T-cell costimulation blockade with abatacept for acute GvHD: Results of the ABA2 trial

There are still no approved agents for preventing severe acute (a) graft-versus-host disease (GvHD), which is a major cause of non-relapse mortality (NRM). For patients undergoing hematopoietic stem cell transplantation, especially human leukocyte antigen (HLA)-mismatched unrelated donor transplant, severe aGvHD remains a significant cause of morbidity and mortality.

In an article published in the Journal of Clinical Oncology, Benjamin Watkins and colleagues describe the results of the phase II ABA2 trial (NCT01743131), which investigated adding abatacept, a T-cell costimulation blocker currently approved for rheumatoid arthritis, to a calcineurin inhibitor and methotrexate (CNI/MTX) for GvHD prophylaxis.¹

Study design

Patients and treatment

Three cohorts were included in this study (see also Table 1). Patients undergoing transplant with stem cells from 8/8 HLA-matched unrelated donors (MUD study cohort, n = 148 enrolled) were randomly assigned (1:1) to receive CNI/MTX plus placebo or abatacept. Patients transplanted with 7/8 HLA-mismatched donor stem cells (MMUD study cohort, n = 46 enrolled) received CNI/MTX plus abatacept, and these were compared with a prespecified control cohort from the Center for International Blood and Marrow Transplant Research (CIBMTR).

Patients were given:

• Four doses of abatacept: 10 mg/kg on Days -1, +5, +14, and +28, and/or
• CNI (tacrolimus or cyclosporine): given as standard through Day +100 and weaned between Days +100–180 as tolerated, plus
• MTX: 15 mg/m² on Day +1 and 10 mg/m² on Days +3, +6, and +11.

Key endpoints

• Primary endpoint: Cumulative incidence of severe (Grade III–IV) aGvHD at Day +100.
• Secondary endpoint: Day +180 severe aGvHD-free survival (SGFS).

Other endpoints included Grade III–IV aGvHD at Day +180; Grade II–IV aGvHD at Days +100 and +180; cGvHD at 1 year; NRM; relapse; relapse-free survival (RFS); overall survival (OS); cytomegalovirus reactivation and disease; Epstein-Barr virus reactivation and post-transplant lymphoproliferative disease; hematologic recovery; and donor engraftment.

Patient and transplant characteristics

Table 1. Baseline patient and transplant characteristics¹

The most common malignancy experienced in this patient population was acute myeloid leukemia, followed by acute lymphoid leukemia and myelodysplastic syndromes. As listed in Table 1, most patients received myeloablative conditioning. Controls were well matched for patient characteristics, and the majority of patients received tacrolimus as CNI – this was especially prominent in the 8/8 cohort.  

Sex matching of donors and recipients was achieved in roughly 50% of patients across all groups, and graft source was split between peripheral blood and bone marrow across the whole population. However, in the historical CIBMTR control group, data on sex matching was missing in 28% of patients and data on the donor-recipient Epstein-Barr virus status was not available.

Key points

• Abatacept was administered in combination with CNI/MTX to 43 patients in the MMUD cohort and 73 patients in the MUD group.
• To compare data between MMUD transplanted patients, a group of 38 patients out of 43 who had received abatacept was matched to the CIBMTR control group.
• Addition of abatacept significantly reduced the incidence of severe (Grade III−IV) aGvHD in MMUD transplanted patients when compared to the historical CIBMTR controls (p < 0.001).
  • Interestingly, improvement in Grade III−IV aGvHD with abatacept was only significant for patients who had been transplanted with MMUD.
• A benefit was also seen in the MUD cohort randomized to abatacept for reduced incidence of aGvHD Grades II−IV (p = 0.006)(Table 2).
• Similarly, improvements in RFS, NRM, and OS after 2 years were significantly different in the MMUD groups comparing patients treated with abatacept versus historical matched controls. No difference was found for the MUD cohorts, possibly because the study was not powered to show a difference for these outcomes.
• Early treatment failure was assessed by investigating SGFS at Day +180. Again, addition of abatacept significantly improved outcomes in both MMUD and MUD transplanted patients when compared with a historical control/placebo group (MMUD, p < 0.001; MUD, p = 0.05)

Table 2. Outcome of treatment¹

• Addition of abatacept did not affect the incidence or severity of chronic GvHD, with a comparable percentage of patients experiencing mild−severe symptoms in both the treatment and control/placebo groups (MMUD: HR, 1.26; 80% CI, 0.97−1.64; p = 0.74 and MUD: HR, 1.17; 80% CI, 0.85−1.61; p = 0.55).
• Successful donor engraftment, neutrophil and platelet restoration, and leukocyte reconstitution were achieved in all cohorts. In addition, there was no significant difference in cytomegalovirus or Epstein-Barr virus reactivation in MUD transplanted patients. These data were unavailable for the MMUD cohorts.
• There was no significant difference in the rate of relapse in the abatacept group and control/placebo group (MMUD: HR, 0.45; 80% CI, 0.22−0.91; p = 0.21 and MUD: HR, 0.86; 80% CI, 0.54−1.35; p = 0.66).
• Using flow cytometry following treatment levels, investigators found that GvHD prophylaxis with abatacept significantly preserved the proportions of naïve T cells.

Conclusion

Safe and effective GvHD prophylaxis strategies for patients without HLA-matched sibling donors are desperately needed. In ABA2, abatacept was shown to reduce the incidence of aGvHD in patients receiving MMUD and MUD transplants and greatly improve survival outcomes for the MMUD cohort when compared to a matched historical control. The safety profile for this treatment was also shown to be acceptable. However, using a four-dose regimen of abatacept in this trial was not shown to have any impact on the incidence of chronic GvHD. In order to assess if a longer dosing schedule would improve the risk of cGvHD development, further trials are necessary.