Clinical outcomes of bortezomib-based desensitization in highly sensitized living and deceased donor kidney transplantation

Article information

Korean J Nephrol. 2025;.j.krcp.24.077

Publication date (electronic) : 2025 February 19

doi : https://doi.org/10.23876/j.krcp.24.077

Hyeran Park^1,2, Hanbi Lee^1,2, Sang Hun Eum^2,3, Ji-Won Min^2,4, Hye Eun Yoon^2,3, Eun-Jee Oh⁵, Chul Woo Yang^1,2, Byung Ha Chung^,1,2

¹Division of Nephrology, Department of Internal Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

²Transplantation Research Center, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

³Division of Nephrology, Department of Internal Medicine, Incheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

⁴Division of Nephrology, Department of Internal Medicine, Bucheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

⁵Department of Laboratory Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

Correspondence: Byung Ha Chung Division of Nephrology, Department of Internal Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 06591, Republic of Korea. E-mail: chungbh@catholic.ac.kr

Received 2024 March 19; Revised 2024 May 8; Accepted 2024 May 20.

Abstract

Introduction

Kidney transplantation (KT) remains the best therapeutic option for end-stage kidney disease [1,2]. However, the presence of alloantibody to human leukocyte antigen (HLA), so-called “sensitization to HLA”, creates an important immunologic barrier against successful KT [3]. Data from the United States has shown that up to 35% of patients on the waiting list for a transplant are sensitized [4]. The situation is similar in Korea, as 15.4% of patients on the waiting list for KT were shown to be highly sensitized to HLA in terms of a positive crossmatch test [5]. KT in patients with strong donor-specific anti-HLA antibody (HLA-DSA) without any treatment can induce severe antibody-mediated rejection (ABMR), which can result in immediate allograft loss, and even the presence of weak HLA-DSA at baseline can be associated with higher risk for the development of ABMR and lower allograft survival [6,7]. Therefore, proper “desensitization (DSZ) therapy” to remove HLA-DSA before KT is crucial [3,4].

The principle of the current DSZ technique is to remove preexisting antibodies and inhibit the production of alloantibodies, and the most widely used protocol is based on rituximab (RTX), intravenous immunoglobulin (IVIg), and plasmapheresis (PP) [3,4]. Even though this protocol has proven effective in many previous studies, it has been shown to have limited effects in certain cases, particularly in patients with very strong HLA-DSA and positive T-complement dependent cytotoxicity (T-CDC) crossmatch test [8]. This may be attributable to the fact that the treatment only depletes B cells or removes circulating antibodies and is not effective in suppressing plasma cells that directly produce HLA-DSA [9]. Meanwhile, bortezomib (BOZ) is a proteasome inhibitor that was introduced for use in KT several years ago [10]. BOZ inhibits antibody production from plasma cells, stimulates apoptosis of this cell type, and decreases the number of bone marrow-derived plasma cells [11]. Therefore, it is expected that this drug will suppress humoral immunity better than conventional therapies such as RTX.

In our previous preliminary report, we found that BOZ-based DSZ (BOZ-DSZ) was effective in three living donor KT (LDKT) with strong HLA-DSA [12]. After that, we set up BOZ-DSZ protocols appropriate for LDKT and deceased donor KT (DDKT), respectively, through combination with IVIg, PP, and RTX, based on previous studies [12,13]. In the current study, we investigate the efficacy and safety of BOZ-DSZ in both LDKT and DDKT cases in terms of the mean fluorescence intensity (MFI) reduction of HLA-DSA, success rate to enter transplantation, and posttransplant outcomes including biopsy-proven allograft rejection (BPAR) rate, infectious complication, and allograft survival.

Methods

Desensitization protocol

The BOZ-DSZ protocol was established based on a previous study [12,13]. In LDKT, the goal of DSZ was to achieve negative conversion of the crossmatch test for patients who tested positive and a reduction of the HLA-DSA MFI value to <5,000 by LSA assay [15]. At 18 days before transplantation, we administered a 375 mg/m² dose of RTX, and PP/IVIg was performed eight or more times until the HLA-DSA MFI level approximated or dropped below the target level (MFI <5,000) (Fig. 1A). At each PP session, one plasma volume was removed and substituted with either 5% albumin or fresh frozen plasma, and IVIg (100 mg/kg) was infused 1 hour after each PP session. BOZ at a dose of 1.3 mg/m² was given intravenously on pretransplant days 15, 12, 8, and 5 after performing PP. On the other hand, in DDKT, BOZ-DSZ was conducted on an outpatient basis according to a DDKT protocol for 31 days (Fig. 1B). High-dose IVIg at a dosage of 1 g/kg/day was administered on DSZ days 1, 2, 30, 31 followed by a 375 mg/m² dose of RTX on DSZ day 3. BOZ at a dosage of 1.3 mg/m² was given on DSZ days 15, 18, 21, and 24. The HLA-DSA tests were performed both before and after DSZ treatment to assess the effectiveness of the DDKT protocol.

