that are sufficient for clinical PK monitoring.

Environmental extremes, such as high-altitude hypoxia, are known to significantly modulate the absorption, distribution, metabolism, and excretion (ADME) of drugs, potentially altering their efficacy and safety profiles [6,7,8,9]. Hypoxia can influence hepatic blood flow and metabolic enzyme activity (e.g., CYP450), renal perfusion and glomerular filtration rate (GFR), plasma protein binding, and gastrointestinal function [10,11,12]. Consequently, PK studies under hypoxic conditions are crucial for dose optimization in populations residing at or traveling to high altitudes. Ultra-performance liquid chromatography–tandem mass spectrometry (UPLC-MS/MS) remains the gold standard for such bioanalytical investigations owing to its unparalleled sensitivity, specificity, and throughput [13,14,15].

Despite the known effect of hypoxia on drug disposition, its specific impact on the PK of AMA remains unexplored. This gap is noteworthy, given AMA’s primary reliance on renal elimination, a process susceptible to hypoxia-induced reductions in renal blood flow and GFR [16,17]. Furthermore, investigating the PK of AMA under hypoxic conditions presents an opportunity to examine the differential effects of this stressor on drugs with distinct ADME properties. For instance, our parallel research on memantine (MEM), a structural analog of AMA used in Alzheimer’s disease, under identical hypoxic conditions revealed profound increases in systemic exposure: the area under the concentration–time curve (AUC) and maximum concentration (Cmax) increasing by 72.15% and 131.52%, respectively, while the apparent clearance (CLz/F) decreased by 41.99% [18]. These changes starkly contrast with the minimal alterations observed in the primary PK parameters of AMA in the present study, prompting a deeper inquiry into the underlying mechanisms. Recent reviews and studies on sedative–hypnotics under hypoxia have consistently reported that drugs metabolized by Cytochrome P450 (CYP450) enzymes, such as midazolam (a CYP3A4 substrate), exhibit significantly increased AUC, prolonged half-life, and reduced clearance owing to hypoxia-induced downregulation of metabolic enzymes [19,20]. The differential response between AMA and MEM (or other central nervous system drugs) underscores the hypothesis that a drug’s primary elimination route dictates its PK sensitivity to hypoxia.

Therefore, the objectives of this study were to (1) develop and validate a rapid, sensitive, and robust UPLC-MS/MS method for the quantification of AMA in rat plasma, addressing the need for a reliable assay in preclinical hypoxia research; (2) apply this method to characterize the PK of AMA in a validated rat model of acute high-altitude hypoxia (simulating 6500 m); and (3) compare and mechanistically interpret the differential PK responses of AMA and its analog MEM to hypoxic stress from an ADME perspective, thereby enhancing the understanding of hypoxia–PK relationships and providing insights for precise dosing in high-altitude populations.

2. Results

2.1. LC-MS/MS Optimization

The structures of AMA and MEM are presented in Figure 1. Quantification of AMA in rat plasma was performed using an internal standard (IS) calibration method with MEM as a structural analogue IS. Analysis was conducted by selective reaction monitoring (SRM). High-intensity and selective ion transitions were optimized for both the analyte and the IS, along with corresponding MS/MS parameters, including collision energy (CE) and tube lens offset. The quantitative SRM transitions were m/z 152.1 → 135.2 for AMA and m/z 180.1 → 163.2 for MEM. For confirmation, qualifying transitions were monitored at m/z 152.1 → 93.2 (AMA) and m/z 180.1 → 107.2 (MEM). All optimized MS parameters are summarized in

Table 1. Mass spectrometry parameters for amantadine (AMA) and memantine (MEM).

Analytes	Precursor Ion (amu)	Product Ion (amu)	CE (eV)	Tube Lens (V)
AMA	152.1	135.2	16	66
	152.1	93.2	28
MEM	180.1	163.2	15	58
	180.1	107.2	25

CE, collision energy.

