A Comparison of Plasmapheresis Versus High-Dose IVIG Desensitization in Renal Allograft Recipients with High Levels of Donor Specific Alloantibody


*Corresponding author: Mark D. Stegall, stegall.mark@mayo.edu


Several protocols allow for the successful transplantation of sensitized renal allograft recipients, yet no one best method has emerged. The aim of the current study was to compare the efficacy of high-dose IVIG with two different plasmapheresis (PP)-based regimens in kidney transplant recipients with high levels of donor specific alloantibody (DSA) defined as a positive T-cell cytotoxicity crossmatch. With the primary goal of achieving a negative crossmatch, we employed three protocols sequentially between April 2000 and May 2005: (i) PP, low-dose IVIG, anti-CD20 antibody (n = 32); (ii) high-dose IVIG (n = 13); and (iii) PP, low-dose IVIG, anti-CD20 antibody and pre-transplant Thymoglobulin combined with post-transplant DSA monitoring (n = 16). IVIG decreased DSA activity in all treated patient, yet only 38% (5/13) achieved a negative crossmatch. In contrast, a negative crossmatch was achieved in 84% in PP group and 88% in the PP/monitoring group (p < 0.01 vs. IVIG). Even with a negative crossmatch, the rejection rates were 80% (IVIG), 37% (PP) and 29% (PP/monitoring), respectively, (p < 0.05 IVIG vs. PP). We conclude that multiple PP treatments leads to more reproducible desensitization and lower humoral rejection rates than a single high-dose of IVIG, but that no regimen was completely effective in preventing humoral rejection.


The transplantation of renal allografts into patients with high levels donor specific alloantibody (DSA) defined by a positive T-cell antiglobulin enhanced complement dependent cytotoxicity crossmatch (T-AHG XM) has become increasingly successful over the past several years with reports of greater than 80% graft survival at 1 year (1–5). In order to prevent antibody-mediated graft damage, desensitization therapy must decrease DSA to safe levels at the time of transplant and to maintain these safe levels after re-exposure to alloantigen.

The two major methods of desensitization have been high-dose intravenous immunoglobulin (IVIG) and plasmapheresis (PP)/low-dose IVIG. One protocol incorporates high-dose IVIG to desensitize patients pre-screened for responsiveness using an in vitro assay (6). Other groups, including our own, have employed desensitization using PP followed by low-dose IVIG as the major mode of antibody reduction. While each protocol has allowed successful transplantation, comparisons have been difficult because of significant differences in the patients treated, the assays used to define DSA levels and the outcomes studied. To date, there have been no direct comparisons of IVIG to PP-based protocols.

The aim of the current study was to compare the efficacy of a single high dose of IVIG to two PP/low-dose IVIG-based regimens in achieving a negative T-AHG XM in kidney transplant recipients with high levels of DSA. A secondary endpoint was the incidence of humoral rejection in patients transplanted after achieving a negative crossmatch with these regimens.



These studies were carried out with informed consent using protocols approved by the Institutional Review Board of the Mayo Foundation and Clinic. During the study period between April 2000 and May 2005, we evaluated 103 patients who at baseline had a positive crossmatch against their potential living donor. Patients who had a positive crossmatch only by either T- or B-cell flow cytometric techniques but a negative cytotoxic crossmatches were transplanted without specific pre-transplant desensitization and were excluded from this analysis (n = 39). Similarly, patients whose primary DSA was directed against Class II (positive B-cell crossmatch combined with a negative T-cell crossmatch) also were excluded from this analysis (n = 3). After these exclusions, 61 patients were identified who had a positive T-AHG XM against their living donor and were included in this analysis.

Detection of donor specific antibody (DSA)

Isolation of donor cells:  The techniques used to identify DSA have been reported in detail (7). In brief, donor cells were collected in ACB solution, and T and B lymphocytes were isolated for cytotoxicity crossmatching using immunomagnetic beads and Ficoll Hypaque density gradient technique for flow cytometric testing.

T-cell antiglobulin enhanced complement dependent cytotoxicity crossmatch:  T-cell antiglobulin enhanced complement dependent cytotoxicity crossmatch (T-AHG-CDC) crossmatch was performed using serial doubling dilutions of recipient serum undiluted through 1:512. Prior to performing the crossmatch, the patient's serum was aliquotted into centrifuge tubes and placed in a 62–64°C heat block for exactly 13 min for the removal of IgM antibodies. The last reaction resulting in greater than 20% cell death above background was considered to be the anti-donor T-AHG XM titer. The contribution of IgM was ruled out using heat-treated serum. We do not routinely perform an autocrossmatch. In all cases in this study, the crossmatch was found to be positive due to IgG and not due to IgM.

