The n e w e ng l a n d j o u r na l
original article
From the Departments of Surgery (R.A.M.,
B.E.L., L.M.K., J.E.L., D.S.W., C.E.S., N.N.D.,
A.L.S., D.L.S.), Pathology (K.E.K.), and
Medicine (E.S.K., A.A.Z.), Johns Hopkins
Medical Institutions; and the Department
of Epidemiology, Johns Hopkins University
School of Public Health (L.M.K., D.L.S.)
both in Baltimore. Address reprint re-
quests to Dr. Montgomery at 720 Rut-
land Ave., Ross Research 765, Baltimore,
MD 21205, or at rmonty@jhmi.edu.
N Engl J Med 2011;365:318-26.
Copyright 2011 Massachusetts Medical Society.
318
Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
of m e dic i n e
Desensitization in HLA-Incompatible Kidney
Recipients and Survival
Robert A. Montgomery, M.D., D.Phil., Bonnie E. Lonze, M.D., Ph.D.,
Karen E. King, M.D., Edward S. Kraus, M.D., Lauren M. Kucirka, Sc.M.,
Jayme E. Locke, M.D., M.P.H., Daniel S. Warren, Ph.D.,
Christopher E. Simpkins, M.D., M.P.H., Nabil N. Dagher, M.D.,
Andrew L. Singer, M.D., Ph.D., Andrea A. Zachary, Ph.D.,
and Dorry L. Segev, M.D., Ph.D.
A bs t r ac t
Background
More than 20,000 candidates for kidney transplantation in the United States are
sensitized to HLA and may have a prolonged wait for a transplant, with a reduced
transplantation rate and an increased rate of death. One solution is to perform live-
donor renal transplantation after the depletion of donor-specific anti-HLA antibod-
ies. Whether such antibody depletion results in a survival benefit as compared with
waiting for an HLA-compatible kidney is unknown.
Methods
We used a protocol that included plasmapheresis and the administration of low-dose
intravenous immune globulin to desensitize 211 HLA-sensitized patients who sub-
sequently underwent renal transplantation (treatment group). We compared rates of
death between the group undergoing desensitization treatment and two carefully
matched control groups of patients on a waiting list for kidney transplantation who
continued to undergo dialysis (dialysis-only group) or who underwent either dialysis
or HLA-compatible transplantation (dialysis-or-transplantation group).
Results
In the treatment group, KaplanMeier estimates of patient survival were 90.6% at
1 year, 85.7% at 3 years, 80.6% at 5 years, and 80.6% at 8 years, as compared with
rates of 91.1%, 67.2%, 51.5%, and 30.5%, respectively, for patients in the dialysis-
only group and rates of 93.1%, 77.0%, 65.6%, and 49.1%, respectively, for patients in
the dialysis-or-transplantation group (P<0.001 for both comparisons).
Conclusions
Live-donor transplantation after desensitization provided a significant survival bene
fit for patients with HLA sensitization, as compared with waiting for a compatible
organ. By 8 years, this survival advantage more than doubled. These data provide
evidence that desensitization protocols may help overcome incompatibility barriers in
live-donor renal transplantation. (Funded by the NationalInstitute of Diabetes and
Digestive and Kidney Diseasesand the Charles T. Bauer Foundation.)
n engl j med 365;4 nejm.org july 28, 2011
The New England Journal of Medicine
Copyright 2011 Massachusetts Medical Society. All rights reserved.
 Desensitization in HLA-Incompatible Kidney Recipients
R
enal-replacement therapy can be
achieved by means of transplantation or
dialysis. Transplantation offers clear ben-
efits in terms of longevity, lifestyle, and savings in
health care costs.1-4 However, organs are scarce,
and the rate of death among patients on the kidney-
transplant waiting list is high. In 2008, of the
82,000 patients on the waiting list in the United
States, 16,520 received a renal transplant, where-
as 4800 died while waiting for one.5 Patients who
have become sensitized to HLA as a result of pre-
vious transplantation, pregnancy, or blood trans-
fusion have an increased likelihood of a positive
cross-match, indicating the presence of donor-
specific anti-HLA antibody. The presence of circu-
lating donor-specific anti-HLA antibody has been
associated with hyperacute rejection, antibody-
mediated rejection, and unacceptably high rates
of organ loss, and the presence of this antibody
has historically been considered a contraindica-
tion to transplantation.6-9 Many patients with such
sensitization have a willing live donor but are un-
able to realize this benefit because of a positive
cross-match.
Among patients undergoing transplantation
across HLA barriers, the use of preconditioning,
either with high-dose intravenous immune globu-
lin or with plasmapheresis plus low-dose intrave-
nous immune globulin, has had promising short-
term outcomes, although rates of short-term and
long-term antibody-mediated rejection have been
high.10-13 Crossing the HLA barrier is a relatively
recent phenomenon, and data on long-term out-
comes are limited.14-18 In recent studies, rates of
graft and patient survival in such cases have been
well below national averages for compatible live-
donor transplantation.19 Although it is clear that
desensitization increases transplantation rates and
reduces the waiting time among patients with HLA
sensitization, the inferior outcomes that have been
reported to date raise the question of whether
these patients would be better served by waiting
for a compatible organ.
Me thods
Study Population
From February 1998 through December 2009, a
total of 211 patients underwent HLA-incompatible
live-donor kidney transplantation, with HLA in-
compatibility defined by three non-overlapping
categories of antibody strength: positive comple-
n engl j med 365;4 nejm.org july 28, 2011
The New England Journal of Medicine
Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
ment-dependent cytotoxic or flow-cytometric cross-
matching or detectable donor-specific anti-HLA
antibody on multiplex bead assay (Luminex). All
patients provided written informed consent to un-
dergo plasmapheresis and the administration of
albumin, fresh-frozen plasma, and intravenous im-
mune globulin before transplantation. The con-
senting procedure included a description of pos-
sible adverse events. The patients were treated with
a desensitization treatment protocol approved by
the institutional review board at the Johns Hop-
kins Medical Institutions. Initially, the protocol
was deemed to be innovative therapy, and a pro-
spective clinical database was maintained. The
board then granted approval to convert the clini-
cal database to a research database, which under-
goes annual review. The study protocol is avail-
able with the full text of this article at NEJM.org.