.

2.2. Method Validation Results

2.2.1. Results of Selectivity, Specificity, and Carry-Over

This method was optimized for speed and sensitivity. Positive electrospray ionization (ESI) provided strong ionization for both compounds. Using an optimized gradient, baseline separation of AMA (retention time 2.09 min) and MEM (retention time 2.27 min) was achieved within a 3.5 min total run time, requiring only 50 µL of plasma per sample. Selectivity was confirmed by the absence of significant interfering peaks from endogenous substances at the retention times of AMA and MEM in blank rat plasma. Representative chromatograms for the quantitative and qualitative ions are shown in Figure 2 and Figure 3, respectively. No significant carry-over was observed in the blank samples injected immediately after the upper limit of quantitation (ULOQ).

2.2.2. Linearity and Lower Limit of Quantification (LLOQ) Results

The calibration curve for AMA in rat plasma demonstrated linearity over the range of 5–1000 µg/L. A weighted (1/x) linear regression model was applied with the regression line forced through the origin (Y = a × X). Three independent validation runs were performed to assess the reproducibility and robustness of the calibration model. This model yielded correlation coefficients (r²) greater than 0.995 across all runs. The regression equations and correlation coefficients are summarized in

Table 2. Linear regression equations and correlation coefficients for the amantadine calibration curve.

Analytes	IS	Regression Equation	R2
AMA	MEM	Y = 0.0125069 × X	0.9997
		Y = 0.0122042 × X	0.9996
		Y = 0.0123832 × X	0.9994

. The LLOQ (5 μg/L) was confirmed with both accuracy and precision within ±20%. The signal-to-noise ratio for the quantitative ion transition was well above 10, ensuring reliable quantification at all sampling time points, including the 24 h time point post-administration. A limit of detection was not determined, as the validated LLOQ of 5 μg/L was sufficient for the purposes of this pharmacokinetic study.

2.2.3. Accuracy and Precision Results

The accuracy and precision of the method were assessed by analyzing replicates at the LLOQ and at low-, medium-, and high-quality control (QC) concentrations. As presented in

Table 3. Intra-day and inter-day accuracy and precision for the determination of amantadine in rat plasma.

Nominal Concentration (μg/L)	Intra-Day (n = 6)			Inter-Day (3 Days, n = 18)
	Measured Concentration (μg/L)	Accuracy (%)	Precision (CV, %)	Measured Concentration (μg/L)	Accuracy (%)	Precision (CV, %)
5.0	5.62 ± 0.08	112.38	1.42	5.54 ± 0.25	110.72	4.51
15.0	15.07 ± 0.85	100.47	5.64	15.04 ± 0.81	100.25	5.39
150.0	143.33 ± 1.30	95.55	0.91	147.69 ± 7.54	98.46	5.11
750.0	737.84 ± 14.03	98.38	1.90	757.67 ± 23.44	101.02	3.09

, both intra-day and inter-day accuracy met the acceptance criteria, with mean deviations within ±15% of the nominal values (±20% for LLOQ). The precision, expressed as the coefficient of variation (CV), was ≤15% for all QC levels (20% for LLOQ). These results demonstrate that the method is reliable, reproducible, and suitable for the quantitative analysis of AMA in rat plasma.

2.2.4. Results for Recovery and Matrix Effect

As summarized in

Table 4. Extraction recovery and matrix effect for amantadine in rat plasma (n = 6).

Nominal Concentration (μg/L)	IS Normalized Recovery (%)	CV of IS Normalized Recovery (%)	IS Normalized Matrix Factor (%)	CV of IS Normalized Matrix Factor (%)
15.0	99.44 ± 5.28	5.31	94.16 ± 3.17	3.37
150.0	94.61 ± 3.47	3.67	88.50 ± 3.28	3.71
750.0	99.85 ± 3.62	3.63	92.13 ± 1.40	1.52

, the extraction recoveries were high and consistent, ranging from 94.61% to 99.85% across the QC levels, which demonstrates the efficiency and robustness of the protein precipitation method. The IS-normalized matrix factors for AMA ranged from 88.5% to 94.16%, with associated CVs below 3.70%. This indicates that ion suppression was minimal and consistent across different plasma lots.

