Development and Clinical Application of a Simultaneous Liquid Chromatography–Tandem Mass Spectrometry Method for Patients with Graft-versus-Host Disease to Quantify the Plasma Concentrations of Ruxolitinib and Posaconazole

Abstract

Graft-versus-host disease (GVHD) is a clinically significant problem with high mortality that is gradually increasing. Ruxolitinib (RUX) is the only drug used for steroid-refractory GVHD treatment and is thereby crucial. Therapeutic drug monitoring of RUX may be effective because of the relationship between the plasma RUX concentration and treatment outcomes. Posaconazole (PCZ) has also been the recent focus of combined treatment with RUX owing to its pharmacokinetics. We established a simultaneous LC–tandem MS (LC–MS/MS) method and performed plasma drug concentration measurements and monitoring using clinical laboratory values for both RUX and PCZ. We also compared our technique to a simple LC–MS/MS method for clinical application. Moreover, the utility of the automated pretreatment LC–MS/MS (auto-LC–MS/MS) method was tested for further applications. The simultaneous quantification LC–MS/MS method satisfied analytical validation criteria under clinical conditions. Our method demonstrated linearity over the range of 0.3–500 ng/mL for RUX and 3–5000 ng/mL for PCZ, with intra- and inter-day precision and accuracy within ±15%. A possible correlation between plasma RUX concentration and kidney injury was observed in 1 of the 6 patients. Notably, plasma PCZ concentrations were decreased by changing the administration route. Moreover, the plasma concentration levels obtained using the auto-LC–MS/MS method were highly concordant with those obtained using the LC–MS/MS method. The validated LC–MS/MS method was found to be useful in clinical applications; thus, further research into its applications in clinical practice is desirable.

INTRODUCTION

Graft-versus-host disease (GVHD), which can occur after allogeneic hematopoietic cell transplantation (HCT), is a clinically important problem caused by an immunological attack on the recipient’s body.^1–3) Real-world data from the Center for International Blood and Marrow Transplant Research registry indicate an increasing trend in allogeneic HCT, which is a primary risk factor for GVHD.⁴⁾ GVHD can be categorized into 2 types: acute GVHD (aGVHD) and chronic GVHD (cGVHD).³⁾ In aGVHD, patients exhibit various symptoms, such as those affecting the skin, gastrointestinal tract, and liver, whereas any organ in the body can be affected by cGVHD.^1,2) More than 30% of patients with GVHD die due to disease progression.⁵⁾ Steroid treatment is the primary therapy; however, 30–40% of patients exhibit steroid-refractory disease.⁶⁾ Only 3 drugs, ruxolitinib (RUX), ibrutinib, and belumosudil, have been approved by the U.S. Food and Drug Administration (FDA) for the treatment of aGVHD or cGVHD.⁷⁾

RUX, an orally administered Janus-associated kinase 1/2 (JAK1/2) inhibitor, is the only drug approved by the FDA for the treatment of steroid-refractory aGVHD and cGVHD.^8,9) Recent real-world studies and meta-analyses have revealed the clinical efficacy and safety of RUX.^10–12) For instance, a single-center real-world study showed that RUX combined with steroids achieved overall response rates of 86 and 92% at 7 and 14 d, respectively, with acceptable adverse events, including neutropenia and cytomegalovirus infection.¹⁰⁾ An individual patient data meta-analysis in pediatric patients reported overall response rates of 74% for aGVHD and 78% for cGVHD, with manageable toxicity.¹¹⁾ Furthermore, maintenance therapy with RUX after allogeneic hematopoietic stem cell transplantation has been associated with a lower incidence of cGVHD.¹²⁾

RUX is mainly metabolized by CYP3A4, which is an enzyme that contributes to the metabolism of >30% of clinically used drugs.^9,13) Various CYP3A4 inducers and inhibitors influence drug–drug interactions (DDIs);^14–16) therefore, polypharmacy management is important for more efficient treatment. Antifungals, such as itraconazole, fluconazole, voriconazole, and posaconazole (PCZ), are often administered during GVHD treatment to prevent infections; however, the DDIs among them should be considered owing to their potential as CYP3A4 inhibitors.^17–19) Among triazole-containing azoles, PCZ has been reported to exhibit particularly strong inhibition, with an in vitro half-maximal inhibitory concentration of 2–8 μM.²⁰⁾ In clinical studies, the area under the curve of simvastatin and midazolam increased approximately 5–11- and 3–6-fold, respectively, when co-administered with 50–200 mg/d of PCZ.¹⁹⁾ According to the updated guidelines of the European Conference on Infections in Leukemia, PCZ is strongly recommended for use in GVHD; thus, the combined use of RUX and PCZ can be considered in clinical practice.²¹⁾

To improve treatment efficacy and safety, therapeutic drug monitoring (TDM) of certain drugs is performed using LC–tandem MS (LC–MS/MS) or enzyme-linked immunosorbent assay.^22,23) Recently, the correlation between plasma RUX and PCZ concentrations and adverse events was evaluated in several studies.^24–26) In particular, the TDM guidelines from the British Society for Medical Mycology recommend a target trough PCZ concentration of 700 ng/mL for prophylaxis and 1000 ng/mL for patients with an established infection.²⁴⁾ Therefore, plasma concentrations of RUX and PCZ as clinical laboratory values would be useful for clinical care management.

