Immunoadsorption as a Rescue Therapy for Very High Anti-B Titer in ABO-Incompatible Kidney Transplantation

Abstract

Patient: Female, 51-year-old

Final Diagnosis: ABO-incompatible kidney transplantation with antibody titer 1:2048

Symptoms: ESKD

Clinical Procedure: Immunoadsorption therapy • plasma exchange

Specialty: Nephrology • Transplantology

Objective: Unusual clinical course

Background

ABO-incompatible (ABOi) kidney transplantation has been performed for several years, with the understanding that elevated antibody titers against the donor blood group are associated with a heightened risk of antibody-mediated rejection. The primary strategy for managing ABOi transplantation involves desensitization through antibody removal and the administration of rituximab. Effective methods for antibody removal include plasmapheresis, double filtration plasmapheresis, and immunoadsorption, using specialized absorbent columns.

Case Report

The patient was a 51-year-old woman with end-stage kidney disease who received an ABO-incompatible (B to O) kidney allograft from her husband. Her initial anti-B antibody titer was 1: 2048 (by gel column-agglutination). During the pre-conditioning phase, sessions of double filtration plasmapheresis (DFPP) combined with rituximab failed to reduce the anti-B antibody titer below 1: 512, and a bleeding complication occurred as a result of DFPP. Subsequently, a rescue approach involving specific anti-B antibody immunoadsorption was implemented, achieving an anti-B titer of 1: 32 on the day of the operation. The living donor kidney transplantation was performed successfully without significant complications.

Conclusions

We present the first case using immunoadsorption as a rescue therapy for DFPP-resistant anti-B titers prior to ABO-incompatible kidney transplantation. Among the available antibody removal protocols, immunoadsorption has demonstrated favorable outcomes. This technique can be performed over extended durations to accommodate larger plasma volumes while minimizing the risk of bleeding complications, making it an effective rescue strategy for cases that are resistant to traditional apheresis methods.

Introduction

Kidney transplantation is currently the most effective renal replacement therapy, offering superior solute clearance, volume control, and reduced mortality for patients with end-stage kidney disease [1,2]. Living donor kidney transplantation, in particular, yields better allograft and patient outcomes than does deceased donor transplantation [3]. However, because living donors are limited and not every patient with end-stage kidney disease has an available donor, deceased donor transplantation remains crucial to supplement the donor pool. One strategy to expand living donation is ABO-incompatible (ABOi) kidney transplantation, which has been performed for over 30 years and has demonstrated successful graft accommodation and favorable outcomes that are superior to those of deceased donor kidney transplantation [4].

A critical aspect of ABOi kidney transplantation is the removal of anti-A and/or anti-B antibodies from the recipient’s plasma, depending on the donor’s and recipient’s blood groups. Established protocols for antibody removal include plasma exchange, double filtration plasmapheresis (DFPP), and immunoadsorption [5]. In this report, we describe an exceptional case of ABOi kidney transplantation in a recipient with a very high initial antibody titer. The management of this case required the use of multiple apheresis techniques, including immunoadsorption rescue, to reduce the antibody titer to levels that permitted safe transplantation.

Case Report

The patient was a 51-year-old woman with blood group O, Rh-positive, and diagnosis of end-stage kidney disease of unknown etiology. Because she had anuria, she had been maintained on hemodialysis via a left brachiocephalic arteriovenous fistula (AVF) for 3 years, without complications. Her husband, with blood group B, Rh-positive, was identified as a potential kidney donor. The human leukocyte antigen (HLA) mismatch was 1-2-0 for HLA-A, -B, and -DR, respectively. Screening for anti-HLA antibodies using Luminex screening beads was negative, and the complement-dependent cytotoxicity assay with anti-human globulin crossmatching was negative.

During the transplantation evaluation, the recipient’s initial serum anti-B IgG antibody titer was 1: 2048. According to current literature and our center’s pre-conditioning protocol, the starting anti-A/B IgG antibody titer should not exceed 1: 512 to 1: 1024, due to the risk of unsuccessful antibody removal and rebound. Despite this elevated titer, the couple opted to proceed with transplantation rather than await a deceased donor. It is worth noting that our center uses an automated gel column-agglutination method for titer measurement. As a result, titers obtained can differ from those measured by other techniques, such as the conventional manual tube method or the solid-phase automated microplate assay [6,7]. Accordingly, the titer thresholds reported here should be interpreted with caution, since assay methodology varies across transplant centers and may require center-specific cutoffs [8,9].