Flow cytometric crossmatch:  For flow cytometric crossmatching (FXM) a three-color crossmatch was performed. Alloantibody binding was assessed using fluorescein isothiocyanate (FITC) conjugated F(ab)'2 goat anti-human IgG serum by way of indirect immunofluorescence and two other fluorescence parameters (CD3, PerCP and CD19) were used to identify T and B cells. The interpretation of the FXM was performed by directly comparing the fluorescence intensity of donor T and B lymphocytes after treatment with patient serum to the intensity of donor cells after treatment with a negative control serum. Pronase pre-treatment was used to treat the lymphocytes used in the flow cytometric crosmsatch to prevent a false positive result due to the presence of anti-CD20 antibody as previously described (7). Donor lymphocytes were treated by re-suspending the cells with 1.0 mL of pronase (1 mg/mL). The cell preparation was incubated at 37°C for 30 min. Following the pronase treatment, it is essential that the lymphocyte preparation is washed three times using a PBS/NaN3 working solution. A mean channel fluorescence shift greater than 52 was interpreted as positive for T cells and greater than 106 for B cells.

Single-antigen “flowbead” analysis of alloantibody:  Flow PRA I and II and Class I and II single antigen beads (One Lambda, Inc.,< Canoga Park, CA) as previously described (8). In each case in this study, the recipient with a positive crossmatch against their donor at baseline was also found to have donor-specific alloantibody by flow beads or too many specificities to determine.

Impact of high-dose IVIG on crossmatch assays:  We have shown that high-dose IVIG does not interfere with the T-cell AHG crossmatch using three different approaches. First, we added IVIG directly (1 μL volume) to the PRA “FLAP” T-cell AHG panel. In three different assays, the PRA was negative demonstrating that IVIG did not bind to any of the panel of antigens. Similarly, we “spiked” 2 patients with a known PRA of zero with 1:10 dilution of IVIG. Again, the T-cell AHG PRA remained negative despite this lower level of antibody. Finally, we tested the T-cell PRA assay prior to IVIG treatment and the day following IVIG infusion in three sensitized patients. In 2 patients, the PRA remained the same and in 1 patient the PRA decreased by 5%.

Desensitization protocols

The three general approaches to desensitization that were used sequentially during the study period are summarized in Table 1. The primary endpoint of each desensitization protocol was a negative T-AHG XM at the time of transplantation. The secondary endpoint was the diagnosis of humoral rejection in the first 6 weeks after transplantation.

Table 1.  Desensitization protocols in sensitized renal allograft patients
ProtocolN = 61Pre-conditioning therapy
  1. *2 patients resistant to desensitization protocol, not transplanted.

PP/IVIG/anti-CD20 from April 2000 to July 200332PP/IVIG (100 mg/kg) daily Rituximab 375 mg/m2 Splenectomy 19/32 patients
High-dose IVIG from August 2003 to July 200413IVIG 2 gm/kg × 1 prior to Tx (3 gm/kg in 2 patients)
PP/IVIG/anti-CD20/monitoring form August 2004 to May 200516 (2*)PP/IVIG (100 mg/kg) daily Rituximab 375 mg/m2× 1 Intensive post-transplant DSA monitoring

Plasmapheresis/low-dose IVIG + anti-CD20 (PP-IVIG/anti-CD20):  Using this protocol, 32 patients with a positive T-AHG XM received a single pre-transplant dose of rituximab (375 mg/m2; Rituxan Genentech, Inc., San Francisco, CA) 4–7 days prior to transplantation. Patients subsequently underwent daily 1-plasma volume exchange using 5% albumin followed by administration of IVIG (100 mg/kg) as has been previously reported (9). The minimum number of pre-transplant PP in this protocol was 4 and 28 patients required additional PP to achieve a negative crossmatch. A negative crossmatch was documented on the morning of transplant after the last PP. In addition, patients underwent 2–3 post-transplant PP/IVIG on post-operative day 1–3. A protocol allograft biopsy was performed on post-transplant days 4, 7 and 14. Additional clinically-indicated biopsies were performed in the setting of renal allograft dysfunction.