In clinical practice, a minimally invasive method for measuring plasma drug levels is preferred because pretreatment processing requires considerable time and depends on the number of samples, thereby resulting in excessive time consumption for the measurer.^27,28) Recently, the utility of an automated pretreatment LC–MS/MS (auto-LC–MS/MS) method was reported.²⁹⁾ This method allowed TDM of 4 drugs (voriconazole, sunitinib, clozapine, and mycophenolic acid) through automated pretreatment procedures. Existing LC–MS/MS methodologies have mainly focused on the quantification of drugs within the same pharmacological class, and few reports have addressed the simultaneous determination of drugs with different mechanisms of action. Furthermore, the incorporation of automated sample preparation into clinical TDM workflows has been rarely reported. Therefore, the development of a method capable of simultaneously quantifying RUX and PCZ with automated pretreatment is expected to be of particular value in clinical practice.

In this study, a simultaneous LC–MS/MS method for the quantification of RUX and PCZ was developed and validated for clinical applications. The concentrations of RUX alone or in combination with PCZ were measured in the plasma obtained from patients who were administered these drugs. Moreover, the correlation between plasma RUX concentrations and clinical laboratory values was explored in patients. The utility of the auto-LC–MS/MS method was tested to evaluate its versatility for various drugs.

MATERIALS AND METHODS

Chemicals and Reagents

The following chemicals were purchased from the listed sources: RUX (Selleck Chemicals, Houston, TX, U.S.A.); PCZ (Alsachim, Illkirch Graffenstaden, France); ruxolitinib-²H₉ (RUX-d9; Toronto Research Chemicals, Vaughan, Toronto, Canada); and posaconazole-²H₇ (PCZ-d7; Sussex Research Laboratories, San Diego, CA, U.S.A.). Six lots of human plasma (each, heparin sodium) from different donors were purchased from Cosmo Bio Company, Limited (Tokyo, Japan) and used as blank plasma in this study. All the other chemicals and reagents used were of the highest commercially available quality.

Preparation of Calibration Standards, Quality Controls, and Internal Standard Solutions

Stock solutions of 1 mg/mL RUX, 100 μg/mL PCZ, 2 mg/mL RUX-d9, and 100 μg/mL PCZ-d7 were prepared in methanol and stored at −20°C. Stock solutions were mixed and diluted with human plasma to prepare the calibration standards (CSs) and quality controls (QCs), as listed in Table 1. The upper limits of the CS ranges were determined based on the expected maximum plasma concentrations reported in clinical practice,^30–32) whereas the lower limits were set according to the lower limit of quantification (LLOQ). Internal standard (IS) solution containing 20 ng/mL RUX-d9 and 1000 ng/mL PCZ-d7 in methanol was prepared from each stock solution.

Table 1. Concentration of Each Analyte in the Calibration Standard and Quality Control Samples

	Ruxolitinib (ng/mL)	Posaconazole (ng/mL)
CS1	0.3	3
CS2	1	10
CS3	3	30
CS4	10	100
CS5	30	300
CS6	100	1000
CS7	300	3000
CS8	500	5000
LLOQ	0.3	3
LQC	1	10
MQC	70	700
HQC	400	4000

CS: calibration standard; LLOQ: lower limit of quantification; LQC: low-quality control; MQC: medium-quality control; HQC: high-quality control.

Sample Preparation and Validation

The plasma samples (20 μL) were added to 80 μL of ice-cold IS solution, thereby inducing protein precipitation. The mixtures were then centrifuged at 20000 × g for 5 min at 4°C, and 10 μL of the supernatant was injected into the LC–MS/MS system. RUX and PCZ contents were measured in positive-ion detection mode using the multiple reaction monitoring (MRM) mode. An LCMS-8050 triple quadrupole mass spectrometer connected to a Nexera X2 UHPLC system (Shimadzu, Kyoto, Japan) was used for LC–MS/MS analysis. Chromatogram separation was performed using a CAPCELL PAK INERT ADME-HR column (2.0 mm i.d. × 50 mm, 3 μm; Osaka Soda, Osaka, Japan) maintained at 40°C. The mobile phases were prepared using deionized water containing 0.1% formic acid and methanol as eluents A and B, respectively. The following gradient program was used: elution was initiated using 50% B for 0.5 min, followed by a linear gradient to 98% B from 0.5 to 4 min, held at 98% B for 0.5 min, and then immediately returned to the initial conditions at a flow rate of 0.5 mL/min. Analysis was performed using LabSolutions (Shimadzu), in which ion MRM transitions were monitored (m/z, 306.95 → 186.10; Q1 pre bias (Q1), 15 V; collision energy, 27 V; Q3 pre bias (Q3), 20 V for RUX; m/z, 315.95 → 186.10; Q1, 16 V; CE, 28 V; Q3, 19 V for RUX-d9; m/z, 701.30 → 127.10; Q1, 5 V; CE, 95 V; Q3, 24 V for PCZ; m/z, 708.30 → 127.10; Q1, 5 V; CE, 73 V; Q3, 22 V for PCZ-d7). The optimized MS conditions were as follows: probe voltage, 4000 V; desolvation line temperature, 250°C; block heater temperature, 400°C; interface temperature, 300°C; nebulizing gas flow, 2.5 L/min; drying gas, 10 L/min; and heating gas flow, 10 L/min.