Figure 1.

BOZ-based desensitization protocol.

(A) In living donor kidney transplantation (KT), rituximab (RTX) was administered on pretransplant day 18 and plasmapheresis (PP)/ intravenous immunoglobulin (IVIg) was performed eight times. Intravenous administration of BOZ was initiated on pretransplant days 15, 12, 8, and 5 after the PP. We initiated immunosuppressant treatment 7 days prior to transplantation. (B) In deceased donor kidney transplantation, high-dose IVIg was administered on days 1, 2, 30, and 31 of desensitization treatment, followed by a 375 mg/m² dose of RTX on day 3. BOZ was given on days 15, 18, 21, and 24 of desensitization treatment.

BOZ, bortezomib; HLA-DSA, donor-specific anti-human leukocyte antigen antibody.

Immunosuppressant treatment

The typical immunosuppressant treatment (IST) regimen conducted at our center has been described previously [17]. In LDKT, tacrolimus-based triple IST (tacrolimus, mycophenolate mofetil, prednisolone) began with the initiation of pretransplant DSZ (Fig. 1A). The target through level of tacrolimus for posttransplant at the first month was 8 to 12 ng/mL, while the dose of mycophenolate mofetil during the same period was 1.5 g/day. The dose of prednisolone was set at 20 mg/day at the time of discharge. Among 13 patients who underwent LDKT, 11 patients (84.6%) received induction IST therapy with a 1.25 mg/kg dose of antithymocyte globulin (ATG) for 5 days after KT, and two patients received a 25 mg dose of basiliximab on the day of transplantation and 4 days after KT as induction IST therapy (Table 1).

Table 1.

Baseline characteristics

In DDKT, we initiated tacrolimus-based IST immediately after transplantation while maintaining the same dosage and trough level as the LDKT protocol. All recipients in DDKT received induction therapy with ATG after KT.

Results

Change in HLA-DSA mean fluorescence intensity after bortezomib-desensitization and success rate to enter transplant

In LDKT, the MFI level of HLA-DSA was found to significantly decrease approximately to or below the target level (MFI < 5,000) in 10 out of 14 patients (p < 0.05) (Fig. 2A). Out of 10 patients (patients 1–10) who took BOZ-DSZ due to positive T-CDC-AHG crossmatch at baseline, eight achieved a negative conversion of crossmatch test after BOZ-DSZ. However, two patients (patients 6 and 7) did not show negative conversion, but their DSA MFI had been successfully reduced to zero and 1,352, respectively, allowing transplantation to proceed. Additionally, three did not achieve an MFI below 5,000 (patient 4, 7,176; patient 8, 6,339; patient 10, 6,336); however, all showed a decreasing trend and it was anticipated that their levels would reach the target by the day of transplantation, allowing the procedures to proceed. Four patients (patients 11–14) who took BOZ-DSZ due to being refractory to RTX/PP-based DSZ showed a marginal decrease in the MFI value of HLA-DSA, but transplantation proceeded in these patients as their HLA-DSA targeted DQ (Fig. 2B). In terms of success rate to transplant, among all patients who received BOZ-DSZ, only one patient did not proceed to KT because of general weakness provoked by tacrolimus neurotoxicity; therefore, 13 out of 14 patients (92.9%) in LDKT proceeded to transplantation.

Figure 2.

Changes in HLA-DSA MFI levels after BOZ-based DSZ.

(A, B) Change in the peak MFI level of immunodominant HLA-DSA after BOZ-DSZ in living donor kidney transplantation candidates. (C, D) Change in the peak MFI level of immunodominant HLA-DSA after BOZ-DSZ in deceased donor kidney transplantation candidates.

BOZ, bortezomib; DSZ, desensitization; HLA-DSA, donor-specific anti-human leukocyte antigen antibody; MFI, mean fluorescence intensity.

In DDKT, the peak MFI level of HLA-DSA decreased after DSZ in five out of six patients (p = 0.098) (Fig. 2C). Out of them, patients 1, 2, and 3 yielded negative results in the T-cell crossmatch test for deceased donors, despite the fact that patient 2 demonstrated a peak MFI level of HLA-DSA that did not decrease after BOZ-DSZ (Fig. 2D). Consequently, 50.0% of the patients (3 of 6) underwent DDKT. The time taken from BOZ-DSZ to receiving a DDKT was 3 months, 7 months, and 2 months, respectively, averaging 4 months.