Desensitization Treatment
Plasmapheresis was performed with the use of a
centrifuge-driven cell separator, as described pre-
viously.20 After each plasmapheresis session, pa-
tients received intravenous cytomegalovirus im-
mune globulin (Cytogam, CSL Behring) at a dose
of 100 mg per kilogram of body weight. Escalat-
ing numbers of treatments were performed be-
fore and after transplantation on the basis of the
level of donor-specific anti-HLA antibody at base-
line.21 Briefly, patients with donor-specific anti-
HLA antibody that was detected only by means of
the bead assay and patients with a positive flow-
cytometric assay received between two and four
treatments, depending on the initial level of the
donor-specific anti-HLA antibody and their re-
sponse to the first two treatments. There was an
increased range in the number of treatments be-
fore transplantation in the group with positive
complement-dependent cytotoxicity cross-match-
es, which reflected the increased range in levels of
cross-match reactivity. The goal of treatment was
the conversion to a negative cytotoxic cross-match
before transplantation. However, in a few cases,
the strength of cross-matching for donor-specific
anti-HLA antibody plateaued at a low level of re-
activity (titer, <8) on the complement-dependent
cytotoxicity assay, and we proceeded with trans-
plantation. We required a sustained reduction in
the level of donor-specific anti-HLA antibody in
each patient before transplantation. Desensitiza-
tion was attempted in 215 patients, and 211 un-
derwent transplantation; the other 4 patients had
319
 The n e w e ng l a n d j o u r na l
an inadequate response to plasmapheresis, which
was defined as a complement-dependent cytotox-
icity cross-match titer of 8 or more or a rebound
in the level of donor-specific anti-HLA antibody.
Immunosuppression
Mycophenolate mofetil (at a dose of 2 g per day)
and tacrolimus (target serum level, 8 to 12 ng per
milliliter) were administered with plasmapheresis
before transplantation. Induction therapy consist-
ed of the use of daclizumab, with an initial dose of
2 mg per kilogram and then 1 mg per kilogram
every 2 weeks for a total of five doses, or antithy-
mocyte globulin (Thymoglobulin, Genzyme) at a
dose of 1.5 mg per kilogram per day for 5 days.
Glucocorticoids, including dexamethasone, were
administered at a dose of 100 mg intraoperatively
and at a dose of 25 mg every 6 hours postopera-
tively for six doses, followed by tapering to 5 to
10 mg daily during a 3-month period.
Cross-Matching and Donor-Specific Anti-HLA
Antibody
Complement-dependent cytotoxicity assays and
three-color flow-cytometric cross-matching were
performed, as described previously.22,23 Cytotox-
icity assays were performed with antihuman glob-
ulin augmented for T cells and one wash for B cells,
with a maximum titer of 512 for a positive result.
Flow-cytometric cross-matching was performed
with the use of the FACSCalibur flow cytometer
(Becton Dickinson). The secondary antibody was an
IgG heavy-chainspecific mouse monoclonal anti-
body (BD Biosciences). Analyses to define antibody
specificities were performed on the Luminex plat-
form with the use of an HLA phenotype panel
(Lifematch Class I and Class II ID, Gen-Probe) and
a single-antigen panel (Single Antigen Beads, One
Lambda). Results of bead assays were expressed
as mean fluorescence intensity.
Matched Control Subjects
To determine whether desensitization provided a
survival benefit, we performed a matched control
analysis with deidentified patients drawn from the
United Network for Organ Sharing (UNOS) kidney-
transplant waiting list. For each patient in the
group that underwent desensitization and trans-
plantation (treatment group), on the date of the re-
ceipt of the incompatible transplant (with frame-
shifting to account for differences in follow-up
between our patients and those in the national
320 n engl j med 365;4 nejm.org july 28, 2011
The New England Journal of Medicine
Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
registry), we identified the five patients on the
UNOS waiting list who were most closely matched.
The percentage of panel-reactive antibodies was
matched as follows: for patients with a panel-
reactive antibody level of 0% or 100%, a control
subject with the same panel-reactive antibody level
was identified; for patients with a level ranging
from 98 to 99% or from 95 to 97%, a control sub-
ject with a level in the same range was identified;
for patients with a level ranging from 85 to 94%,
65 to 84%, or 1 to 64%, a control subject with a
panel-reactive antibody level within 2, 5, or 10
points, respectively, of the antibody level in the
study patient was identified. The remaining fac-
tors were matched with the use of iterative, ex-
panding radius matching, as described previous-
ly,24 in the following order of priority: age, blood
type, number of previous transplantations, pro-
portion of years of renal-replacement therapy with
a functioning allograft, total number of years of
renal replacement (including transplantation or di-
alysis), race, sex, and the presence or absence of
diabetes.
To determine whether the relatively small pos-
sibility of finding an HLA-compatible donor would
affect the comparison with subjects on the waiting
list, we divided the matched control subjects on the
waiting list into two groups: those who continued
to undergo dialysis (dialysis-only group) and those
who either continued to undergo dialysis or under-
went HLA-compatible transplantation at any time
after the date of transplantation in the matched
patient treated with desensitization (dialysis-or-
transplantation group). Matched control subjects
in the dialysis-only group were identified for all
but one HLA-incompatible transplant recipient (a
46-year-old woman who had received a transplant
in 1999, had a panel-reactive antibody level of 91%,
and had undergone renal-replacement therapy for
28 years). Matching subjects in the dialysis-or-
transplantation group were identified for all but
three HLA-incompatible recipients.
Statistical Analysis
We used the KaplanMeier method and the log-
rank test to compare rates of survival in the treat-
ment group with those in the two matched con-
trol groups. Comparisons with the control groups
were performed for both the overall group of HLA-
incompatible recipients and the three subgroups
identified on the basis of the level of donor-specif-
ic anti-HLA antibody. When survival curves for two
 Desensitization in HLA-Incompatible Kidney Recipients

