Mini-dose methotrexate combined with methylprednisolone as a first-line treatment for acute graft-versus-host disease: A phase 2 trial

Study design and subjects

The prospective, nonrandomized, open-label study included patients recruited at the Peking University People’s Hospital between December 2020 and May 2021. The study was approved by the ethical committee of Peking University People’s Hospital (2020PHB067-01). All included subjects signed an informed consent. The study protocol was in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University. This study was registered at http://clinicaltrials.gov/NCT 04677868.

The study included patients, aged 16–65 years, with aGvHD after hematopoietic stem cell transplantation (HSCT), who had not received drug treatments and whose peripheral blood absolute neutrophil count was higher than 0.5 × 109/L. Exclusion criteria were patients with severe heart, kidney, or liver disease or a life-threatening infection.

A total of 31 patients developing aGvHD following HSCT were screened for inclusion in the trial to receive mini-dose MTX combined with 1 mg/kg MP as an initial therapy for aGvHD. During the study period, 28 patients solely receiving 1 mg/kg MP as a first-line treatment of aGvHD at the physicians’ discretion (owing to grade I acute GvHD, early GvHD onset time, or concerns about risks) were selected as the control cohort. Subject characteristics are shown in Table 1. Human leukocyte antigen (HLA)-A, HLA-B, and HLA-DRB1 typing was performed at the allele level using high-resolution techniques. Two siblings and two unrelated donor– recipient pairs were fully HLA matched.

Transplant procedures

Conditioning regimen and GvHD prophylaxis have been previously reported in detail.^[17,18] All transplant recipients received Busulfan+Cyclophosphamide (BuCy) -b a s ed my e l o a b l a t ive co n d it ion in g regimens.^{[19, 20, 21, 22]} The conditioning regimen included cytarabine (4 g/m²/day, day -9), busulfan (3.2 mg/kg/day, intravenously, days -8 to -6), cyclophosphamide (1.8 g/m²/day, days –5 to –4), semustine (250 mg/m², day –3), and rabbit ATG (thymoglobulin; Imtix Sangstat, Lyon, France; 2.5 mg/kg/day, days –5 to –2). Cyclosporine A (CsA), MMF, and short-term MTX were administered as GvHD prophylaxis. Haploidentical and unrelated patients received ATG.^[23] The dosage of MTX was 15 mg/m², intravenously, on day 1, followed by 10 mg/m² on days 3 and 6 after matched sibling transplantation or on days 3, 6, and 11 after mismatched/haploidentical or unrelated transplantation.

Study drug administration regimens

Patients received intravenous MTX at a dose of 5 mg/m² and MP at a dose of 1 mg/kg/day. MTX was administered on days 1, 3, 8, and 15 and once in every 7 days afterward. The programmed dose of MP was as follows: days 1–7, 1 mg/kg/day; days 8–14, 0.5 mg/kg/day; days 15–21, 0.25 mg/kg/day, and the dose was reduced by half after 5–7 days until it was stopped. Patients were scheduled to receive at least two doses (number of MTX administrations) for evaluation of the efficacy of the drug. If patients responded and tolerated the toxicity, additional doses were used for consolidation, and the scheduled maximal doses were capped at six.

Patients were observed for 5–7 days and switched to the second-line treatment if there was NR to the initial therapy. The second-line treatment was basiliximab (Novartis Pharma AG, Basel, Switzerland) at 20 mg/day on days 1, 3, 8, and weekly afterward for as long as it was clinically indicated. This second-line treatment was administered to patients for 5 days with progression of aGvHD or for 7 days with no improvement after initial therapy or for 14 days with partial response (PR) after the initial therapy, not including the patients who initially responded to MTX/MP and then flared.^[14]

Definition and evaluation

Before treatment initiation, patients underwent a thorough evaluation to ascertain the severity and extent of GvHD, including a physical examination, laboratory evaluations, and a consultation without tissue biopsy results. Each organ (skin, liver, and gut) was staged 1 through 4 for acute GvHD according to the modified criteria based on the schema of the Mount Sinai Acute GvHD International Consortium (MAGIC). Moreover, patients were assigned a grade of acute GvHD (I–IV) based on overall severity. Minnesota GvHD risk status was also evaluated.^[11] The time that elapsed between the onset of aGvHD and HSCT was defined as the time from HSCT to the onset of any grade of aGvHD.