Posttransplant allograft function recovery and bleeding complications

None of the 16 cases showed hyperacute rejection after KT. In LDKT, allograft function recovered immediately, maintaining the mean value of estimated glomerular filtration rate (eGFR) of 66.3 ± 20.8 mL/min/1.73 m² at the first posttransplant year (Fig. 3A). In DDKT, all recipient’s allograft function also recovered immediately, and no delayed graft function was found. The mean value of eGFR was 82.4 ± 29.0 mL/min/1.73 m² at the first posttransplant year. No peri- or postoperative bleeding events requiring blood transfusion were observed in any of the patients who underwent transplantation after BOZ-DSZ in either LDKT or DDKT.

Figure 3.

Posttransplant clinical outcomes in LDKT and DDKT.

(A) Change in allograft function. (B) Incidence and type of biopsy-proven allograft rejection (BPAR). (C) Incidence and subtype of infection.

ABMR, antibody-mediated rejection; BKV, BK polyomavirus; CMV, Cytomegalovirus; DDKT, deceased donor kidney transplantation; eGFR, estimated glomerular filtration rate; LDKT, living donor kidney transplantation; TCMR, T cell-mediated rejection; UTI, urinary tract infection.

Discussion

We investigated the efficacy and safety of BOZ-DSZ in both highly sensitized LDKT and DDKT candidates. The most important finding to note in this study is that BOZ-DSZ is not only effective in achieving an acceptable immunologic status for KT but also safe in terms of posttransplant bleeding or infectious complications. Our results suggest that BOZ-DSZ can be safely recommended for DSZ in highly sensitized patients both in LDKT and DDKT.

First, we intended to establish an effective BOZ-DSZ protocol that was appropriate for both LDKT and DDKT based on previous studies [12,13]. Prior research reported that proteasome inhibitor-based combination therapy provides a potential means for rapid DSA elimination in ABMR [18]. We reported on the efficacy of BOZ in three DSZ cases before KT and six ABMR treatments for LDKT recipients in 2014 [12]. In 2017, another study was published indicating that the largest trial involving 44 patients over 3 years used a combination of PP, a single RTX dose, and multiple doses of 1.3 mg/m² of BOZ, ultimately leading to a substantial reduction of anti-HLA levels in 86% of KT candidates [19,20]. In DDKT, a prospective, open-labeled clinical trial published in 2016 indicated that PRA was reduced by 15.5% after BOZ-DSZ in the highly sensitized 19 patients awaiting DDKT, and the transplant rate reached 42.1% in the treated group, compared to the corresponding value of 23.5% in the control group [13].

Our next consideration was to investigate whether BOZ-DSZ was an appropriate protocol for each of LDKT and DDKT, and we ultimately observed a reduction of HLA-DSA in both LDKT and DDKT. The degree of decrease of HLA-DSA is far more significant in both groups. However, the success rate of entering KT was higher in LDKT than it was in DDKT. This may be attributable to the difference in the protocol. In LDKT, since the rapid elimination of HLA-DSA is important, we used PP, which enabled the rapid elimination of HLA-DSA to reach a target level within a specified period, i.e., until the scheduled date of KT, so we used a DSZ protocol including PP in LDKT. By contrast, in the case of DDKT, since the exact date of KT is not yet known, it is more important to maintain the decrease in HLA-DSA for a long time without rebound, so we used a protocol including high-dose IVIg instead of PP in DDKT. In this context, the lower success rate of entering KT in DDKT can be attributed to the unpredictability of the timing of transplantation in DDKT. It is possible that the DDKT candidate may not receive a chance to undergo KT while the effects of BOZ-DSZ are still present, thereby negatively impacting the success rate. It therefore seems necessary to develop a DSZ strategy that provides a longer duration of DSZ effects for patients awaiting DDKT with high sensitization.

Meanwhile, the decrease in the MFI value of HLA-DSA showed a differential pattern according to the HLA-DSA type. The patient group that was refractory to RTX/PP-based DSZ in LDKT showed a marginal decrease in HLA-DSA after BOZ-DSZ, the specificity of which was HLA-DQ (Fig. 2B). Nevertheless, all of these patients proceeded successful KT, which is attributed to the fact that HLA-DQ mismatched kidney has been reported to be associated with a ‘late rejection’ rather than ‘hyperacute’ or ‘acute rejection’ after KT [21]. As a result, all 16 KT recipients did not experience hyperacute rejection and no peri- or postoperative bleeding events requiring blood transfusion were observed. Further, the recovery of allograft function from transplantation to discharge was steady in both LDKT and DDKT.