2.2.5. Stability Test Results

The stability of AMA in rat plasma under various storage and handling conditions is summarized in

Table 5. Stability of amantadine in rat plasma under various conditions (n = 6).

Times	Nominal Concentration (μg/L)	RT (%)	+4 °C (%)	−80 °C (%)	RFT (%)
24 h	15.0	2.04	-	-	-
	150.0	−3.37	-	-	-
	750.0	3.33	-	-	-
48 h	15.0	-	3.23	-	-
	150.0	-	5.63	-	-
	750.0	-	2.77	-	-
7d	15.0	-	-	2.71	−2.23
	150.0	-	-	4.88	2.02
	750.0	-	-	3.21	2.22
21d	15.0	-	-	7.64	-
	150.0	-	-	9.95	-
	750.0	-	-	5.51	-

RT, room temperature; RFT, repeatedly frozen and thawed; -, not applicable.

. The mean measured concentrations for all stability tests were within ±10% of the nominal values. Specifically, AMA was stable in processed samples for 24 h at ambient temperature and for 48 h in the autosampler at 4 °C. The analyte exhibited satisfactory stability in plasma when stored at −80 °C for 7 and 21 days. Furthermore, the AMA plasma samples remained stable over three complete freeze–thaw cycles conducted between −80 °C and room temperature. Collectively, these results confirm that AMA is stable under the typical sample handling and storage conditions encountered in this study.

2.2.6. Dilution Integrity Results

The 10-fold dilution of a sample with a concentration exceeding the ULOQ (10,000 μg/L) resulted in an accuracy between 96.28% and 110.5% and a precision between 2.15% and 4.62%, confirming the integrity of the dilution procedure.

2.3. Pharmacokinetic Analysis

The validated method was applied to determine the plasma concentration of AMA in rats following oral administration under different simulated altitude conditions.

The mean plasma concentration–time profiles of AMA after a single oral dose (10 mg/kg) in the low-altitude control (LAC) and high-altitude exposure (HAE) groups are shown in Figure 4. The profiles were largely superimposable, suggesting a minimal overall impact of acute hypoxic exposure on systemic AMA exposure.

Key pharmacokinetic parameters are summarized in

Table 6. Key pharmacokinetic parameters of amantadine.

Parameter	Unit	LAC Group	HAE Group	p-Value
AUC(0–t)	μg/L×h	2791.02 ± 674.13	2951.26 ± 734.64	0.656
Cmax	μg/L	715.28 ± 157.19	776.09 ± 196.68	0.506
Tmax	H	1.22 ± 0.49	0.94 ± 0.48	0.264
t1/2z	H	2.06 ± 0.58	3.20 ± 1.11	0.021 *
λz	1/h	0.36 ± 0.08	0.24 ± 0.09	0.019 *
Vz/F	L/kg	10.86 ± 2.60	17.20 ± 10.29	0.113
CLz/F	L/h/kg	3.74 ± 0.79	3.62 ± 1.23	0.826
MRT(0–t)	H	3.36 ± 0.73	3.50 ± 0.37	0.645
MRT(0–∞)	H	3.39 ± 0.78	3.66 ± 0.52	0.424
VRT(0–t)	h^2	9.11 ± 3.95	13.69 ± 4.43	0.046 *
VRT(0–∞)	h^2	9.76 ± 5.28	18.00 ± 9.12	0.044 *

AUC, area under the concentration–time curve; Cmax, maximum concentration; Tmax, time to maximum concentration; t_1/2z, terminal elimination half-life; λz, terminal elimination rate constant; Vz/F, apparent volume of distribution; CLz/F, apparent clearance; MRT, mean residence time; VRT, variance of residence time. Data are presented as mean ± SD (n = 8). * p < 0.05 for LAC vs. HAE.