The LC–MS/MS method was validated according to the U.S. FDA guidelines for bioanalytical method validation, including accuracy, precision, matrix effect, stability, dilution integrity, carryover, and selectivity (https://www.gmp-compliance.org; accessed January 19, 2024). The calibration curves for each analyte were evaluated over the concentration range between CS1 and CS8. Each calibration point was independently measured 3 times. To determine matrix effects, blank plasma samples derived from 6 different individuals containing neither analyte nor IS and containing middle-quality control (MQC) samples with IS were prepared and analyzed. The intra- and inter-day precision and accuracy (N = 6 and N = 18, respectively) were evaluated by measuring the QC samples at the 4 different concentrations listed in Table 1 (LLOQ; low-quality control, LQC; MQC; and high-quality control, HQC). The stability of each analyte in plasma was assessed after incubation at room temperature (25°C) for 7 d, 4°C for 28 d, and −20°C for 28 d; after freeze–thaw cycles (repeated 3 times) for QC solutions (LQC and HQC), each prepared and measured 3 times. Dilution tests were performed to evaluate accuracy at high concentrations (1 μg/mL RUX and 10 μg/mL PCZ), followed by a 1 : 9 (v/v) dilution in the blank plasma independently measured 3 times. The recovery and relative standard deviation percentages of each analyte under the above conditions were calculated based on the relative error (RE) and coefficient of variation (CV), respectively.

Clinical Application

The validated method was applied in a pharmacokinetic study approved by the Ethics Review Board of Tohoku University (Approval No. 2023-1-965). This study was performed in accordance with the ethical principles for medical research outlined in the Declaration of Helsinki 1964 and its subsequent revisions (https://www.wma.net/; accessed January 19, 2024). Patient consent was obtained via an opt-out procedure for samples measured during routine clinical laboratory testing. Blood samples from patients administered RUX were collected between March 2024 and August 2024 under real-world clinical conditions, without standardization of dietary intake or daily activities. Briefly, samples were collected at trough levels for inpatients and approximately 12–15 h post-dose for outpatients. The heparinized blood was centrifuged at 1580 × g for 5 min at 4°C and then used for quantitative analysis.

An automated sample preparation method using a CLAM-2030 (Shimadzu; auto-LC–MS/MS method) was tested for the quantification of plasma RUX and PCZ concentrations. First, the IS solution and each of the CS and plasma samples (>200 μL) were set in the CLAM-2030. A polytetrafluoroethylene membrane filter activated by 20 μL of methanol was used for pretreatment. The IS solution (100 μL) and samples (30 μL) were dispensed into a dedicated vial and stirred. After filtration, 10 μL of the sample was injected into the LC–MS/MS system described above. The concordance between the validated method and auto-LC–MS/MS method was evaluated using Passing–Bablok and Bland–Altman analyses,^33,34) which were conducted using MedCalc Statistical Software version 22 (MedCalc, Ostend, Belgium).

RESULTS

RUX, PCZ, and the ISs were successfully separated and exhibited elution profiles with sharp peaks (Fig. 1). Signals indicating RUX and PCZ were not detected in the control plasma samples, and among clinically used drugs, no compounds are known to have the exact same precursor and fragment ions as RUX or PCZ. The calibration curves for each analyte exhibited a linear concentration range between CS1 and CS8 with a coefficient of determination greater than 0.999. Carryover was assessed by injecting a blank after CS8, and no RUX or PCZ peaks were detected, thereby indicating negligible carryover. Moreover, the accuracy at each CS level was within 15% of the nominal values. Thus, these conditions were then applied for subsequent validation experiments. For the matrix effects, the RE and CV values indicated −1.13 and 7.51% for RUX, and 9.36 and 8.87% for PCZ, respectively. The accuracy and precision of the intra- and inter-day LLOQ, LQC, MQC, and HQC are summarized in Table 2. All the values of average RE and CV were within ±15%. Stability tests showed that the concentrations of RUX and PCZ remained unaltered under all the tested conditions (Table 2). For the dilution test, the results satisfied the adequacy criteria for dilution. The RE and CV values were −0.56 and 1.14% for RUX, and −14.36 and 9.54% for PCZ, respectively. Taken together, all the values satisfied the standards proposed by the FDA, although some individual measurements showed larger variability.

Fig. 1. Analytes and Representative Chromatogram

(A) Chemical structures of analytes. (B) Multiple reaction monitoring chromatograms were injected with a control plasma sample containing ruxolitinib (30 ng/mL), posaconazole (300 ng/mL), and internal standard solution.

Table 2. Assay Performance and Stabilities for Each Analyte

	Intra-day (N = 6)		Inter-day (N = 18)		Room temperature (25°C) for 7 d (N = 3)		4°C for 28 d (N = 3)		−20°C for 28 d (N = 3)		Three times freeze and thaw (N = 3)
	RE (%)	CV (%)	RE (%)	CV (%)	RE (%)	CV (%)	RE (%)	CV (%)	RE (%)	CV (%)	RE (%)	CV (%)
Ruxolitinib
LLOQ	2.89	5.92	5.59	6.63	—	—	—	—	—	—	—	—
LQC	−0.48	3.91	3.67	4.52	9.53	1.84	2.43	2.24	−1.63	2.89	5.63	4.40
MQC	−0.95	5.31	2.42	5.76	—	—	—	—	—	—	—	—
HQC	2.44	3.81	5.80	4.90	9.87	1.15	−9.20	0.58	−9.47	9.60	−6.79	8.81
Posaconazole
LLOQ	9.03	3.74	3.81	13.85	—	—	—	—	—	—	—	—
LQC	10.89	5.48	7.07	6.61	−7.75	6.80	5.00	9.55	12.32	4.27	12.22	4.06
MQC	8.76	4.57	9.37	4.18	—	—	—	—	—	—	—	—
HQC	10.52	4.32	11.67	4.04	−14.32	4.10	1.71	2.82	−0.28	8.38	9.04	5.68

Data represent the RE and CV values of samples prepared independently. RUX: ruxolitinib; PCZ: posaconazole; LLOQ: lower limit of quantification; LQC: low-quality control; MQC: medium-quality control; HQC: high-quality control; RT: room temperature; RE: relative error; CV: coefficient of variation.

