Treatment of chronic graft-versus-host disease with bortezomib

Abstract

Chronic graft-versus-host disease (cGVHD) following allogeneic hematopoietic stem cell transplantation (HSCT) has emerged as a predominant complication following HSCT and has a distinct etiology. We and others have previously demonstrated that bortezomib, a proteasome inhibitor, can prevent but not treat acute GVHD in mice. To assess the effects of bortezomib on cGVHD, a mouse minor histocompatibility antigen-mismatched strain combination was used to mimic clinical cGVHD sclerodermatous pathogenesis and phenotype. Treatment of ongoing cGVHD with bortezomib ameliorated cutaneous lesions, which were also associated with a reduction in total numbers of germinal center B cells and lower B-cell activating factor gene expression levels in cutaneous tissues. Importantly, lymphoma-bearing mice receiving allogeneic HSCT with bortezomib preserved graft-versus-tumor (GVT) effects. Based on these animal studies, we initiated an intrapatient dose escalation clinical trial in patients with extensive steroid-intolerant, dependent, or resistant cGVHD. Marked clinical improvement was observed in patients, which was also associated with reductions of peripheral B cells and minimal toxicity. These results indicate that bortezomib can be of significant use in the treatment of cGVHD and may also allow for maintenance of GVT. This trial was registered at www.clinicaltrials.gov as #NCT01672229.

Figure 1

Therapeutic bortezomib administration protects mice from sclerodermatous cGVHD responses. Irradiated (800 cGy) recipient BALB/c mice received bone marrow cells (8 million) with or without spleen cells (25 million) intravenously from donor B10.D2 mice. Mice were then randomized allocated to treat with either vehicle (PBS) or bortezomib (0.1 mg/kg) intraperitoneally at day 20 after transplant and every 5 days thereafter. (A) Skin clinical scores (on a scale of 3.9) were evaluated twice a week. (B) Photographs were taken at day 55 after HSCT from either bone marrow only or GVHD mice treated with vehicle or bortezomib at day 20. (C) (Upper) Pathologic examination of skin by hematoxylin and eosin stain. (Lower) Collagen deposition and fibrosis were examined by Masson’s trichrome stain. (D) Pathological scores (on a scale of 10) were evaluated by pathologists in a blind code fashion. Data are shown as mean ± standard error of the mean (SEM) and analyzed by 1- or 2-way ANOVA with a Tukey post hoc test to compare between individual groups. *P < .05, **P < .01, and ***P < .001 were considered significant. Data were collected from 2 independent experiments with 8 mice per group.

Figure 2

Time-dependent administration of bortezomib produces differential scleroderma GVHD responses. Irradiated (8 Gy) BALB/c recipient mice were transplanted with bone marrow cells with or without spleen cells from donor B10.D2 mice. (A) Bortezomib or vehicle administration schema during cGVHD pathogenesis. (B) Body weight changes among different bortezomib regimen groups. (C) Comparison between early therapeutic bortezomib treatments vs vehicle control groups. (D) Comparison between vehicle control groups vs delayed bortezomib treatment groups. (E) Comparison between continuous therapeutic bortezomib treatments vs intermittent therapeutic bortezomib treatment groups. All the groups were tested simultaneously and repeated twice with 8 mice per group. Bortezomib was given at a dose of 0.1 mg/kg for all the groups except bone marrow only and vehicle control groups. The data are shown as mean ± SEM and were analyzed by 2-way ANOVA with a Tukey post hoc test to compare between groups. *P < .05 and ***P < .001 were considered significant.

Figure 3

B-cell population decreases after bortezomib treatment. Irradiated BALB/c mice transplanted with bone marrow and spleen cells were injected with either bortezomib (0.1 mg/kg) or vehicle control starting at day 20. (A) Spleen cells were collected at day 55 and stained for CD45⁺CD19⁺ populations. (B-C) Data showing B-cell populations as percentage and total numbers from spleen. (D) Total numbers of CD45⁺CD19⁺MHC II⁺ B-cell populations in the skin samples harvested on day 55. (E) reverse transcriptase-polymerase chain reaction results detecting BAFF gene expression levels from skin samples collected at day 55. (F) Serum BAFF levels were detected by enzyme-linked immunosorbent assay and calculated as log (BAFF/B cell) ratio. (G) Total numbers of CD45⁺B220⁺PNA⁺ germinal center B cells in the spleen. All the data were collected from 2 independent experiments with 8 mice per group. The data are shown as mean ± SEM and analyzed by Student t test to compare between individual groups. *P < .05 and **P < .01 were considered significant.