As part of this protocol, the first 19 patients underwent splenectomy at the time of transplantation and the subsequent 13 did not undergo splenectomy. This protocol was used from April 2000 to July 2003.

High-dose intravenous immunoglobulin (HD IVIG):  In this protocol, 11 of 13 patients received a total of 2.1 g/kg and 2 of 13 received 3 g/kg IVIG administered over 1–3 days immediately prior to transplantation. IVIG was obtained from two different vendors during this study (Gamimune N 10%, Bayer Biological, Elkhart, IN; Venoglobulin-S 10%, Alpha Therapeutic Corporation, Los Angeles, CA). After administration, a T-AHG XM was performed within 24 h, and if negative, the living donor kidney transplant was performed. If positive following IVIG treatment, the T-AHG XM was repeated the following day. If persistently positive, the patient was considered to be a “nonresponder” and was treated using the PP-IVIG/anti-CD20 desensitization protocol described above. Protocol and clinically-indicated allograft biopsies were performed as described above. This protocol was used from August 2003 to July 2004.

Plamapheresis /IVIG + anti-CD20 + pre-transplant Thymoglobulin + intensive post-transplant DSA monitoring (PP/monitoring):  This protocol was similar to the PP/IVIG/anti-CD20 protocol described above except for two additional aspects: (i) pre-transplant rabbit anti-human T-cell polyclonal antibody (Thymoglobulin, Sangstadt, Menlo Park CA, 1.5 mg/kg/day) was given in an attempt to aid in desensitization, and (ii) following transplantation crossmatching (T-AHG XM as well as T and B cell flow cytometric XM) was performed and PP-IVIG administered as needed in an attempt to maintain a low DSA level and prevent humoral rejection. This protocol was used from August 2004 to July 2005.

Pre-transplant conditioning

In this protocol, 16 patients received pre-transplant rituximab (375 m2) 1 day prior to the initiation of PP-IVIG and Thymoglobulin was administered after PP on the first 5 days of the pre-conditioning regimen. The number of PP treatments was estimated based on the pre-transplant T-AHG XM titer expecting the level to decrease by one dilution with each PP, and negativity was documented prior to transplantation. Using this protocol, 14 of 16 patients achieved a negative T-AHG XM and underwent transplantation. Two of 16 failed to achieve a negative crossmatch, and were not transplanted.

Post-transplant DSA monitoring and treatment

All patients received PP/IVIG treatments daily on post-operative days 1–3. Protocol allograft biopsy, T and B FXM, and T-AHG XM were performed on post-operative day 4, and were repeated on days 7 and 14 and at times of allograft dysfunction. After post-operative day 3, if the T-AHG XM was negative and both T- and B-cell FXM channel shifts were less than 300, PP/IVIG was discontinued. Conversely, if at any time point during the first 14 days post-transplant, the T-AHG XM was positive or the T or B FXM channel shifts exceeded 300, PP was reinstituted to reduce DSA levels. The channel shift of 300 was chosen because it generally correlates with the cutoff of a positive T-cell AHG crossmatch. In this study, we excluded patients who had primarily or solely anti-donor Class II as determined by channel shift and Luminex beads. At no time point in the study period, did any patient who had predominantly anti-Class I antibody develop predominantly anti-Class II antibody.

Patient demographics

There were no significant differences in the recipient age, recipient gender or donor type in the three groups. The 32 patients in the PP/IVIG/anti-CD20 group had a mean age of 46.1 + 16.3 years (range 21–64), included 20 females and 12 males and comprised 7 living-unrelated renal donors and 25 living-related donors. The 13 patients in the HD IVIG group had a mean age of 44.7 + 13.1 years (range 34–62), included 7 females and 6 males and comprised 4 living-unrelated renal donors and 9 living-related donors. The 16 patients in the PP/monitoring group had a mean age of 45.2 + 12.8 years (range 22–63), included 12 females and 4 males and comprised 3 living-unrelated donors and 13 living-related donors. All patients were Caucasian except one African American in the high-dose IVIG group.

Induction and maintenance immunosuppression

All patients received induction with Thymoglobulin (1.5 mg/kg/day × 5–7 doses) and maintenance immunosuppression with tacrolimus (Prograf, Fujisawa, Deerfield, IL), mycophenolate mofetil (Cell Cept, Roche, Nutley, NJ) and prednisone (tapered to 5 mg by 3 months) as previously reported (2).