A total of 45 plasma samples were collected from 6 patients who were administered RUX for GVHD treatment. Basic patient characteristics are provided in Table 3. Plasma RUX concentrations ranged between 5.07 and 246.51 ng/mL (interquartile range [IQR], 118.98), and no samples were below or beyond the calibration curve concentration range. Moreover, plasma PCZ concentrations in the patients who were administered PCZ with RUX (N = 17) were within the calibration curve concentration range (141.81–1577.55 ng/mL, IQR = 279.79). Individual quantified concentrations were provided in the Supplementary Tables 1 and 2. RUX and PCZ levels were simultaneously quantified in the patients who were administered both drugs. The plasma RUX and PCZ concentrations for 1 inpatient (Pt4) are chronologically shown in Fig. 2, along with clinical laboratory values. Relatively higher concentrations of RUX were observed on Day 21. The serum creatinine (SCr) concentration gradually increased, whereas the estimated glomerular filtration rate (eGFR) gradually decreased, although remarkable blood toxicity or liver injury was not observed. Thus, plasma RUX concentration may be correlated with kidney injury; however, this observation was limited to a single patient and should be considered exploratory.

Table 3. Characterization of Each Patient Administered with Ruxolitinib

Number	Sex	Age (years)	Ruxolitinib dosage (mg/d)	Posaconazole dosage (mg/d)	Care classification
Pt1	Female	68	20	0	Inpatient
Pt2	Male	47	20	300 (–Day –165)	Outpatient
				0 (Day –164–)
Pt3	Male	27	20	0	Inpatient
			15 (Day 41–)		Outpatient (Day 47–)
Pt4	Male	55	20	300	Inpatient
Pt5	Female	54	10	0	Outpatient
Pt6	Male	40	20	300	Inpatient

Fig. 2. Clinical Laboratory Values of Patient 4

SCr: serum creatinine; eGFR: estimated glomerular filtration rate; AST: aspartate aminotransferase; ALT: alanine aminotransferase; T-Bil: total bilirubin; WBC: white blood cell; RBC: red blood cell; Hb: hemoglobin; PLT: platelet.

Finally, the utility of the auto-LC–MS/MS method for TDM was evaluated by comparing the plasma RUX and PCZ concentrations obtained using the validated method and the auto-LC–MS/MS method, which were measured on the same day to minimize variability due to storage or handling. The calibration curves for each analyte exhibited a linear concentration range between CS1 and CS8 with a coefficient of determination greater than 0.999 using the auto-LC-MS/MS method. Passing–Bablok regression and Bland–Altman analyses were conducted to compare the results obtained by the validated method and auto-LC–MS/MS method using all the plasma samples; the correlations between the 2 methods were highly concordant (Fig. 3). The 95% confidence interval (CI) values for the slope obtained by Passing–Bablok regression analysis ranged from 0.97 to 1.05 in RUX (mean bias, 1.02) and 0.83 to 1.18 in PCZ (mean bias, 0.99), whereas the 95% CI values for the intercepts ranged from −0.79 to 0.79 in RUX (mean bias, 0.18) and −150.35 to 167.68 in PCZ (mean bias, −16.32). The values calculated by the Bland–Altman analysis (upper and lower limits of agreement) were not significantly different between the methods. These results indicate that the quantified concentrations using the auto-LC-MS/MS did not differ from those obtained with the validated method.

Fig. 3. Comparison between Validated and Automated Sample Preparation Methods (Auto-LC–MS/MS)

(A) Passing–Bablok regression analysis for ruxolitinib and posaconazole. Thick solid, dashed, and dotted lines indicate the estimated regression equation, upper and lower limits of the 95% confidence interval, and identity line, respectively. (B) Bland–Altman plots for ruxolitinib and posaconazole. Solid and dashed lines indicate the mean relative difference between the 2 methods and the upper and lower limits of agreement calculated as the mean relative difference ±1.96 standard deviations, respectively.

DISCUSSION

RUX, a selective oral JAK1/2 inhibitor, is the only drug approved by the FDA for the treatment of steroid-refractory aGVHD and cGVHD.^7,9) Although the therapeutic drug window of RUX is still being explored, relationships between exposure–response and toxicity have been suggested.²⁵⁾ In patients with GVHD, antifungals, such as PCZ and fluconazole, are used in combination with RUX.^17,21) Patients administered PCZ also require TDM to adhere to the target trough PCZ levels that have been previously proposed.²¹⁾ Here, a simultaneous LC–MS/MS quantification method for RUX and PCZ was developed and validated. Subsequently, their plasma concentrations in patients with GVHD were quantified, and their correlation with clinical laboratory values was evaluated in some patients. The utility of the auto-LC–MS/MS method was explored by comparing it with the LC–MS/MS method validated in this study.

In clinical practice, simultaneous quantification methods using the same pretreatment process and LC–MS/MS conditions for numerous analytes are desirable to achieve operational standardization.^27,28) The selection of target analytes during method development generally considers pharmacological mechanisms, clinical applications, or structural features.^27,28,35,36) For instance, simultaneous quantification methods for 16 antipsychotics, 7 monoclonal antibodies, and 20 oral molecular-targeted anticancer drugs have been reported.^27,35,36) Although challenges in sensitivity and separation differences can arise when targeting analytes with different structures, a simultaneous method for quantifying RUX and PCZ concentrations in patients with GVHD was successfully developed and validated (Fig. 1B and Table 2), and thus, the assay performance and stability of the analytes were good under clinically assumed conditions. In this study, we selected a metal-free column used in a previous study to minimize undesired interactions with the analytes.^{16,27,29,36,37)} The pK_a values of RUX and PCZ are 5.9 and 4.6, respectively, and both analytes exhibited efficient ionization in the positive ion mode.^38,39) Additionally, their solubility within the clinically relevant concentration ranges was sufficient to allow reliable quantification. Although some drugs can approach solubility limits within the target concentration range—making quantification challenging—optimization of ionization parameters has been reported to overcome such issues.²⁷⁾ These considerations support the rationale for why this combination of analytes could be successfully quantified under unified pretreatment and LC–MS/MS conditions.