. Notably, the key exposure parameters—AUC(0–t), AUC(0–∞), and Cmax—did not differ significantly between the LAC and HAE groups. Similarly, the CLz/F and time to maximum concentration (Tmax) remained unchanged. However, significant alterations were observed in parameters related to the terminal elimination phase and residence time distribution: the terminal elimination half-life (t_1/2z) was prolonged by 55% (from 2.06 ± 0.58 h to 3.20 ± 1.11 h, p = 0.021), and the variance of residence time (VRT) increased significantly (VRT(0–∞) from 9.76 ± 5.28 h² to 18.00 ± 9.12 h², p = 0.044). Thus, while the trend toward prolonged elimination is evident, the precise magnitude of these terminal phase parameters should be interpreted with caution. The apparent volume of distribution (Vz/F) showed a trend toward an increase (58% increase, p = 0.113) in the HAE group.

3. Discussion

This study establishes a fully validated UPLC-MS/MS method for quantifying AMA in rat plasma. Although the methodology is conventional, the rigorous validation according to ICH M10 guidelines and the specific application to a pharmacokinetic study under hypoxic conditions address an existing research gap. Compared with previously reported LC-MS/MS methods for AMA [3,4,5,21,22], the present method offers a shorter run time (3.5 min vs. typically 5–10 min), requires a smaller plasma volume (50 µL vs. often 100–200 µL), and employs a simpler protein precipitation protocol, making it more suitable for high-throughput preclinical PK studies under challenging conditions such as hypoxia.

The most significant finding is the differential impact of acute high-altitude hypoxia on the PK of AMA compared to its structural analog, MEM—studied under identical conditions in our parallel work [18]—and to other centrally acting drugs reported in the literature [19,20,23]. While MEM exhibited marked increases in AUC (72%) and Cmax (132%), accompanied by decreases in CLz/F (42%) and Vz/F (40%), AMA showed no significant changes in these primary exposure and clearance parameters. This clear divergence provides a unique opportunity for a comparative mechanistic discussion from an ADME perspective, which integrates recent insights on the effects of hypoxia.