After a thorough discussion of the risks of antibody-mediated rejection, potential procedure-related complications (such as bleeding or infection associated with plasmapheresis), and the associated costs, a desensitization protocol was formulated. This protocol included a rituximab infusion and 7 sessions of DFPP, with the goal of reducing the anti-B IgG titer to ≤1: 32 by the scheduled transplant day, which had already been fixed in the operative timetable. Due to our center’s tight surgical schedule, patients who fail to meet this threshold will have their transplant procedures cancelled and rescheduled in the following months, thereby forfeiting the benefits of the desensitization process without proceeding to transplantation.

The patient was admitted on day −15 prior to the operation for rituximab administration (Truxima, Celltrion, Incheon, Korea; 600 mg, equivalent to 375 mg/m²) and initiation of oral immunosuppressive therapy, including tacrolimus (target pre-dose [C₀] concentration of 7–12 ng/mL), mycophenolic mofetil (1500 mg daily), and prednisolone (20 mg daily). DFPP sessions were conducted on alternate days starting on day −13 and continued on days −11, −9, and −7 (Figure 1). Each DFPP session processed 4.375 to 5 L of plasma – approximately twice the patient’s total plasma volume.

Figure 1.

Anti-B titer levels in the pre-conditioning period.

During each DFPP session, Plasmaflo was used as the plasma separator and Cascadeflo-EC20W as the plasma component separator. The blood flow rate was maintained at 100 mL/min, with a plasma filtration rate of 30 mL/min and a final drainage rate of 6 mL/min. An 8% albumin solution, equal in volume to the drained plasma, was used as the replacement fluid. Low-dose unfractionated heparin was administered as an anticoagulant, beginning with an initial 1000-unit intravenous bolus, followed by a continuous infusion of 500 units per hour for 2 hours.

Laboratory test results, including anti-B IgM and IgG antibody titers and coagulation parameters (prothrombin time [PT], partial thromboplastin time [PTT], and international normalized ratio), were monitored before and after each DFPP session. Although each session reduced the anti-B antibody titer by 1 to 2 levels, the titers rebounded to 1: 512 by the morning of day −5. Detailed antibody titer levels are presented in Figure 1.

As expected, a bleeding complication occurred due to the large plasma volume treated with DFPP, which removes both the antibodies and coagulation proteins [10]. After the first DFPP session, the patient’s PTT was prolonged to over 200 s, and following the second session, she experienced bleeding from the AVF upon removal of the access needle. After the fourth DFPP session, AVF bleeding occurred that did not respond to compression, necessitating the temporary use of a high-pressure tourniquet.

Given the modest reduction in antibody titer, bleeding complications, and need to lower the antibody titer for transplantation on the fixed operative day, we modified our approach. After discussing options with the donor and recipient, immunoadsorption was reconsidered despite its higher cost, and it was agreed to discontinue using DFPP alone. We adapted the immunoadsorption technique using the Glycosorb-ABO B column [11], which specifically adsorbs antibodies while preserving coagulation proteins.

On day -5, immunoadsorption was performed with a total treated plasma volume of 8.75 L (approximately 3.7 times the patient’s plasma volume), requiring nearly 5 h. Following immunoadsorption, the anti-B IgG titer dropped by 5 log₂, from 1: 512 to 1: 16, and the anti-B IgM titer decreased from 1: 64 to 1: 4, with no prolongation of PT or PTT. However, the anti-B IgG titer rebounded by 2 log₂ to 1: 64 the next day, prompting an additional dose of rituximab (500 mg) on day −4. CD20⁺ B cells were not measured routinely, as the turnaround time at our center is incompatible with the clinical requirements for ABO-incompatible transplantation.

By day −3, the anti-B IgG titer had rebounded to 1: 256; therefore, daily DFPP sessions (6.25 liters of plasma) were prescribed, but they again failed to reduce the titer to 1: 32. Consequently, another immunoadsorption session was performed on day −1, treating 11 L of plasma (approximately 4.3 times the patient’s plasma volume). Immediately after this session, the anti-B IgG titer reached its lowest level, at 1: 8, and the IgM titer was 1: 1. On the morning of the operation, the titers were 1: 32 for IgG and 1: 2 for IgM. Details of the procedures and the pre-conditioning timeline are summarized in Table 1.

Table 1.

Patient’s data on preconditioning before kidney transplantation.