Next, plasma concentrations of RUX and PCZ in patients who were administered these medications were measured. Both concentrations for all samples were within the ranges of the calibration curves. Thus, this validated method is useful in clinical applications. The serum trough RUX concentrations in 29 patients with aGVHD or cGVHD ranged from 5.6 to 99.8 ng/mL.²⁶⁾ Importantly, RUX clearance in patients who were co-administered strong CYP3A4 or CYP2C9 inhibitors decreased by 50% compared with that in patients without any CYP3A4 or CYP2C9 inhibitors.²⁶⁾ In another clinical trial, the maximum plasma RUX concentrations for patients with or without voriconazole (a CYP3A4 inhibitor) co-administration were 72.5 ± 14.7 and 48.2 ± 14.5 ng/mL, respectively.³⁰⁾ A systematic review of RUX pharmacokinetics revealed various values, including maximum RUX concentrations and areas under the curves.³¹⁾ In Pt4, the plasma trough RUX concentration on Day 21 was 127.0 ng/mL, which was approximately 4.8-fold higher than that measured 2 d before (Fig. 2). Simultaneously, deterioration of renal function was observed until day 30, which resulted in death. Although some clinical laboratory values related to liver function and hematology also indicated values outside the reference range, renal injury that may have been associated with RUX could have occurred in Pt4 because both the SCr and eGFR values suddenly and gradually changed; however, causality cannot be formally established based on a single case in this study. Cases of acute kidney injury after RUX administration have been reported.^40,41) Notably, the time-series data (Fig. 2) indicate that the increase in RUX concentration was preceded by the rise in concentration of PCZ (a strong CYP3A4 inhibitor) until around Day 20. This temporal pattern suggests a potential DDI in which elevated PCZ exposure may have reduced RUX clearance, thereby contributing to the subsequent sharp increase in RUX levels. This alternative hypothesis could also be considered when interpreting the cause of renal deterioration in Pt4. Therefore, RUX TDM may be effective for more appropriate therapeutic management.

Antifungals, including PCZ, are used for therapeutic and prophylactic purposes in hematology and oncology treatments.^22,42) For patients administered PCZ, bioavailability reduction is a crucial problem because of the adverse effects of PCZ, such as mucositis and diarrhea.³²⁾ DDIs with PCZ are also important because PCZ is known to be a strong CYP3A inhibitor,¹⁹⁾ and a case affecting the dosage adjustment of venetoclax, which is a CYP3A substrate, by PCZ suspension has been reported.¹⁶⁾ Among antifungals, PCZ is preferred in terms of efficacy and effective fungal spectrum.^43,44) In this study, plasma PCZ concentrations in Pt4 were measured, and the values gradually decreased below the recommended concentrations. Since decreased absorption ability may reflect the concentration of PCZ, TDM has been suggested to be important. Particularly, Pt4 had exhibited diarrhea even before the initiation of PCZ, which may have contributed to impaired gastrointestinal absorption. Therefore, the marked decline in PCZ concentration after the switch from intravenous injection to oral administration may have been at least partly attributable to impaired oral absorption caused by diarrhea. In short, both RUX and PCZ should be administered with TDM to improve pharmacotherapy, even in situations where only PCZ, but not RUX, is administered to patients. Notably, plasma RUX concentrations varied widely among patients, which is likely due to multiple factors, including co-administration of PCZ, renal function, and dosing timing. Although formal evaluation of DDI impact was not feasible in this cohort, these findings underscore the necessity of TDM (Supplementary Tables 1 and 2).

In addition, the utility of the auto-LC–MS/MS method in clinical practice was evaluated. Auto-LC–MS/MS methods have been used for endogenous steroids and various drugs, including beta-lactam antibiotics (ceftazidime, cefotaxime, flucloxacillin, meropenem, and piperacillin).^45,46) Both robustness and routine capability were indicated; thus, the application of routine TDM is expected. Highly concordant results were obtained between the validated and auto-LC–MS/MS methods. However, some issues, including the sample volume used for auto-LC–MS/MS, remain; thus, further improvement would be required to reduce the burden on patients and for application in clinical practice. Meanwhile, the auto-LC–MS/MS method reduces the burden on the measurer and decreases the variation in measurement accuracy caused by inter-individual differences in the measurer’s experience and skill. Additionally, the reduction in manual handling and the standardization of measurements are expected to improve workflow efficiency in clinical practice, although the initial implementation cost remains a concern.

In this study, we developed an analytical method for the simultaneous quantification of RUX and PCZ in patients with GVHD. This method was applied to human plasma samples (N = 45) obtained from 6 patients. Importantly, the possibility of a clinically relevant association between plasma RUX concentrations and kidney injury was recognized, thereby suggesting that TDM of RUX can be effective for therapeutic management. However, this observation was based on only a single patient among the 6 studied patients, and therefore, should be interpreted with caution. Furthermore, the utility of the auto-LC–MS/MS method was demonstrated, and our results, taken together with those of other investigations, could lead to clinical applications. Nevertheless, complete validation of the automated method, including assessment of accuracy and precision, will be necessary for clinical implementation. Outpatients with GVHD can receive treatment because both RUX and PCZ are orally available drugs; therefore, it is important to perform TDM with a minimum burden for patients. Recently, the utility of a plasma trough concentration prediction program using a nonlinear mixed-effects model was also reported.³⁷⁾ Although further studies such as multicenter validation with a large sample size, population-based TDM, and implications for dose adjustment based on PK/PD are required, the most effective TDM strategy will be established by combining these technologies and methods, and these aspects should be considered in future clinical implementation.