• Absorption: Both drugs were well absorbed. Hypoxia-induced gastrointestinal changes (e.g., altered motility, blood flow, and pH) could theoretically affect drug absorption [24,25,26]. The lack of change in Tmax and Cmax of AMA suggests its absorption was largely unperturbed in this acute hypoxic model. In contrast, the significantly higher Cmax of MEM may indicate an enhanced absorption rate or extent under hypoxia, possibly due to its different physicochemical properties or susceptibility to hypoxia-altered intestinal transporters—a hypothesis that requires further investigation.
• Distribution: The response of Vz/F to hypoxia differed between the two drugs; while MEM exhibited a significant decrease, AMA showed only a non-significant increasing trend (p = 0.113). This discrepancy could be attributed partly to their differing lipophilicity (LogP ~2.44 for AMA vs. ~3.28 for MEM), which influences the baseline tissue distribution and may modulate sensitivity to hypoxia-induced changes in perfusion or membrane permeability [27,28]. Furthermore, differences in tissue-binding characteristics and distribution kinetics—potentially including varied sensitivity of blood–brain barrier penetration to hypoxic stress—might also contribute to the opposing trends in Vz/F, although this requires direct confirmation in future studies. Hypoxia-induced alterations in drug distribution are compound-specific. While many central nervous system drugs, such as midazolam and phenytoin, show significantly reduced brain-to-plasma ratios under chronic hypoxia due to upregulated efflux transporters (e.g., P-glycoprotein (P-gp)) at the blood–brain barrier [20,29], the distribution of amantadine appears to be governed by different mechanisms. In this study, amantadine exhibited a significant increase in VRT under high-altitude exposure without marked changes in AUC or Cmax, indicating heightened inter-individual variability in its tissue distribution and/or elimination rather than a uniform shift in clearance or central penetration. This suggests that for amantadine, whose interaction with efflux transporters such as P-gp is less defined than that of classical benzodiazepines or MEM [18], hypoxia may predominantly affect physiological variables such as regional perfusion, plasma protein binding, or local pH partition, leading to more variable distribution kinetics rather than consistent directional changes in brain exposure [30].
• Metabolism: This is the key differentiator that explains contrasting PK outcomes. AMA undergoes minimal hepatic metabolism, with approximately 85% excreted unchanged in the urine. In contrast, approximately 48% of MEM is excreted unchanged, while the remainder is primarily metabolized in the liver to polar metabolites [31,32]. Consequently, systemic exposure to AMA (AUC and Cmax) and its CLz/F remained unchanged in our acute hypoxia model. The observed prolonged elimination half-life and increased variability (VRT) could be consistent with altered renal tubular transport (potentially involving transporters such as organic cation transporters (OCT2)) rather than suppressed metabolism. High-altitude hypoxia potently downregulates key drug-metabolizing enzymes, including CYP3A4, CYP1A2, and UGT1A1, via repression of the pregnane X receptor (PXR) and constitutive androstane receptor (CAR) [33]. The inhibition of its metabolic clearance, coupled with a likely reduction in renal excretion, explains the profound increase in MEM’s systemic exposure (AUC) and the decrease in its apparent clearance (CLz/F) observed under identical conditions. Similarly, diazepam (a CYP3A4/2C19 substrate) showed a 25-fold increase in AUC under hypoxia [19,23]. Therefore, the increased exposure to MEM likely stems largely from inhibited hepatic and possibly cerebral CYP450-mediated metabolism, a mechanism not applicable to AMA. These findings support the concept that hypoxia most profoundly affects the PK of drugs reliant on hepatic metabolism.
• Excretion: Both AMA and MEM are primarily renally eliminated. Hypoxia reduces renal blood flow and GFR [34], which can impair excretion. The fact that AMA’s systemic CLz/F remained unchanged suggests a possible compensatory mechanism or that the 72 h acute hypoxic insult was insufficient to significantly impair the net renal clearance of this compound in rats. However, the significantly prolonged t_1/2z and increased VRT for AMA hint at altered elimination kinetics. This could reflect increased variability in renal tubular handling rather than a consistent reduction in the net excretion rate. AMA is a substrate of renal OCT2 [35]. Hypoxia may variably affect the function or expression of these transporters among individuals, leading to increased variability (higher VRT) in the elimination process without changing the mean clearance. The unchanged AUC supports this interpretation; while the elimination process became more variable and slower in some individuals (prolonged t_1/2z), the overall extent of elimination over 24 h was similar. In contrast, for MEM, reduced metabolism coupled with potentially impaired renal excretion likely contributed synergistically to its markedly reduced CLz/F ratio and increased AUC.

Our findings align with and extend the framework proposed in recent reviews [19]. They demonstrated that, even within the same structural class (adamantane derivatives), PK responses to hypoxia can be diametrically opposed based on the dominant elimination pathway. Taken together, the PK resilience of AMA contrasts sharply with the sensitivity of MEM. This pattern supports the hypothesis that AMA, being largely metabolism-independent, is relatively resilient to hypoxia-induced changes in hepatic CYP450 function. Its PK alterations are subtle, manifesting primarily as increased variability (VRT) and a prolonged terminal phase, which are possibly linked to the variable effects of hypoxia on renal tubular transport (e.g., via OCT2). In contrast, MEM, which is metabolism-dependent, is highly susceptible to hypoxia-induced CYP450 suppression, leading to markedly increased exposure.