Figure 4

T-cell engraftment and Treg cell populations after BMT. To evaluate the effects of bortezomib on T cells, irradiated Balb/c mice were transplanted with bone marrow cells with or without spleen cells and treated with bortezomib (0.1 mg/kg) or vehicle control. Data were collected from 2 to 3 independent experiments with ≥8 mice per group. (A-B) Spleen cells were isolated at either day 11 or day 56 and analyzed for chimerism (Ly9.1⁺ stain for recipient-derived cells) by flow cytometry. (C-D) Thymus engraftment was analyzed by double-positive T cells (CD4⁺CD8⁺) at day 56 by flow cytometry. (E) Cellularity in secondary lymphoid organs at day 56. (F) Spleens were harvested and analyzed for Treg cell populations. (G) Total numbers of Treg cells. (H) Serum IL-6 levels were detected by cytometric bead array. All data were shown as mean ± SEM and analyzed by 1-way ANOVA with a Tukey post hoc test to compare between individual groups. *P < .05, **P < .01, and ***P < .001 were considered significant. Data were collected from 2 independent experiments with 8 mice per group.

Figure 5

Bortezomib administration allows for maintenance of GVT effects while decreasing cGVHD skin lesions in A20 tumor models. Irradiated BALB/c mice were transplanted with bone marrow cells with or without spleen cells at day 0. Six hours later, A20 luciferase-transfected lymphoma cells (1 × 10⁶) were injected through the tail vein into the indicated groups. Bortezomib was administered at day 20 and every 5 days thereafter. (A) Timeline schema for different conditions among groups. (B) Bioluminescent images were acquired to monitor tumor burdens. (C) Survival curves from experimentally treated groups in the cGVHD model. (D) Survival curves from different groups challenged by A20 tumor cell lines at day 0. (E) Skin clinical scores were evaluated twice a week. Data were collected from 1 experiment with 8 mice per group. The data are shown as mean ± SEM and analyzed by 2-way ANOVA with a Tukey post hoc test to compare between individual groups. Survival data were plotted by the Kaplan-Meier method and analyzed by the log-rank test. **P < .01, ***P < .001, and ****P < .0001 were considered significant.

Figure 6

Bortezomib administration allows for maintenance of GVT effects while decreasing cGVHD skin lesions in P815 tumor models. Irradiated BALB/c mice were transplanted with bone marrow cells with or without spleen cells at day 0. P815-luciferase-transfected mastocytoma cells (6 × 10⁵) were injected through the tail vein into the indicated groups at day 20 when bortezomib treatment was initiated. (A) Timeline schema for different conditions among groups. (B) Bioluminescent images were acquired to monitor tumor burdens. (C) Survival curves from experimentally treated groups. (D) Skin clinical scores were evaluated twice a week. Data were collected from 1 experiment with 8 mice per group. The data are shown as mean ± SEM and analyzed by 2-way ANOVA with a Tukey post hoc test to compare between individual groups. Survival data were plotted by the Kaplan-Meier method and analyzed by the log-rank test. **P < .01, ***P < .001, and ****P < .0001 were considered significant.

Figure 7

Treatment effects of bortezomib on clinical cGVHD human patients. A single institution pilot study of bortezomib was initiated in patients with steroid-dependent, -intolerant, or -refractory cGVHD. (A) Patient 4 showed extensive grade III skin sclerodermatous GVHD covering >0% of the body. The abdominal region before and after bortezomib treatments are shown. (B) Representative images of the pretreatment skin biopsies taken from the patient shown in A. (C) Immunohistochemical staining for CD3 and CD20 in pretreatment skin biopsy samples from patient 4. (D) CBC and biochemistry data from patient 5 were collected through the trial period. (E) Total numbers of peripheral blood B cells (CD45⁺CD19⁺) from 3 patients were analyzed by flow cytometry before and after bortezomib treatment. (F-G) Treg cell populations (CD4⁺CD25⁺Foxp3⁺) were analyzed by flow cytometry and shown as total numbers and percentage. All the data were collected from individual cGVHD patients that underwent bortezomib treatment. The data are shown as mean ± SEM and analyzed by Student t test to compare pre- and postbortezomib treatments. *P < .05 was considered significant.