Diagnosis and treatment of rejection

All humoral rejection episodes were diagnosed by biopsy. The histologic criteria for humoral rejection included both light and immunofluorescence findings as delineated by the Banff 97 classification of allograft rejection (10). C4d immunostaining in the absence of light microscopic abnormalities was not considered rejection. However, the presence of peritubular capillary neutrophils, glomerular mesangiolysis and fibrin deposition either alone or in combination with C4d deposition was considered rejection. All rejections were treated with high-dose methylprednisolone and the reinstitution of PP/IVIG. In some instances of severe rejection accompanied by marked allograft dysfunction, high-dose IVIG (2 g/kg) also was administered.


Differences in the rates of desensitization, humoral rejection and the number of pre- and post-transplant PP/IVIG sessions in the PP/IVIG/anti-CD20 and PP/IVIG/monitoring groups were compared using Student's t-test. For some comparisons, IVIG was compared to the combined PP protocols. Actuarial patient and graft survival was determined using the Kaplan-Meier test.


Response to desensitization protocols

Table 2 shows the baseline AHG titers in patients treated with the various protocols and their response to desensitization. The overall success of desensitization was 36% (5/13) in the high-dose IVIG group, 84% in the PP/IVIG/anti-CD 20 group (27/32) and 88% (14/16) in the PP/IVIG/monitoring group (p < 0.05 IVIG vs. PP). The success of desensitization correlated with baseline antibody level in each groups. All patients with a baseline crossmatch titer <1:4 achieved a negative crossmatch with all three protocols. Conversely, only 1 of 10 patients with a baseline titer >1:32 achieved a negative crossmatch regardless of the pre-conditioning method utilized. In patients with baseline titers of 1:8 and 1:16, the desensitization rate in the IVIG group was 33% (2/6). When the two PP groups were combined, the success of desensitization with baseline titers of 1:8 and 1:16 was 87% (13/15; p < 0.05 compared to IVIG). Of note, all patients with a persistently positive T-AHG XM also had a persistently positive T-cell flow crossmatch when tested. We previously have shown that the majority of patients who achieved a negative T-cell AHG crossmatch with PP remained positive by T-cell flow cyometric crossmatch after desensitization (8). The outcomes of desensitized patients who achieved a negative T-cell flow crossmatch immediately after transplant was not different from that of patients who had persistent, low levels of donor-specific alloantibody.

Table 2.  Success of desensitization with various protocols stratified by baseline crossmatch titer
AHG Xmatch titer UndiluteIVIGPPPP/monitoring X
  1. X = achieved a negative crossmatch.

  2. O = crossmatch remained positive despite desensitization.

  3. *Nonresponsive to desensitization protocol, not transplanted.

1:32OO X
1:256O O*O*

Table 3 shows that all but 1 patient treated with IVIG had a decrease in their AHG titer. However, patients who achieved a negative crossmatch with high-dose IVIG tended to have lower AHG titers at baseline than nonresponders. Two patients (baseline T-AHG XM titers 1:16) who failed to achieve a negative crossmatch after 2 g/kg IVIG received a second dose of 1 g/kg. One patient converted to a negative crossmatch, while 1 patient remained unresponsive. The 8 patients unresponsive to high-dose IVIG subsequently underwent PP/IVIG/anti-CD20 antibody conditioning and 3 of the 8 patients (those with baseline titers 1:8, 1:16 and 1:256) achieved a negative crossmatch.

Table 3.  Change in T AHG XM titers pre- and post-IVIG treatment
 Pre-IVIG titerPost-IVIG titer
  1. *Received 3 gm/kg total IVIG.


Humoral rejection

While the primary endpoint of this study was a negative T-AHG XM prior to transplant, another measure of the efficacy of the desensitization protocols is their impact on post-transplant antibody-mediated damage. Table 4 shows the humoral rejection rates in patients who achieved a negative crossmatch and subsequently underwent transplantation. The humoral rejection rate was 80% (4/5) in the high-dose IVIG responder group, 37% (11/30) in the PP/IVIG/anti-CD20 group (including 3 high-dose IVIG nonresponders) and 29% (4/14) in the PP/monitoring group (p < 0.05 IVIG vs. PP groups). The time to the diagnosis of rejection was 2.5 days, 4.7 days and 6.4 days, respectively (P = NS IVIG vs. PP). No humoral rejections occurred more than 6 weeks post-transplantation. Importantly, acute cellular rejection has been diagnosed in only 2 patients in our entire series.