This study has several limitations. First, the mechanistic interpretation primarily relies on comparative PK parameters and known ADME properties, lacking direct support from measured data on urinary excretion, GFR, or the activity/expression of relevant renal transporters (e.g., OCT2) and hepatic enzymes. While the stark contrast between AMA and MEM strongly suggests that the primary elimination pathway dictates hypoxic susceptibility, conclusive evidence requires the inclusion of these additional biomarkers. Second, brain concentrations were not measured, limiting insights into whether hypoxia altered central nervous system distribution. Third, the use of an acute (72-h) hypoxia model may not fully capture adaptive changes occurring during chronic exposure. Future studies should incorporate urine collection for clearance calculations, assess specific transporter function, measure tissue drug concentrations, and employ chronic hypoxia models to delineate the precise mechanisms governing renal and metabolic drug handling at high altitude.

4. Materials and Methods

4.1. Chemicals and Reagents

The amantadine hydrochloride standard (99.0% purity; lot no. B30M11D114662) and memantine hydrochloride (98.0% purity; lot no. J21HB189391) were purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). Amantadine hydrochloride tablets (Batch No. 25A0609) were obtained from Jiangsu Pengyao Pharmaceutical Co., Ltd. (Yixing, China). Normal saline (Batch No.: 2503122) was supplied by Guangdong Otsuka Pharmaceutical Co., Ltd. (Foshan, China). Methanol and acetonitrile (both HPLC grade, 99.9% purity) were purchased from Merck Company, Inc. (Darmstadt, Germany). Formic acid (HPLC-grade) was obtained from Beijing MREDA Technology Co., Ltd. (Beijing, China). Ultrapure water was prepared using a Millipore water purification system (Billerica, MA, USA).

4.2. Instrumentation

Analyses were performed using a UPLC-MS/MS system (Thermo Fisher Scientific, Waltham, MA, USA) consisting of an UltiMate 3000 chromatograph coupled to a Quantum Access triple quadrupole mass spectrometer equipped with an ESI source. Sample preparation utilized a Microfuge^® 22R centrifuge (Beckman Coulter, Brea, CA, USA), an Eppendorf micropipette (Eppendorf SE, Hamburg, Germany), and polypropylene centrifuge tubes (Axygen; Corning Inc., Corning, NY, USA). Data acquisition, mass spectrometry optimization, and data analysis were conducted using Xcalibur (v2.2 SP1), TSQ Tune Master (v2.3.0.1214 SP3), and LCquan (v2.8.0.51) software packages (Thermo Fisher Scientific), respectively.

4.3. LC-MS/MS Analytical Conditions

Chromatographic separation was achieved using a Hypersil GOLD^TM C18 column (2.1 × 100 mm, 3.0 μm, Thermo Scientific) maintained at 25 °C. The mobile phase, delivered at a flow rate of 0.30 mL/min, consisted of 0.2% formic acid in water (solvent A) and acetonitrile (solvent B). A gradient elution program was employed as follows: 0~1.3 min, 10–90% B; 1.3~1.7 min, 90% B; 1.7~2.1 min, 90–10% B; and 2.1~3.5 min, 10% B. All analytes were eluted within 3.5 min, enabling high-throughput bioanalysis. The autosampler temperature was set at 4 °C, and the injection volume was 1 μL. The ion source temperature and spray voltage were set at 300 °C and 4.0 kV, respectively. The vaporizer temperature was set to 100 °C. Nitrogen was used as the sheath gas (35 Arb) and auxiliary gas (10 Arb). Argon was used as the collision gas at a pressure of approximately 1.5 mTorr. Quantification was performed using SRM mode. Compound-specific MS parameters are detailed in Table 1.