Day	IgM	IgG	Post IgM	Post IgG	Procedure/ medication	Total plasma volume (L)	Estimated PV	Tacrolimus C0 (ng/mL)	Complications
−15					Rituximab 600 mg
−14	256	2048						6.3
−13	256	2048	64	512	DFPP	4.375	2.0
−12	128	512
−11	128	1024	32	512	DFPP	4.375	1.9		AVF compression >30 min, need adrenaline gauze pack
−10	256	512						14.1
−9	64	512	16	128	DFPP	4.375	1.8		AVF compression >30 min
−8	32	256
−7	64	512	16	128	DFPP	5.000	2.1	9.6	AVF bleeding, need tourniquet compression
−6	32	256
−5	64	512	4	16	Glycosorb-ABO B column	8.750	3.7
−4	16	64			Rituximab 500 mg
−3	32	256	8	64	DFPP	6.250	2.5	5.9
−2	16	64	8	32	DFPP	6.250	2.4
−1	16	64	1	8	Glycosorb-ABO B column	11.000	4.3
0	2	32			Kidney transplantation			10.5

AVF – arteriovenous fistula; C₀ – pre-dose concentration; DFPP – double-filtration plasmapheresis; post IgG – post-antibody removal IgG titer; post IgM – post-antibody removal IgM titer.

The induction regimen included basiliximab (20 mg) administered intravenously on days 0 and 4, and methylprednisolone (1 g) given intravenously on days 0 to 2, followed by oral prednisolone at 1 mg/kg/day. The patient also received a standard maintenance immunosuppressive regimen comprising tacrolimus, mycophenolate mofetil, and oral prednisolone.

Kidney transplantation was successfully performed without complications. The cold ischemic time was 31 min, and the warm ischemic time was 46 min, resulting in a total ischemic time of 1 h and 17 min. The allograft demonstrated immediate function upon reperfusion, with urine flow observed intraoperatively. Urine output initially reached 8 L per day and gradually decreased to approximately 1.5 to 2 L per day. Serum creatinine levels declined steadily from 4.15 mg/dL preoperatively to 0.6 to 0.7 mg/dL within 2 days after surgery. Tacrolimus trough levels (C₀) were maintained at 9 to 10 ng/mL during the first month after transplantation. The patient’s post-transplantation data are presented in Table 2.

Table 2.

Patient’s data after living donor kidney transplantation.

	IgM	IgG	Urine output (mL/day)	Drain volume (mL/day)	Creatinine (mg/dL)	Tacrolimus C0 (ng/mL)	Procedure
Day
0	2	32	8750	315	4.15	10.5	Kidney transplantation
1	2	16	7740	155	1.41
2	4	16	5350	445	0.7	12.7
3	4	32	2720	290	0.67
4	4	32	2050	610	0.72	7.8
5	4	32	2400	415	0.69
6	4	32	3450	470	0.71
7	4	32	2900	410	0.64	7.0	Off Foley catheter
8	4	32	2700	520	0.64
9	4	32	2300	430	0.60
10	8	32	2250	345	0.60
11	8	32	2800	340	0.59	9.2
12	8	32	2600	236	0.61
13	8	32	2650	239	0.61
14	8	32			0.61	8.6	Discharged from hospital
Week
3	16	16	2105	150	0.62	11.6
4	8	32	1640	90	0.76	9.1
5	8	16	1785	50	0.65	9.0	Flexible cystoscope with double-J stent removal
6	16	32	1900	-	0.72	10.7	Off drain
8	8	32	2525	-	0.77	8.3
10	8	16	1775	-	0.77	6.3

C₀ – pre-dose concentration.

The 1-month protocol surveillance allograft biopsy was unremarkable, with histological scores of g0, ptc0, t0, i0, and v0, according to the Banff criteria [12]. Additionally, infection screening – including BK virus and cytomegalovirus viral load assessments – yielded negative results, as part of a preemptive strategy. The only post-transplant complication was a high drainage volume. The fluid was confirmed to be serous, ruling out urine leakage. Initially, the drainage volume ranged between 150 and 180 mL per day but gradually decreased, allowing for the drain to be successfully removed at 6 weeks after surgery.

Discussion

We presented a case of ABO-incompatible kidney transplantation in a patient with an exceptionally high anti-B antibody titer of 1: 2048, which was well above the established threshold of 1: 1024 [13–15]. High anti-A/B antibody titers are associated with an increased risk of desensitization failure [16]. Notably, our innovative approach incorporated immunoadsorption as an add-on therapy, following unsuccessful desensitization with plasmapheresis. To the best of our knowledge, this is the first report describing the use of immunoadsorption in combination with plasmapheresis to achieve the target antibody titer in a plasmapheresis-resistant ABOi kidney transplant.

A meta-analysis by de Weerd et al found that ABOi kidney transplantation had significantly lower uncensored graft survival at 1 year than did center-matched ABO-compatible living donor transplants (relative risk, 0.97; 95% CI, 0.96–0.98; P<0.001). Moreover, ABOi transplantation is associated with increased rates of infectious complications, bleeding, and antibody-mediated rejection [17]. Nevertheless, several studies have demonstrated that outcomes of ABOi transplantation are superior to those of patients remaining on the waitlist or receiving an ABO-compatible deceased donor transplant, supporting the decision to proceed with ABOi transplantation even in the presence of high anti-B titers [4,18,19].