Table 4.  Humoral rejection rates
Negative crossmatch at transplant NHumoral rejection
 High-dose IVIG responders54/5 (80%)
PP/IVIG/anti-CD20 (includes 3 IVIG nonresponders)3011/30 (37%)
PP/IVIG/monitoring144/14 (29%)
Positive crossmatch at transplantHigh-dose IVIG, PP/IVIG/anti-CD20 nonresponders107/10 (70%) 5/10 allograft loss (50%)

Transplantation with a positive crossmatch

Not all patients achieved a negative crossmatch in response to the pre-conditioning regimen. Early in our experience, we transplanted 10 patients who were able to achieve low titers, (undilute – 1:8) but not a completely negative crossmatch. Table 4 shows that after transplantation, 70% of these patients developed humoral rejection and allograft loss occurred in 50% (5/10) with two of these due to hyperacute rejection (both were in the PP/IVIV/Anti-CD20 group).

Comparison of post-transplantation plasmapheresis/IVIG treatments in PP/IVIG/anti-CD20 versus PP/IVIG/monitoring groups

All patients in the PP/IVIG/anti-CD20 and PP/IVIG/monitoring groups who achieved a negative T-AHG XM and underwent transplantation underwent post-transplant PP/IVIG. Patients treated with PP/IVIG/anti-CD20 underwent PP/IVIG on post-operative days 1 and 3, and at the time of diagnosis of humoral rejection. Individuals in the PP/IVIG/monitoring group received PP/IVIG on post-operative days 1 and 3, and additional treatments upon detection of an increase in DSA (detected by T-AHG XM positivity or by an increase in the FXM channel shift). Table 5 shows that the use of post-transplant DSA monitoring significantly increased the number of post-transplant PP treatments (PP treatments = in PP/IVIG/anti-CD20 group vs. PP/IVIG/monitoring group; p < 0.01)

Table 5.  Pre- and post-transplant plasmapheresis/IVIG treatments
 Pre-transplant PP/IVIGPost-transplant PP/IVIG
  1. p < 0.001 for both pre- and post-transplant values.

PP/IVIG/anti-CD204.0 ± 1.63.5 ± 3.1
PP/IVIG/monitoring9.8 ± 3.29.4 ± 7.7

Graft and patient survival

Figure 1 shows patient and graft survival for all patients transplanted. The overall 1-year actuarial patient survival was 93% and the 1-year actuarial graft survival was 82% (follow-up 127–1915 days). At 6 months, the serum creatinine in the patients who achieved desensitization, underwent transplant and still had a functioning graft was similar in the three groups including: 1.09 ± 0.71 mg/dL in the PP/IVIG/anti-CD20 group; 1.75 ± 0.65 mg/dL in the high-dose IVIG group and 1.64 ± 0.38 mg/dL in the PP/IVIG/monitoring group (P = NS for all comparisons).

Figure 1.

Patient and graft survival of positive crossmatch recipients—all groups combined.


The current study demonstrates that both a single high dose of IVIG and PP-based regimens can achieve desensitization and successful transplantation in patients with high levels of DSA. However, both are complicated by the failure to successfully desensitize all patients and a high rate of humoral rejection after transplantation. These data also suggest that the greatest predictor of a successful desensitization is the baseline crossmatch titer. Specifically, patients with positive T-AHG XM against their donors can be desensitized with either IVIG or PP, but patients with high titers do not respond well to either regimen.

This study is the first to allow for a comparison of high-dose IVIG to PP-based protocols for desensitization. While the number of patients in each group is small and the subjects were not randomized, we believe that several important points can be gleaned regarding the comparative efficacy of high-dose IVIG versus PP/low-dose IVIG as used in this study. First, it appears that PP/low-dose IVIG, albeit using multiple treatments, is more likely to achieve desensitization in patients with crossmatch titers in the 1:8 to 1:16 range when compared to a single 2 gm/kg dose of IVIG. Second, the finding that at least some patients who failed IVIG were able to achieve a negative crossmatch with PP also suggests that in some patients multiple PP treatment might be effective in IVIG “nonresponders”. Finally, the humoral rejection rate in patients desensitized with a single dose of high-dose IVIG appears higher than those desensitized with multiple PP treatments. However, it is important to note that the current study was designed primarily to compare IVIG to PP-based protocols and did not aim to “optimize” IVIG desensitization. Thus, it is possible that multiple doses of IVG including the post-transplant period might show similar rates of desensitization and decrease post-transplant humoral rejection. The use of a screening in vitro assay to identify potential IVIG responders and accepting higher DSA levels at the time of transplant might have allowed us to achieve desensitization rates similar to those previously reported.