4.4. Preparation of Standard and Quality Control Samples

Individual stock solutions of AMA and MEM were prepared at a concentration of 1000 mg/L by dissolving the reference standards in 50% (v/v) methanol in water. Stock and working solutions were stored at −80 °C until use. Calibration standards and QC samples were prepared by spiking appropriate volumes of standard working solutions into blank rat plasma. Briefly, a standard working solution was mixed with blank rat plasma at a ratio of 1:9. The final concentrations of the plasma samples containing the analytes are listed in

Table 7. Concentrations of amantadine in calibration standards and quality control samples.

Calibration Concentration (μg/L)								QC Concentrations (μg/L)
1	2	3	4	5	6	7	8	LLOQ	Low	Medium	High
5	10	25	50	100	250	500	1000	5	15	150	750

.

4.5. Sample Preparation

A 50 μL aliquot of thawed rat plasma was accurately transferred into a 1.5 mL capped PVC centrifuge tube. Then, 150 μL of a protein precipitation solution (methanol–acetonitrile = 1:4, v/v) spiked with MEM (20 μg/L in the precipitation solvent) was added as the IS. This resulted in a final approximate MEM concentration of 15 μg/L in the processed sample mixture prior to injection. The mixture was vortexed for 30 s and then centrifuged at 14,000× g for 3 min. Subsequently, 50 μL of the supernatant was transferred to an autosampler vial with an inner liner tube, and a 1 μL aliquot was injected into the LC-MS/MS system for analysis.

4.6. Method Validation

Method validation was conducted in accordance with the International Conference on Harmonization (ICH) M10 guidelines on bioanalytical method validation [36]. The validation included assessments of selectivity, specificity, carry-over, matrix effects, calibration curve linearity, LLOQ, accuracy, precision, dilution integrity, and stability.

4.6.1. Selectivity, Specificity, and Carry-Over

Specificity and selectivity were assessed using six individual sources of blank rat plasma. Interference was considered negligible if the response in blank plasma at the retention times of the analyte and internal standard was less than 20% of the LLOQ response for the analyte and less than 5% for the IS. Carry-over was evaluated by sequentially injecting a blank sample after the ULOQ sample. The acceptance criteria for carry-over were the same as those applied for the assessment of selectivity.

4.6.2. Linearity and LLOQ

Linearity was evaluated by analyzing eight non-zero calibration standards in triplicate across three separate runs. A calibration curve was constructed by plotting the peak area ratio (AMA/IS) against the nominal concentration using a weighted (1/x) least-squares linear regression model. A correlation coefficient (r²) of ≥0.990 was required. The LLOQ was defined as the lowest concentration on the calibration curve that could be quantified with acceptable accuracy and precision (within ±20% of the nominal value). For concentrations above the LLOQ, the deviation from the nominal value had to be within ±15%. During validation, the LLOQ was confirmed by analyzing six replicate samples, all of which met the predefined accuracy and precision criteria.

4.6.3. Accuracy and Precision

Intra-day accuracy and precision were evaluated by analyzing six replicates of QC samples at the LLOQ and at low, medium, and high QC concentrations in a single run. Inter-day accuracy and precision were determined by analyzing the same QC levels over three independent runs. Accuracy was calculated as the percentage deviation of the mean measured concentration from the nominal concentration, and precision was expressed as the CV. The acceptance criteria required accuracy and precision to be within ±15% for all QC levels (except ±20% for LLOQ).

4.6.4. Recovery and Matrix Effect

The matrix effect was evaluated at low, medium, and high QC levels using six different lots of blank rat plasma. The matrix factor (MF) for the analyte and the IS was calculated as the peak area in the presence of matrix divided by the peak area in neat solution. The IS-normalized matrix factor (IS-MF) of the analyte was then derived by dividing the analyte MF by the IS MF. The CV of the IS-MF across the six matrix batches was required to be ≤15%. Recovery was determined by comparing the peak responses of samples spiked with the analyte before extraction to those spiked after extraction.