Acknowledgments

We are grateful to all the patients who provided plasma samples. Part of this study was supported by grants from the Japan Society for the Promotion of Science (Grant Numbers: 22K15333 and 24K18325 to M. Kumondai).

DECLARATIONS

Author Contributions

Conceptualization: Masaki Kumondai and Yasushi Onishi. Methodology: Masaki Kumondai, Yoshihiro Hayakawa, Yu Sato, Yuji Sato, Toshihiro Sato, Mayumi Sato, Masafumi Kikuchi, Masamitsu Maekawa, and Nariyasu Mano. Validation and analysis: Masaki Kumondai, Nagomi Hayashi, and Yu Sato. Investigation: Masaki Kumondai, Daisuke Kobayashi, Ayaka Otsuki, Yugo Ueki, and Yasushi Onishi. Data curation: Masaki Kumondai, Nagomi Hayashi, Yu Sato, Daisuke Kobayashi, Ayaka Otsuki, and Yugo Ueki. Writing—original draft: Masaki Kumondai. Writing—review and editing: Nagomi Hayashi, Yu Sato, Daisuke Kobayashi, Ayaka Otsuki, Yugo Ueki, Yasushi Onishi, Taku Tsukamoto, Kohei Yoshikawa, Yoshihiro Hayakawa, Yuji Sato, Toshihiro Sato, Mayumi Sato, Masafumi Kikuchi, Masamitsu Maekawa, and Nariyasu Mano. Visualization: Masaki Kumondai, Nagomi Hayashi, and Yu Sato. Resources: Taku Tsukamoto, Kohei Yoshikawa, Yoshihiro Hayakawa, Masafumi Kikuchi, Masamitsu Maekawa, and Nariyasu Mano. Supervision: Masamitsu Maekawa and Nariyasu Mano. Project administration: Masaki Kumondai. All authors have read and agreed to the published version of the manuscript.

Conflict of Interest

Masaki Kumondai, Yu Sato, Yuji Sato, Toshihiro Sato, Mayumi Sato, Masamitsu Maekawa, and Nariyasu Mano have a joint research agreement with the Shimadzu Corporation. Taku Tsukamoto, Kohei Yoshikawa, and Yoshihiro Hayakawa are employed by the Shimadzu Corporation (Kyoto, Japan). The other authors declare no conflict of interest.