A complete response (CR) was classified as complete disappearance of all clinical signs of skin, liver, and/or gut GvHD. It was assumed that a PR occurred if GvHD symptoms in the patient had not completely disappeared, but at least one target organ decreased in grading by at least one stage without deterioration or emergence of GvHD in other organs. Overall responses (ORs) included CR and PR. NR or treatment failure was defined as absence of improvement in any organ involved by aGvHD or worsening in one or more organs by one or more stages, requiring additional systemic GvHD therapy. Meanwhile, changes in white blood cell count and drug side effects were evaluated to assess drug safety. The Common Terminology Criteria (CTC) for adverse events version 3.0 were used to grade the severity of side effects.

Intracellular flow cytometry

Peripheral blood was harvested from patients before therapy and at 7 and 28 days after therapy. Anti-CD3-APC/H7, anti-CD8- eFluor450, anti-IFN-γ-Percp/Cy5.5, and anti-Ki67 monoclonal antibodies (BD Biosciences, San Jose, CA, USA) were used to stain cell surface markers and intracellular cytokines. CD3+ T cells were cultured with phorbol myristate acetate (100 ng/mL) and ionomycin (2 μg/mL) (both from Sigma-Aldrich, St. Louis, MO, USA) for 4 hours. During this incubation period, GolgiStop (0.7 μL/mL) was added to the samples to sequester cytoplasmic proteins. As previously described, Th1 and Tc1 cells were identified as CD3+CD8−IFN-γ+ and CD3+CD8+IFN-γ+ cells, respectively. Owing to difficulties in obtaining samples, we reserved the peripheral blood of 12 patients who were administered mini-dose MTX combined with 1 mg/kg MP as the initial treatment for aGvHD and six patients who were administered 1 mg/kg MP as the first-line treatment of aGvHD. The samples from MP + MTX group and MP group were matched for age, sex, and time after transplantation.

The proliferation of T cells was examined using a Ki67 antibody (BD Biosciences). First, the cells were harvested and incubated with anti-CD3-APC/H7 antibodies (BD Biosciences) at room temperature (24℃) for 15 minutes. Then, they were fixed and permeabilized with FOXP3 Fix/Perm buffer set (BD Biosciences). Subsequently, the cells were resuspended and incubated with prediluted Ki-67 antibody for 30 minutes on ice in the dark. The proliferation was monitored by the percentage of Ki67-positive cells by flow cytometric analysis.

Xenogeneic model of aGvHD

G-CSF–mobilized peripheral blood was obtained from healthy donors after they signed a written informed consent approved by our ethical committee. G-CSF–mobilized peripheral blood mononuclear cells (G-PBMC) were harvested from peripheral blood by Ficoll-Paque gradient centrifugation. The current study was approved by the ethical committee of Peking University People’s Hospital (2020PHB067-01).

A humanized xenogeneic aGvHD model was established to further evaluate the effect of MP and MTX on aGvHD, according to a previous report.^[24] Six-to eight-week-old NOD.Cg-Prkdc^scid Il2rg^tm1Vst/Vst (NPG; Beijing Vitalstar Biotechnology Co., Ltd., Beijing, China) mice were sublethally irradiated with 1.5 Gy total body irradiation via X-ray on day –1, followed by intravenous infusion of 5 × 10⁶ G-PBSCs in the caudal vein, and then were intraperitoneally injected with PBS (CTL), MP (2 mg/kg/day), and MTX (1 mg/kg/day) daily beginning on day 1 after allo-HSCT for 4 weeks. Disease severity was monitored using a scoring system that features six parameters: weight loss, posture, activity, fur ruffling, skin integrity, and diarrhea.^[25] Each parameter was assigned a score from 0 (no symptoms) to 2 (severe symptoms), which resulted in a total score from 0 to 12. Tissue samples were prepared, stained with hematoxylin, eosin, and safran, and imaged with microscope (Olympus, Tokyo, Japan) to evaluate GvHD target organ pathology scores.

Endpoints and statistical methods

The primary endpoint was the ORR to therapy at 7 days after treatment. The study would provide an 80% power with a sample size of 30 patients to detect a 25% benefit in ORR with the mini-dose MTX plus MP, compared to a reference rate of 50% with MP alone that was derived from previous reports. Type I and II error rates were controlled at 5% and 20%, respectively. The secondary endpoints were ORR at 28 days after treatment, incidence of bacterial infections, fungal infections, cytomegalovirus (CMV) and Epstein–Barr virus (EBV) infections, transplant-related mortality (TRM), relapse of original disease, failure-free survival (FFS; defined as alive without relapse, requirement for additional therapy for acute GvHD, or signs or symptoms of moderate-to-severe chronic GvHD), overall survival (OS, time from HSCT to death from any cause), and chronic GvHD incidence.