Higher baseline antibody titers are associated with an increased risk of rejection. For instance, Masterson et al observed that ABO-incompatible kidney transplantation can be successfully performed without antibody removal in patients with low-titer anti-B antibodies (1: 1 to 1: 16, as measured by the automated column agglutination technique), suggesting that the baseline titer can serve as a reliable predictor of antibody-mediated rejection [20]. Moreover, a case report documented successful ABOi kidney transplantation in a patient with an extremely high anti-B antibody titer (>1: 8196) [21]. In that instance, a desensitization protocol – comprising low-dose rituximab, 8 sessions of standard plasmapheresis, and intravenous immunoglobulin – reduced the titer to 1: 32 prior to transplantation, resulting in the absence of clinical antibody-mediated rejection and good allograft function at 1 year. In our patient, whose initial anti-B titer was 1: 2048, transplantation remains a viable option provided that the titer can be reduced to below 1: 32 before the procedure. As noted above, these studies used different titer measurement techniques – including the conventional tube method, automated gel column agglutination (as in our report), and solid-phase automated microplate assays [6,7] – which limits comparability across centers. Currently, no universally adopted external quality-assurance framework exists. Establishing such an external quality-assurance would reduce inter-laboratory variability and enable the adoption of standardized titer thresholds for clinical practice [22]. Additionally, our fixed operative schedule necessitated achieving the target titer within a specific timeframe, in contrast to the previous case report that benefited from a flexible schedule and non-DFPP-resistant antibody reduction [21]. Cancellation of a transplant procedure in our patients would incur substantial financial losses and waste the equipment allocated for the desensitization protocol.

In the past, splenectomy was considered a prerequisite for successful ABOi kidney transplantation [23,24]. However, multiple studies have demonstrated that ABOi kidney transplants can be successfully performed without splenectomy, through the use of rituximab, plasmapheresis, and immunoglobulin therapy [11,25,26]. For instance, Lo et al conducted a systematic review and meta-analysis involving 4,810 ABOi kidney transplant recipients, reporting an overall graft survival rate of 94.5% in patients treated with rituximab, compared with 79.7% in those who underwent splenectomy, with a mean follow-up period of 28 months [14]. Consequently, splenectomy is now considered an outdated practice in this context.

The American Society for Apheresis recommends therapeutic plasma exchange as a first-line intervention for desensitization in living donor ABOi kidney transplantation [27]. However, plasmapheresis is associated with the loss of fibrinogen, von Willebrand factor, and factor XIII during each session, which can increase the risk of bleeding complications [28,29].

Specific immunoadsorption techniques have increasingly been used, demonstrating successful outcomes [30,31]. For example, a systematic review and meta-analysis reported an overall graft survival rate of 94.1% (95% CI, 88.2–97.1%) in recipients who underwent immunoadsorption, compared with 88.0% (95% CI, 82.6–91.8%) in those who received apheresis. Moreover, some reports suggest that immunoadsorption is associated with less depletion of clotting factors than is plasmapheresis, potentially reducing the risk of hemorrhagic events [32,33].

In our center, the cost of an immunoadsorption column is approximately $3000 USD (per session), whereas DFPP costs around $1500 USD. Despite the higher cost, immunoadsorption achieves significantly greater antibody removal per session than does DFPP, likely due to the longer duration of treatment and the larger plasma volume processed, as demonstrated in our case. Moreover, immunoadsorption preserves coagulation factors, thereby avoiding the increased bleeding risk associated with DFPP or standard plasmapheresis when plasma volume is expanded [32,34,35]. As such, immunoadsorption can serve as an effective rescue therapy when additional antibody removal is required.

Conclusions

We present a case of an ABO-incompatible living donor kidney transplant in a patient with an extremely high antibody titer, in whom initial antibody removal with DFPP resulted in significant bleeding complications and failed to achieve the target titer. In light of this inadequate response, we implemented a rescue strategy combining DFPP with immunoadsorption – a widely used technique in ABOi kidney transplantation that specifically targets anti-A/B antibodies and effectively reduces antibody levels to acceptable thresholds. Unlike plasmapheresis, which carries a higher risk of bleeding due to coagulation factor loss, immunoadsorption can be performed over extended periods to treat larger plasma volumes, ensuring more comprehensive antibody clearance while minimizing adverse bleeding events. This multimodal approach successfully reduced antibody levels and ultimately led to a favorable transplant outcome, demonstrating the efficacy of combined desensitization techniques in overcoming challenging cases.