4.6.5. Stability

Analyte stability in plasma was evaluated using three QC levels under four different storage conditions. Freeze–thaw stability was assessed by subjecting the samples to three full freeze–thaw cycles (from −80 °C to 25 °C). Short-term stability was then determined by analyzing QC samples after storage at 25 °C for 24 h. Long-term stability was determined following storage of plasma samples at −80 °C for 7 and 21 days. Post-preparative stability was investigated by analyzing QC samples stored in the autosampler at 4 °C for 48 h after extraction. Stability was deemed acceptable if the mean measured concentration was within ±15.0% of the nominal value.

4.6.6. Dilution Effects

Using blank rat plasma as the diluent, dilution integrity was evaluated by diluting a sample with a concentration of 10,000 μg/L (above the ULOQ). The accuracy and precision of the diluted samples were required to be within ±15%.

4.7. Pharmacokinetics Study Under Simulated High-Altitude Conditions

4.7.1. Animal Group Allocation and High-Altitude Exposure Protocol

Sixteen healthy male Sprague Dawley rats (200 ± 20 g, 8–9 weeks old) were supplied by Lanzhou Veterinary Research Institute (license no. SCXK (Gan) 2020-0002). After one week of acclimatization, the rats were randomly divided into two groups (n = 8 per group): an LAC group (simulating 1500 m, approximating Lanzhou’s local altitude) and an HAE group. Rats in the HAE group were housed in a computer-controlled hypobaric chamber for 72 h to simulate an altitude of 6500 m. The chamber conditions were set to a barometric pressure of 354 mmHg (equivalent to 6500 m) with 21% O₂, a temperature of 22 ± 2 °C, and a relative humidity of 50 ± 10%. This 72 h acute exposure protocol is a well-established model for inducing significant physiological adaptations to hypoxia relevant to drug PK studies [18]. The LAC group was kept at ambient pressure (~630 mmHg, 1500 m equivalent). All rats were fasted for 12 h prior to drug administration.

4.7.2. Drug Administration and Plasma Sample Collection

A single dose of AMA (10 mg/kg, calculated as the free base) was administered to all rats via gavage. The dose was prepared as a suspension in normal saline by grinding commercial amantadine hydrochloride tablets. The final concentration of AMA in saline was 2 mg/mL. Blood samples (approximately 0.2 mL) were collected from the orbital venous plexus of each rat at 0, 0.08, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 8, 12, and 24 h post-administration. Blood was immediately transferred into a 0.5 mL centrifuge tube containing 10 μL sodium heparin. Plasma was separated by centrifugation at 2000× g for 5 min and stored at −80 °C until analysis.

4.8. Statistical Analysis

Pharmacokinetic parameters were calculated using non-compartmental analysis with DAS software (version 3.2.8). Data are presented as the mean ± standard deviation (SD). Statistical comparisons between the LAC and HAE groups were performed using an unpaired Student’s t-test. Statistical significance was set at p < 0.05.

5. Conclusions

In this study, a rapid, sensitive, and robust UPLC-MS/MS method for quantifying amantadine in rat plasma was successfully developed and validated. Its application in a simulated high-altitude hypoxia model revealed that acute hypoxia does not significantly alter the overall systemic exposure (AUC, Cmax) or clearance of AMA, but increases the variability and prolongs the terminal phase of its elimination. This pattern starkly contrasts with the profound PK changes observed for its metabolically cleared analog MEM under identical conditions, and with the established pattern for CYP450-metabolized sedative–hypnotics. This study underscores the critical influence of a drug’s primary elimination pathway (renal excretion versus hepatic metabolism) as the key determinant of its pharmacokinetic susceptibility to hypoxic stress. For renally excreted drugs, such as AMA, hypoxia may introduce greater inter-individual variability in elimination rather than a systematic reduction in clearance. The developed bioanalytical method is suitable for further PK/PD studies of AMA, and the findings contribute to a more nuanced, mechanism-based understanding of drug disposition in high-altitude environments, which is essential for personalized dosing strategies.