Data were collected on case report forms by medical record reviews. The time course for aGvHD response and FFS was estimated using the method of Kaplan– Meier. The surviving patients were followed up, and the results of the follow-up examinations were analyzed on December 10, 2021. Unless otherwise specified, all the reported P values were based on two-sided hypothesis tests. Statistical Package for the Social Sciences (SPSS) 19.0 (Mathsoft, Seattle, WA, USA) and R version 3.4.4 (The R Foundation for Statistical Computing) were used for data analyses.

Patients

Thirty-one patients were enrolled in the study. Most patients were female (68%) and underwent haploidentical HSCT (87%). Median age was 34 (16–65) years. Most of the patients had grade I (19%) or II (71%) aGvHD as per the MAGIC criteria and standard risk (90%) as per the Minnesota risk stratification. Organ involvement was primarily skin (80%), followed by lower gastrointestinal tract (GI) (32%) and liver (6%) (Table 1).

Responses

aGvHD occurred at a median of 22 (11–85) days after HSCT among the 31 patients in the study group. The drugs were immediately administered after aGvHD was diagnosed. The ORR and CR rate at 7 days after MTX + MP treatment were 100% (31/31 patients) and 83% (26/31), respectively. Among the five patients achieving PR at 7 days after treatment, three achieved CR at 9, 14, and 24 days after MTX + MP treatment without additional second-line treatment and the other two received second-line treatment at 14 and 21 days after MTX + MP therapy. By day 28 after drug administration, among the 29 patients achieving CR after MTX + MP therapy without additional second-line treatment, two flared and received second-line treatment. Therefore, the ORR and CR rate on day 28 were 87% (27/31). Responses in different grades of GvHD (the grades at onset), individual organs involved, and the number of involved organs are shown in Table 2.

MTX was administered at a median of 3 (2–4) times. The median time to show improvement (at least reaching PR) for the 31 patients was 3 (2–7) days after application. The median time needed to achieve a maximal response (CR or PR) was 3 (2–24) days. The time course for maximal response is shown in Figure 1A.

Figure 1

Clinical responses of MP combined with MTX as a first-line therapy in patients with aGvHD. (A) The time course for maximal clinical response of the combination therapy. (B) The representative phenotypes of Th1 and Tc1 cells. Consistent with the clinical responses, the activation, differentiation, and proliferation of T cells in patients were significantly decreased after combined treatment, which was characterized by reduced levels of (C) Th1, (D) Tc1, and (E) T cell proliferation. Data are presented as the means ± SEM (*P ≤ 0.05, **P ≤ 0.01). SSC: side scatter; IFN-γ: interferon γ; Th1: T helper type 1; aGvHD: acute graft-versus-host disease; MP: methylprednisolone; MTX: mini-dose methotrexate; SEM: standard error of the mean.

In the MP monotherapy group, the ORR and CR rate at 7 days after treatment were 58% (16/28 patients) and 50% (14/28), respectively, and the P values were 0.011 and < 0.001, respectively, compared to the combination treatment group. The difference was higher when considering that nearly half of the patients (46%, 13/28) had grade I aGvHD in the MP alone group, whereas the proportion was only 19% (6/31) in the combined treatment group. For grades II–IV aGvHD, the ORR and CR rate at 7 days after treatment were 40% (6/15) vs. 100% (25/25, P < 0.001) and 33% (5/15) vs. 80% (20/25, P = 0.006), respectively.

Immunophenotyping analysis

Consistent with the clinical responses, after treatment with MP and MTX, significantly fewer activated T cells were demonstrated in patients with aGvHD than in those treated with MP alone (Figure 1B–1D), characterized by a reduced proportion of Th1 (Figure 1B; 14.3% ± 1.2% vs. 19.2% ± 1.9%; P = 0.04) and Tc1 (Figure 1C; 22.8% ± 2.9% vs. 32.1% ± 45.5%; P = 0.03) cells in patients with aGvHD on day 7 after treatment. Additionally, compared to T cells in patients with aGvHD treated with MP alone, T cells in aGvHD patients treated with a combination of MP and MTX had decreased proliferation (Figure 1D; 3021.7 ± 461.9 vs. 4496.0 ± 792.3) on day 7 after treatment. Consistently, the frequencies of proinflammatory cells, such as Th1 (Figure 1B; 9.6% ± 0.9% vs. 15.0% ± 1.4%; P = 0.01) and Tc1 (Figure 1C; 13.8% ± 2.2% vs. 24.8% ± 6.1%; P = 0.03) were markedly decreased in the combined treatment group on day 28 after treatment. In addition, degraded T-cell proliferation (Figure 1D; 1846.7 ± 282.8 vs. 3792.3 ± 429.2; P = 0.003) was observed in the combined treatment group on day 28 after treatment.