Data Availability

The dataset is available on request from the authors.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1)  Malard F, Holler E, Sandmaier BM, Huang H, Mohty M. Acute graft-versus-host disease. Nat. Rev. Dis. Primers, 9, 27 (2023).
• 2)  Amanam I, Otoukesh S, Al Malki MM, Salhotra A. Chronic GVHD: review advances in prevention, novel endpoints, and targeted strategies. Hematology (Am. Soc. Hematol. Educ. Program), 2023, 164–170 (2023).
• 3)  Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol. Blood Marrow Transplant., 11, 945–956 (2005).
• 4)  Auletta JJ, Kou J, Chen M, et al. Real-world data showing trends and outcomes by race and ethnicity in allogeneic hematopoietic cell transplantation: a report from the Center for International Blood and Marrow Transplant Research. Transplant. Cell. Ther., 29, 346.e1–346.e10 (2023).
• 5)  Ramdial JL, Mehta RS, Saliba RM, Alousi AM, Bashir Q, Hosing C, Kebriaei P, Olson AL, Oran B, Qazilbash MH, Srour SA, Andersson BS, Champlin RE, Popat U. Acute graft-versus-host disease is the foremost cause of late nonrelapse mortality. Bone Marrow Transplant., 56, 2005–2012 (2021).
• 6)  Axt L, Naumann A, Toennies J, Haen SP, Vogel W, Schneidawind D, Wirths S, Moehle R, Faul C, Kanz L, Axt S, Bethge WA. Retrospective single center analysis of outcome, risk factors and therapy in steroid refractory graft-versus-host disease after allogeneic hematopoietic cell transplantation. Bone Marrow Transplant., 54, 1805–1814 (2019).
• 7)  Martini DJ, Chen YB, DeFilipp Z. Recent FDA approvals in the treatment of graft-versus-host disease. Oncologist, 27, 685–693 (2022).
• 8)  De Togni E, Cole O, Abboud R. Janus kinase inhibition in the treatment and prevention of graft-versus-host disease. Front. Immunol., 15, 1304065 (2024).
• 9)  Becker H, Engelhardt M, von Bubnoff N, Wäsch R. Ruxolitinib. Recent Results Cancer Res., 201, 249–257 (2014).
• 10)  Yang Y, Huo Y, Zhou D, Kang Z, Huang Y, Wang Y, Fan G, Ma L, Yan J. Efficacy and safety of Ruxolitinib combined with steroids for first-line treatment of acute graft-versus-host disease after hematopoietic stem cell transplantation: a single-center, real-world experience. Front. Immunol., 16, 1621708 (2025).
• 11)  Baccelli F, Gottardi F, Muratore E, Leardini D, Grasso AG, Gori D, Belotti T, Prete A, Masetti R. Ruxolitinib for the treatment of acute and chronic graft-versus-host disease in children: a systematic review and individual patient data meta-analysis. Bone Marrow Transplant., 59, 765–776 (2024).
• 12)  DeFilipp Z, Kim HT, Knight LW, O’Connor SM, Dhaver SE, White M, Dholaria B, Schroeder MA, Vasu S, Abedin S, Chung J, El-Jawahri A, Frigault MJ, McAfee S, Newcomb RA, O’Donnell PV, Spitzer TR, Chen YB, Hobbs GS. Low rates of chronic graft-versus-host disease with Ruxolitinib maintenance following allogeneic HCT. Blood, 145, 2312–2316 (2025).
• 13)  Zanger UM, Schwab M. Cytochrome P450 enzymes in drug metabolism: regulation of gene expression, enzyme activities, and impact of genetic variation. Pharmacol. Ther., 138, 103–141 (2013).
• 14)  Hendriks DFG, Vorrink SU, Smutny T, Sim SC, Nordling Å., Ullah S, Kumondai M, Jones BC, Johansson I, Andersson TB, Lauschke VM, Ingelman-Sundberg M. Clinically relevant cytochrome P450 3A4 induction mechanisms and drug screening in three-dimensional spheroid cultures of primary human hepatocytes. Clin. Pharmacol. Ther., 108, 844–855 (2020).
• 15)  Kumondai M, Kikuchi M, Mizuguchi A, Hayashi N, Ui M, Hirama T, Okada Y, Sato Y, Sato T, Maekawa M, Mano N. Therapeutic drug monitoring of blood sirolimus and tacrolimus concentrations for polypharmacy management in a lymphangioleiomyomatosis patient taking two cytochrome P450 3A inhibitors. Tohoku J. Exp. Med., 260, 29–34 (2023).
• 16)  Otsuki A, Kumondai M, Kobayashi D, Kikuchi M, Ueki Y, Sato Y, Hayashi N, Yagi A, Onishi Y, Onodera K, Ichikawa S, Fukuhara N, Yokoyama H, Maekawa M, Mano N. Plasma venetoclax concentrations in patients with acute myeloid leukemia treated with CYP3A4 inhibitors. Yakugaku Zasshi, 144, 775–779 (2024).
• 17)  Ullmann AJ, Lipton JH, Vesole DH, Chandrasekar P, Langston A, Tarantolo SR, Greinix H, Morais de Azevedo W, Reddy V, Boparai N, Pedicone L, Patino H, Durrant S. Posaconazole or fluconazole for prophylaxis in severe graft-versus-host disease. N. Engl. J. Med., 356, 335–347 (2007).
• 18)  Hayashi Y, Kanda Y, Nakamae H, Kanamori H, Ohashi K, Hidaka M, Yano S, Hatanaka K, Kohno A, Moriuchi Y, Ago H, Yamashita T, Takata T, Yoshida M, Hino M, Yamaguchi T, Fukuda T. Voriconazole vs. itraconazole for antifungal prophylaxis in patients with GVHD: a randomized trial. Biol. Blood Marrow Transplant., 20 (supple.), S91 (2014).
• 19)  Krishna G, Ma L, Prasad P, Moton A, Martinho M, O’Mara E. Effect of posaconazole on the pharmacokinetics of simvastatin and midazolam in healthy volunteers. Expert Opin. Drug Metab. Toxicol., 8, 1–10 (2012).
• 20)  Godamudunage MP, Grech AM, Scott EE. Comparison of antifungal azole interactions with adult cytochrome P450 3A4 versus neonatal cytochrome P450 3A7. Drug Metab. Dispos., 46, 1329–1337 (2018).
• 21)  Maertens JA, Girmenia C, Brüggemann RJ, Duarte RF, Kibbler CC, Ljungman P, Racil Z, Ribaud P, Slavin MA, Cornely OA, Peter Donnelly JP, Cordonnier C. European guidelines for primary antifungal prophylaxis in adult haematology patients: summary of the updated recommendations from the European Conference on Infections in Leukaemia. J. Antimicrob. Chemother., 73, 3221–3230 (2018).
• 22)  Kably B, Launay M, Derobertmasure A, Lefeuvre S, Dannaoui E, Billaud EM. Antifungal drugs TDM: trends and update. Ther. Drug Monit., 44, 166–197 (2022).
