This single-center, retrospective, observational study was conducted at The University of Tokyo Hospital (Tokyo, Japan). Patients who underwent allo-HSCT and received therapeutic IV-MTX administration during immunosuppressive therapy with TCR (administered as CIV) from April 2004 to December 2021 were assessed for eligibility. Each therapeutic IV-MTX administration was defined as a case, when a patient received therapeutic IV-MTX administration more than once during the study period, each case was independently assessed for eligibility. When conducting the PK analysis, the cases that met the following exclusion criteria were excluded: 1) oral TCR was administered between 2 days before and 4 days after the therapeutic IV-MTX administration, 2) TCR CIV administration was suspended between 2 days before and 4 days after the therapeutic IV-MTX administration due to steroid pulse therapy for acute GVHD, 3) MTX was re-administered within 4 days after the previous therapeutic IV-MTX administration because it was difficult to judge which therapeutic IV-MTX administration affected C/D, 4) Css was not measured before (on the day or one day before the day) and/or after (from 2 to 4 days after) the therapeutic IV-MTX administration, 5) azole antifungals were newly started or stopped between 2 days before and 4 days after the therapeutic IV-MTX administration, 6) the kind of azole antifungals was switched to another azole antifungal between 2 days before and 4 days after the therapeutic IV-MTX administration, and 7) aprepitant, well-known CYP3A inhibitor and inducer [10, 11] was administered within 3 weeks before therapeutic IV-MTX administration.

Allo-HSCT was performed after conditioning pretreatment with chemo(radio)therapy in The University of Tokyo Hospital (Tokyo, Japan), following which, immunosuppressive treatment with TCR and MTX was started to prevent GVHD [5, 6]. TCR was generally started from the day before allo-HSCT (day -1) at a dose of 0.02 mg/kg/day CIV and the dose was adjusted at the discretion of the attending physician based on TDM and clinical condition. MTX was administered intravenously on days 1, 3, 6, and 11 at doses of 15, 10, 10, and 10 mg/m2, respectively, or on days 1, 3, and 6 at doses of 10, 7, and 7 mg/m2, respectively.

The therapeutic IV-MTX was administered to the patients who developed GVHD at a dose of 3–10 mg/m2 weekly until the GVHD resolved [7,8,9].

The following demographic and clinical data were collected using the electronic medical record system: age at the time of allo-HSCT, sex, body weight, primary disease, conditioning pretreatment, classification of allo-HSCT, the doses of TCR and MTX, Css, concomitant use of azole antifungals and/or aprepitant, red blood cell (RBC) count, hematocrit (Hct), hemoglobin (Hb), serum albumin (Alb), aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin (T-bil), and serum creatinine (Cre) levels. In the hospital, blood concentrations of TCR were measured using chemiluminescent immunoassay (CLIA, ARCHITECT Tacrolimus assay, Abbott) (~ June 13th, 2021) and affinity column-mediated immune assay (ACMIA, Dimension Tacrolimus Flex reagent cartridge, Siemens Healthineers) (June 14th, 2021 ~).

Clinical laboratory data were obtained on the same day as Css measurements. For clinical laboratory data after MTX administration, average values from 2 to 4 days after therapeutic IV-MTX administration were used. When more than one case of therapeutic IV-MTX administration was eligible in a single patient, the mean value of clinical data of all cases in the patient was used in the analysis of changes in clinical laboratory data after therapeutic IV-MTX administration.

The C/D was calculated by the following equation [12]:

$$C/D [(ng/mL)/(mg/kg)]=\frac_\left[ng/mL\right]}$$

In this equation, the daily dose of TCR on the day before the Css measurement was used. The C/D values were calculated before and after therapeutic IV-MTX administration to evaluate its effects on the clearance of TCR. The C/D before therapeutic IV-MTX administration was calculated using the Css measured immediately before (or measured on the previous day when Css on the day of MTX administration was not measured). In contrast, the mean C/D value from days 2 to 4 after therapeutic IV-MTX administration was used as the C/D value after therapeutic IV-MTX administration. When more than one case of therapeutic IV-MTX administration was eligible in a single patient, the mean value of C/D and fold changes in C/D of all cases in the patient were used in the analysis of changes in C/D after therapeutic IV-MTX administration.

The daily doses of TCR from the next day to 7 days after therapeutic IV-MTX administration were compared with those on the day of MTX administration to investigate changes in TCR doses. Dose changes were classified into three types: dose reduction, dose increase, and no change compared with the previous day. When both dose reduction and increase occurred from the next day to 7 days after therapeutic IV-MTX administration, the first change in the TCR dose was analyzed. When the TCR doses did not change from the next day to 7 days after therapeutic IV-MTX administration, these cases were classified as “no change”.

Continuous variables are shown as median (range) unless otherwise stated. Because the number of cases of therapeutic IV-MTX administration differed among patients, the geometric mean values of C/D before and after therapeutic IV-MTX administration in each case within the same patient were subjected to the following statistical analysis. Similarly, the relative frequency of each type of dose change (dose reduction, dose increase, or no change) for each patient was calculated. Wilcoxon signed-rank tests were used to compare the C/D and clinical laboratory data before and after therapeutic IV-MTX administration. A permuted Brunner-Munzel test was performed to compare the relative frequencies of each type of dose change [13].