Bortezomib in late antibody-mediated kidney transplant rejection (BORTEJECT Study): study protocol for a randomized controlled trial

Abstract

Background: Despite major advances in transplant medicine, improvements in long-term kidney allograft survival have not been commensurate with those observed shortly after transplantation. The formation of donor-specific antibodies (DSA) and ongoing antibody-mediated rejection (AMR) processes may critically contribute to late graft loss. However, appropriate treatment for late AMR has not yet been defined. There is accumulating evidence that the proteasome inhibitor bortezomib may substantially affect the function and integrity of alloantibody-secreting plasma cells. The impact of this agent on the course of late AMR has not so far been systematically investigated.

Methods/design: The BORTEJECT Study is a randomized controlled trial designed to clarify the impact of intravenous bortezomib on the course of late AMR. In this single-center study (nephrological outpatient service, Medical University Vienna) we plan an initial cross-sectional DSA screening of 1,000 kidney transplant recipients (functioning graft at ≥180 days; estimated glomerular filtration rate (eGFR) >20 ml/minute/1.73 m2). DSA-positive recipients will be subjected to kidney allograft biopsy to detect morphological features consistent with AMR. Forty-four patients with biopsy-proven AMR will then be included in a double-blind placebo-controlled intervention trial (1:1 randomization stratified for eGFR and the presence of T-cell-mediated rejection). Patients in the active group will receive two cycles of bortezomib (4 × 1.3 mg/m2 over 2 weeks; 3-month interval between cycles). The primary end point will be the course of eGFR over 24 months (intention-to-treat analysis). The sample size was calculated according to the assumption of a 5 ml/minute/1.73 m2 difference in eGFR slope (per year) between the two groups (alpha: 0.05; power: 0.8). Secondary endpoints will be DSA levels, protein excretion, measured glomerular filtration rate, transplant and patient survival, and the development of acute and chronic morphological lesions in 24-month protocol biopsies.

Discussion: The impact of anti-humoral treatment on the course of late AMR has not yet been systematically investigated. Based on the hypothesis that proteasome inhibition improves the outcome of DSA-positive late AMR, we suggest that our trial has the potential to provide solid evidence towards the treatment of this type of rejection.

Trial registration: Clinicaltrials.gov: NCT01873157.

Figure 1

Study flowchart of Part A (cross-sectional screening). All kidney transplant recipients from our outpatient service with a functioning allograft at 6 months and estimated glomerular filtration rate >20 ml/minute/1.73 m² will be considered for study inclusion. AMR, antibody-mediated rejection; DSA, donor-specific antibodies.

Figure 2

Study flowchart of Part B (randomized controlled trial). Forty-four transplant recipients with late biopsy-proven antibody-mediated rejection (AMR) will be randomized to receive either bortezomib or placebo. The primary endpoint, the estimated glomerular filtration rate (eGFR), will be evaluated at 0, 6, 12, 18 and 24 months. Major secondary endpoints are the measured glomerular filtration rate (GFR), protein excretion, patterns of human leukocyte antigen reactivity and results obtained with 24-month protocol biopsies. Cr-EDTA, chromium ethylenediamine tetraacetic acid; DSA, donor-specific antibodies; MFI, mean fluorescence intensity.

Figure 3

Course of kidney allograft function in relation to late biopsy results. In a retrospective cohort analysis, 344 consecutive long-term kidney allograft recipients with a functioning graft at 4 years (transplantation at the Vienna transplant unit) were evaluated for the course of estimated glomerular filtration rate (eGFR). Kidney function is shown in relation to C4 complement split product deposition (C4d) staining results in late indication biopsies performed >6 months after transplantation.

Figure 4

Sample size determination and power calculation. Power calculation using five different assumptions of correlation in a covariance matrix. Simulation analyses using 100 permutations of randomly selected 2 × 22 samples of first and second transplants respectively from the Austrian Dialysis and Transplant Registry and the earlier described assumption underline the robustness of our analyses. Seventy-seven of the 100 permutations showed statistically significant delta estimated glomerular filtration rate (eGFR) thresholds of 5 ml/minute/year. Ar1, first-order autoregressive model.