Side effects

Three of the 31 (9.7%) patients developed severe leukopenia (CTC Grade 4, white blood cell < 1 × 109/L) or severe thrombocytopenia (CTC Grade 4, platelets < 25 × 109/L) compared to baseline value after three, one, and two doses of MTX administration, respectively. The white blood cell counts of both the patients developing severe leukopenia were 4.85 × 10⁹ /L and 4.49 × 10⁹ /L before MTX administration, and the nadir values after MTX administration were 0.63 × 109/L and 0.80 × 109/L, respectively. The platelet count of the single patient developing severe thrombocytopenia was 133 × 109/L before MTX administration, and the nadir value after MTX administration was 24 × 109/L. No patients were withdrawn from the study because of hematologic side effects, and their blood cell counts quickly returned to baseline values with a median of 2 days (1–19 days) after the last dose of MTX. In addition, another 12 patients had increased white blood cell and/or platelet counts for at least one lower CTC grade compared to baseline value after MTX administration. Levels of blood glucose and blood pressure remained normal during treatment. During MTX administration, three patients (9.7%) had CMV and/or EBV antigenemia; other infectious complications or nonhematologic toxicities were not observed.

Follow-up and survival

Up to December 10, 2021, the median follow-up time was 289 (204–421) days from HSCT and 269 days (range: 182–352) from the onset of aGvHD among survivors. Chronic GvHD and moderate-to-severe chronic GvHD occurred in 16 (51.6%) and 6 (19.4%) patients in the MTX group, whereas the incidence was 53.6% (n = 15) and 13.4% (n = 4) in the control group.

Leukemia relapse occurred in six (19.3%) patients at a median of 103 (82–153) days from HSCT, three of whom eventually died from leukemia relapse. One patient (3.2%) died of lung infection at 154 days from HSCT. Twenty-four patients (77%) remained alive without leukemia relapse with a median survival of 287 (204–421) days from HSCT and 263 (182–352) days from the onset of aGvHD. In the control group, leukemia relapse occurred in two (7.1%) patients at 224 and 282 days from HSCT and four patients (14.3%) died of TRM at a median of 136 (204–421) days from HSCT (two from lung infection and two from GvHD). The 6-month FFS was 61% (95% confidence interval [CI]: 44%–78%) for MTX and 32% (15%–49%) for controls.

MP combined with MTX demonstrated synergistic effects in humanized aGvHD mice

To further evaluate the effect of MP and MTX on humanized aGvHD mice model, humanized aGvHD mice were treated with PBS (CTL), MP, and MP and MTX (MP + MTX), respectively. aGvHD occurred in the humanized mice receiving donor G-PBSCs during 1–4 weeks after transplantation, as evidenced by weight loss, disease score (hunching, activity, ruffling, and diarrhea), and death. Similar to patients with aGvHD, humanized aGvHD mice showed severe inflammation, leukocyte infiltration, necrosis, and tissue damage in target organs, such as the skin, lung, liver, and gut, in the G-PBSC group (Figure 2). MP + MTX treatment demonstrated synergistic effects on ameliorating aGvHD, as characterized by improving OS and aGvHD pathological scores in mice compared to those in the CTL or MP alone groups (Figure 2). Consistent with the results of the clinical trial, our data indicated that MP combined with MTX treatment demonstrated better anti-GvHD effect in humanized aGvHD mice model.

Figure 2

MP combined with MTX synergistically alleviated aGvHD in humanized mice. (A) Protocol for the establishment of the humanized aGvHD mouse model and the different treatment groups. (B) Median survival of the recipient mice in different treatment groups. (C) Representative histology of the target organs harvested from the recipient mice of the different treatment groups. (D) Pathological score of the recipient mice in different treatment groups. Data are presented as the means ± SEM (**P ≤ 0.01). SSC: side scatter; IFN-γ: interferon γ; Th1: T helper type 1; aGvHD: acute graft-versus-host disease. MP: methylprednisolone; MTX: mini-dose methotrexate; CTL: treated with PBS; SEM: standard error of the mean.