This is also the first study to measure the DSA response to IVIG using crossmatch titers as an endpoint. These data show that almost all patients experience a decrease in crossmatch titers during IVIG treatment and that the “response” is usually 1–3 dilutions. Thus, it is possible that the “nonresponders” in previous studies of IVIG were patients with very high levels of DSA. These data further suggest that the response to IVIG is actually quite reproducible, i.e. almost all patients with low crossmatch titers develop a negative crossmatch and those with high titers do not. The response to multiple high doses of IVIG was only tested in 2 patients and although only one responded, this might be an area of further study.

From these and other studies, we suggest that none of the current desensitization regimens appears to have profound effects on DSA production (8,11). While a small minority of patients may demonstrate significant and long-term decreases in DSA levels, most patients have persistent DSA at least at low levels (8). Most “successful” transplants occur in patients who begin with relatively modest crossmatch titers. The high rate of humoral rejection and graft loss in the patients who were transplanted with a persistently positive T-AHG XM suggests that any impact on DSA production in these patients is insufficient to achieve a successful transplant. Our current practice is to attempt desensitization only in patients who have T-AHG XM titers <1:16. Fortunately, this includes the majority of sensitized patients. One of the most important lessons learned from this clinical experience is that the selection of patients with lower levels of DSA activity is the key to improved graft survival. The use of intensive post-transplant monitoring of DSA levels to maintain these at safe levels was not successful in higher titer patients, but may deserve further study in low-titer recipients. Overall, this approach did appear to delay the onset of rejection but it also increased the number of PP treatments. In the absence of a therapy that effectively controls DSA production, the ability to monitor DSA levels may have little utility. Although not addressed in this study, it is likely that the renal allograft does “accommodate” to DSA as has been reported previously (12), but that this process requires a period of protection from high levels of DSA to allow for long-term graft survival.

These data would agree with studies of desensitization without transplantation. IVIG has been tested in a large prospective, randomized trial (13). Its use led to a transient decrease in Panel Reactive Antibody, but these levels increased to baseline after cessation of treatment. The transient reduction in sensitization with IVIG allowed for more crossmatch-negative transplants to occur, yet the acute rejection rate was higher in the transplanted group compared to the placebo group. This result, similar to the results of the current study, suggests that IVIG only transiently affects DSA levels. Anti-CD20 was tested in 9 sensitized dialysis patients (14). Only a small percentage of patients showed any effect. These studies emphasize the fact that no therapy has been shown to have a significant effect on DSA production.

Our understanding of desensitization is hampered the lack of definitive studies in several areas. While the current study seeks to improve our understanding, it still has many limitations. First, like many other previous studies, the number of patients in each group is relatively small. Second, even though we confined our study to patients with a positive T-cell AHG crossmatch, the levels of DSA varied markedly although they were generally similar in each group. The role of Class II antibody is still quite controversial and our ability to differentiate between Class I and Class II using the B-cell crossmatch is limited. The current study did not include patients who at baseline had a negative T-cell AHG crossmatch, but a positive T-cell flow cytometric crossmatch. However, we believe that these patients also may be at risk for antibody-mediated allograft damage. Finally, the current study is not randomized, but rather three sequential clinical series with all of the inherent biases in such a retrospective study.

In summary, we believe that the current study demonstrates that when compared to a single high dose of IVIG, protocols employing multiple PP treatments lead to more reproducible desensitization and lower humoral rejection rates. Post-transplant monitoring of DSA with intensive PP appears to delay the onset, but may not reduce the incidence of humoral rejection. No current therapy appears to be able to significantly reduce antibody production or to reduce DSA to safe levels in patients with high titers. We conclude that a major focus of future research in this area should emphasize methods to decrease DSA production and/or to ameliorate its effect on the graft.