• 23)  Mohammadpour N, Elyasi S, Vahdati N, Mohammadpour AH, Shamsara J. A review on therapeutic drug monitoring of immunosuppressant drugs. Iran. J. Basic Med. Sci., 14, 485–498 (2011).
• 24)  Ashbee HR, Barnes RA, Johnson EM, Richardson MD, Gorton R, Hope WW. Therapeutic drug monitoring (TDM) of antifungal agents: guidelines from the British Society for Medical Mycology. J. Antimicrob. Chemother., 69, 1162–1176 (2014).
• 25)  van der Kleij MBA, Guchelaar NAD, Mathijssen RHJ, Versluis J, Huitema ADR, Koolen SLW, Steeghs N. Therapeutic drug monitoring of kinase inhibitors in oncology. Clin. Pharmacokinet., 62, 1333–1364 (2023).
• 26)  Isberner N, Kraus S, Grigoleit GU, Aghai F, Kurlbaum M, Zimmermann S, Klinker H, Scherf-Clavel O. Ruxolitinib exposure in patients with acute and chronic graft versus host disease in routine clinical practice-a prospective single-center trial. Cancer Chemother. Pharmacol., 88, 973–983 (2021).
• 27)  Hirasawa T, Kikuchi M, Takasaki S, Kumondai M, Sato Y, Sato T, Imoto E, Hayakawa Y, Maekawa M, Mano N. High throughput LC/ESI-MS/MS method for simultaneous analysis of 20 oral molecular-targeted anticancer drugs and the active metabolite of sunitinib in human plasma. Heliyon, 9, e16926 (2023).
• 28)  Rappold BA. Review of the use of liquid chromatography-tandem mass spectrometry in clinical laboratories: Part II—Operations. Ann. Lab. Med., 42, 531–557 (2022).
• 29)  Sato Y, Kondo H, Sato Y, Abe A, Kikuchi M, Sato T, Kumondai M, Yoshikawa K, Hayakawa Y, Maekawa M, Mano N. Development of simultaneous drug concentration measurement method using an automated pretreatment liquid chromatography/tandem mass spectrometry system for therapeutic drug monitoring. Pharmaceutics, 16, 1138 (2024).
• 30)  Zhao Y, Chen P, Dou L, Li F, Li M, Xu L, Chen J, Jia M, Huang S, Wang N, Luan S, Yang J, Bai N, Liu D. Co-administration with voriconazole doubles the exposure of Ruxolitinib in patients with hematological malignancies. Drug Des. Devel. Ther., 16, 817–825 (2022).
• 31)  Appeldoorn TYJ, Munnink THO, Morsink LM, Hooge MNL, Touw DJ. Pharmacokinetics and pharmacodynamics of Ruxolitinib: a review. Clin. Pharmacokinet., 62, 559–571 (2023).
• 32)  Selby PR, Heffernan AJ, Yeung D, Warner MS, Peake SL, Hahn U, Westley I, Shakib S, Roberts JA. Population pharmacokinetics of posaconazole in allogeneic haematopoietic stem cell transplant patients. J. Antimicrob. Chemother., 79, 567–577 (2024).
• 33)  Passing H, Bablok W. A new biometrical procedure for testing the equality of measurements from two different analytical methods: application of linear regression procedures for method comparison studies in clinical chemistry, Part I. J. Clin. Chem. Clin. Biochem., 21, 709–720 (1983).
• 34)  Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 327, 307–310 (1986).
• 35)  Amrani ME, Gerencser L, Huitema ADR, Hack CE, van Luin M, van der Elst KCM. A generic sample preparation method for the multiplex analysis of seven therapeutic monoclonal antibodies in human plasma or serum with liquid chromatography-tandem mass spectrometry. J. Chromatogr. A, 1655, 462489 (2021).
• 36)  Maekawa M, Yokota M, Sato T, Sato Y, Kumondai M, Sato Y, Suzuka M, Kobayashi D, Sakamoto K, Matsuura M, Kikuchi M, Komatsu H, Fujii K, Ozeki Y, Tomita H, Mano N. Development of a simultaneous LC-MS/MS analytical method for plasma: 16 antipsychotics approved in Japan and 4 drug metabolites. Anal. Sci., 40, 1749–1763 (2024).
• 37)  Kumondai M, Ogawa R, Hayashi N, Ishida Y, Oshikiri H, Sato Y, Kikuchi M, Sato Y, Sato T, Maekawa M, Mano N. Relevance of plasma lenvatinib concentrations and endogenous urinary cytochrome P450 3A activity biomarkers in clinical practice. Pharmacol. Res. Perspect., 12, e1241 (2024).
• 38)  Shi JG, Fraczkiewicz G, Williams WV, Yeleswaram S. Predicting drug-drug interactions involving multiple mechanisms using physiologically based pharmacokinetic modeling: a case study with Ruxolitinib. Clin. Pharmacol. Ther., 97, 177–185 (2015).
• 39)  Nijhawan M, Godugu M, Saxena T, Farheen T, Dwivedi K. Pharmaceutical co-crystals of posaconazole for improvement of physicochemical properties. Braz. J. Pharm. Sci., 58, e191024 (2022).
• 40)  Jagasia M, Perales MA, Schroeder MA, Ali H, Shah NN, Chen YB, Fazal S, Dawkins FW, Arbushites MC, Tian C, Connelly-Smith L, Howell MD, Khoury HJ. Ruxolitinib for the treatment of steroid-refractory acute GVHD (REACH1): a multicenter, open-label phase 2 trial. Blood, 135, 1739–1749 (2020).
• 41)  Ali H, Tsai NC, Synold T, Mokhtari S, Tsia W, Palmer J, Stiller T, Al Malki M, Aldoss I, Salhotra A, Rahmanuddin S, Pullarkat V, Cai JL, Stein A, Forman SJ, Marcucci G, Mei M, Snyder DS, Nakamura R. Peritransplantation Ruxolitinib administration is safe and effective in patients with myelofibrosis: a pilot open-label study. Blood Adv., 6, 1444–1453 (2022).
• 42)  Greer ND. Posaconazole (Noxafil): a new triazole antifungal agent. Proc. Bayl. Univ. Med. Cent., 20, 188–196 (2007).
• 43)  Nagappan V, Deresinski S. Reviews of anti-infective agents: posaconazole: a broad-spectrum triazole antifungal agent. Clin. Infect. Dis., 45, 1610–1617 (2007).
• 44)  Kung HC, Johnson MD, Drew RH, Saha-Chaudhuri P, Perfect JR. Clinical effectiveness of posaconazole versus fluconazole as antifungal prophylaxis in hematology-oncology patients: a retrospective cohort study. Cancer Med., 3, 667–673 (2014).
• 45)  Fraissinet F, Pereira T, Violin A, Feugray G, Bach-Ngohou K, Brunel V. A novel fully-automated method to measure steroids in serum by liquid chromatography-tandem mass spectrometry. J. Mass Spectrom. Adv. Clin. Lab, 27, 24–32 (2023).
• 46)  Khromov T, Dihazi GH, Brockmeyer P, Fischer A, Streit F. 24/7 therapeutic drug monitoring of beta-lactam antibiotics with CLAM-2000. Antibiotics, 12, 1526 (2023).