There is a potential concern about the high risk for allograft rejection in highly sensitized patients, indeed, half of all patients (eight out of 16) were confirmed to have BPAR in this study, and as expected, most of them were ABMR (six out of eight). Although there was no significant difference in the nonparametric test between LDKT and DDKT, the small sample size made statistical comparison challenging. However, it is noteworthy that among the DDKT recipients, only one out of three cases showed evidence of TCMR, and there were no instances of ABMR. The main reason for this may be the lower immunologic risk at transplant in DDKT compared to LDKT. We proceeded with KT in the case of DDKT only when T-CDC-AHG was shown to be negative, so only 50.0% (3 of 6) of patients who took BOZ-DSZ could take transplantation. Regarding the response to antirejection therapy, all patients showed successful recovery of allograft function to a baseline level and maintained stable allograft function with an eGFR of 60 or above afterward without recurrence of BPAR. Interestingly, none of the patients had de novo DSA at the time of graft biopsy, but six out of all eight BPAR patients (75.0%) had preformed DSA with a specificity of HLA-DQ, DR, CW. As preformed DSA ABMR is known to have a better prognosis than de novo DSA ABMR [22], the clinical course of the patients was favorable after antirejection treatment. Cases involving highly sensitized patients warrant close surveillance with proper monitoring through short-term follow-up strategies. This vigilance will facilitate the early detection of allograft rejection, and prompt treatment can significantly enhance allograft outcomes.

The concurrent use of RTX, PP, and BOZ may potentially worsen infections by promoting hypoglobulinemia and B-cell depletion after DSZ [23,24]. As the safety of BOZ remains a subject of concern, appropriate dose modification of BOZ for DSZ is necessary. Based on our BOZ-DSZ protocol, of the 16 transplanted patients, the overall incidence of graft infection complications, including CMV, BKV, and urinary tract infection, was 75% during the follow-up duration, and it was higher in LDKT than DDKT. This may be because LDKT was done immediately after BOZ-DSZ, in contrast to DDKT, which was performed after some time had elapsed from BOZ-DSZ. However, infections were effectively managed with appropriate treatment among total KT recipients, except for one patient who expired due to pneumonia sepsis, leading to a mortality rate of 6.3% (1 of 16). Further, as PP is known to be associated with an increased risk of bleeding [25], our study observed no instances of bleeding requiring transfusion after KT in patients who underwent BOZ-DSZ. While life-threatening infections associated with BOZ-DSZ appear to be relatively low, it is crucial to emphasize vigilant short-term monitoring and infection prophylaxis treatment in highly sensitized patients after BOZ-DSZ to prevent infection events.

To investigate the efficacy and clinical outcomes of BOZ-DSZ compared to conventional DSZ in LDKT, 10 patients with positive crossmatch tests in LDKT were categorized into a BOZ-subgroup. As a control group, 19 highly sensitized patients with positive crossmatch tests who underwent conventional DSZ-based RTX/PP at The Catholic University of Korea, Seoul St. Mary’s Hospital from 2010 to 2019, were compared. After DSZ, both groups showed significant reductions in HLA-DSA MFI levels to below the target level (p = 0.67). In terms of BPAR within 1 year, in the BOZ-subgroup, the BPAR rate was 44.4% (four out of nine patients), including three cases of ABMR and one case of mixed rejection. The conventional group experienced a BPAR rate of 52.6% (10 out of 19 patients), with nine cases of ABMR and one case of TCMR (Supplementary Fig. 1, available online). However, hyperacute rejection leading to graft nephrectomy within 1 month of KT occurred in 15.8% of the conventional group but not in the BOZ-subgroup. Since RTX does not suppress plasma cells and memory B cells [8], there is a possibility of insufficient efficacy, but we cannot exclude the possibility that this could be influenced by factors such as the low frequency of PP sessions [26], dominant use of basiliximab as induction, and the likelihood of retransplantation may have influenced these outcomes (Supplementary Table 1, available online). Larger prospective studies are needed to substantiate these findings and refine DSZ practices for highly sensitized patients.

Our study has some limitations, such as its small sample size obtained from a single center, and retrospective study which presented challenges in establishing statistical significance. Second, we did not employ the C1q binding assay for all participants in this study to assess DSA-HLA antibodies capable of complement activation, mainly due to its recent introduction [27]. Further studies incorporating the precise immunologic risk stratification of pretransplantation including C1q binding assay will be beneficial for the DSZ of highly sensitized patients in the future.

In conclusion, we can suggest that BOZ-DSZ is not only effective but also safe for both highly sensitized LDKT and DDKT recipients. In LDKT, highly sensitized patients, or those who are refractory to previous RTX-based DSZ, may be actively considered before transplantation. Similarly, in DDKT, it can be considered for patients who are anticipated to undergo transplantation imminently; however, there is a need to develop treatment methods that offer sustained efficacy beyond that offered by current therapies. In this regard, our study is expected to serve as a foundational cornerstone that can contribute to the improvement of posttransplant outcomes for highly sensitized patients in the future.

Supplementary Materials

Article information Continued