The phase 3 GRAPHITE study showed that vedolizumab plus standard prophylaxis after unrelated allogeneic hematopoietic stem cell transplantation was more effective vs placebo for the prevention of lower gastrointestinal graft-vs-host disease.
Treatment with vedolizumab (Entyvio) plus standard prophylaxis following unrelated allogeneic hematopoietic stem cell transplantation (HSCT) showed superiority vs placebo at preventing lower gastrointestinal (GI) acute graft-vs-host disease (aGVHD), according to findings from the global, phase 3 GRAPHITE study (NCT03657160) presented at the 2023 Transplantation & Cellular Therapy Meetings.1
“Vedolizumab was more effective than placebo for the prevention of lower GI GVHD after unrelated allogeneic transplant when added to a standard calcineurin inhibitor–based GVHD platform,” said Yi-Bin Chen, MD, director of the Hematopoietic Cell Transplant & Cell Therapy Program and the Allen B. Rogers Jr and Cara J. Rogers Endowed Chair at Massachusetts General Hospital.
“Its safety profile was comparable to placebo with no new safety signals identified, given as part of GVHD prophylaxis,” he added. “This is the first positive phase 3 study for this specific prevention of lower GI GVHD.”
In patients assigned to vedolizumab, the rate of intestinal aGVHD-free survival by 180 days following allogeneic HSCT, the primary end point, was 85.52% (95% CI, 79.17%-90.05%) vs 70.85% (95% CI, 61.63%-77.22%) for those assigned to placebo (HR, 0.45; 95% CI, 0.27-0.73; P < .001). Sixteen (9.7%) patients died in the placebo arm vs 12 (7.1%) in the experimental arm. Fourteen (8.5%) patients in the placebo arm developed stage 2 to 4 intestinal aGVHD compared with 4 (2.4%) for vedolizumab.
In the sensitivity analysis, which excluded clinical stage 0 lower GI GVHD events, 13.7% of patients in the experimental arm had GI aGVHD by day 180 following allogeneic HSCT, compared with 27.3% in the placebo arm (HR, 0.44; 95% CI, 0.27-0.73; P = .001).
Despite progress in prophylaxis and treatment, approximately 40% to 70% of patients who undergo allogenic HSCT will develop grade 2 to 4 aGVHD. Investigators have found mixed results in small studies assessing vedolizumab, a gut-selective anti-ß integrin antibody that inhibits migration of Gl-homing T lymphocytes across the gut endothelium, as a preventative for aGVHD. The agent is currently approved to treat ulcerative colitis and Crohn’s disease.2
Investigators conducted this global investigation from February 2019 through May 2022 at 94 centers in North and South America, Europe, Asia, and Australia. Investigators hoped to enroll 558 patients, but the study closed after enrolling 343 patients because the COVID-19 pandemic hampered recruitment. A total of 169 patients were randomly assigned to placebo and 165 were included in the safety and efficacy populations. In the vedolizumab arm, 169 patients were evaluated for safety and 168 for efficacy.
Patients aged 12 years or older undergoing a first allogeneic HSCT for treatment of hematological malignancies from unrelated donors were eligible. Unrelated donors were human leukocyte antigen (HLA)–matched (8/8) or 7/8 matched (a single mismatch at HLA-A, -B and -C, and HLA-DRB1) were allowed. All patients received a standard GVHD prophylaxis regimen consisting of a combination of a calcineurin inhibitor plus methotrexate or mycophenolate mofetil.
All patients were assigned to standard of care prophylaxis on day 13, day 41, day 69, day 97, day 125, and day 153 after allogeneic HSCT plus placebo (n = 169) or intravenous 300 mg vedolizumab on day –1 before allogeneic HSCT (n = 174).
In the placebo group, the median age was 55.0 years (range, 16-74), 64.2% of patients were male, and the most common primary disease was acute myeloid leukemia (AML; 43.6%). More than half (52.4%) of patients had an ECOG performance status of 1, 39.6% had an ECOG performance status of 0, and 7.9% had a score of 2.
Stem cells came from the peripheral blood in 86.6% of patients, and 90.9% were 8/8 matched for HLA compatibility. Eighty-four percent of patients were in complete remission 1.
In the experimental group, the median age was 53.0 years (range, 19-74), 61.3% of patients were male, and the most common primary disease was AML (43.5%). More than half (53.0%) of patients had an ECOG performance status of 1, 38.1% had an ECOG performance status of 0, and 8.3% had a score of 2.
Stem cells came from the peripheral blood in 83.9% of patients and 90.5% were 8/8 matched for HLA compatibility. More than two-thirds of patients (71.4%) patients were in complete remission 1.
Chen noted that vedolizumab induced superior results for the primary end point, irrespective of conditioning, prophylaxis, calcineurin inhibitor, or HLA match.
Vedolizumab outperformed placebo across key secondary end points including intestinal aGVHD-free and relapse-free survival events (20.8% for vedolizumab vs 33.9% for placebo; HR, 0.56; 95% CI, 0.37-0.86; P =.0043), grade C-D aGVHD-free survival events (20.8% vs 31.5%; HR, 0.59; 95% CI, 0.39-0.91; P =.0204), non-relapse mortality events (6.0% vs 11.5%; HR, 0.48; 95% CI, 0.22-1.04; P = .0668), overall survival events (10.1% vs 15.2%; HR, 0.63; 95% CI, 0.34-1.17; P = .1458), and grade B-D aGVHD survival events (33.3% vs 46.7%; HR, 0.64; 95% CI, 0.46-0.91; P .0105). Intestinal aGVHD-free and relapse free survival events excluding clinical grade 0 patients also favored vedolizumab (20.2% vs 32.7%; HR, 0.56; 95% CI, 0.36-0.86; P = .0062).
All patients in both groups experienced any-grade adverse effects (AEs). Grade 3 or higher AEs were 89.1% in the placebo group and 92.3% in the experimental arm. Drug-related grade 3 or higher AEs occurred in 11.5% and 10.7% of patients in the vedolizumab and placebo arms, respectively. AEs leading to discontinuation occurred in 30.9% of patients in the placebo arm and 26.0% in the experimental arm.
Serious AEs leading to discontinuation occurred in 23.0% of patients in the placebo arm and 23.1% in the experimental arm. Drug-related serious AEs occurred in 8.5% and 6.5% of patients, respectively. Twenty-seven (16.4%) patients in the placebo arm experienced AEs leading to death compared with 21 (12.4%) in the experimental arm.
Approximately two-thirds (67.3%) of the placebo arms experienced serious infections, 82.4% experienced hypersensitivity/injection site reactions, and 41.8% experienced liver injury. Those results were 74.0%, 79.3%, and 40.2% in the vedolizumab arm, respectively.
Chen said that vedolizumab should be evaluated with combination with post-transplant cyclophosphamide. He added that vedolizumab requires more evaluation to define its optimal role in treatment.
“Some further analysis we'll need to do is…look at maximum stage of GI GVHD to truly understand the impact of what we presented here and also perhaps the impact on chronic GVHD, as well,” he said. “With the changing landscape of GVHD prevention, as we’ve seen recently and even at this meeting, it’s unclear how to incorporate this and the other new agents into what is a rapidly changing landscape, which is interesting. It's also great for our field.”