Table 1. Baseline Characteristics of HLA-Incompatible Kidney-Transplant Recipients, Stratified According to Strength

of Donor-Specific Anti-HLA Antibody.*

All Patients
Characteristic (N=211) Positive Results on Cross-Matching Assay
CDC FCXM Multiplex Bead
(N=74) (N=95) (N=42)
Age (yr) 4413 4414 4612 4214
Female sex (%) 66.8 70.3 66.3 62.0
Race or ethnic group (%)
White 78.2 82.4 81.1 64.3
Black 13.7 10.8 12.6 21.4
Hispanic 1.9 2.7 0 4.8
Asian 0.5 0 0 2.4
Other 5.7 4.1 6.3 7.1
Blood-type incompatibility with donor (%) 10.9 5.4 14.7 11.9
Calculated panel-reactive antibody (%) 8223 9015 8027 7324
Donor-specific anti-HLA antibody (%)
HLA class I 41.2 33.8 44.2 47.6
HLA class II 25.6 24.3 25.3 28.6
HLA class I and II 33.1 41.9 30.5 23.8
Previous kidney transplants (%)
0 45.5 41.9 47.4 47.6
1 39.8 43.2 36.8 28.6
2 12.8 10.8 14.7 11.9
3 1.9 4.1 1.1 0
Plasmapheresis sessions (no.)
Before transplantation 44 65 32 34
After transplantation 54 86 43 53

* Plusminus values are means SD.

Positive CDC denotes positive cross-matching on complement-dependent cytotoxicity assay, positive FCXM positive
flow-cytometric cross-matching with negative CDC cross-matching, and positive multiplex bead positive bead cross-
matching with negative FCXM.
Race or ethnic group was self-reported.

comparison populations crossed (i.e., when haz-78.2% were white, 13.7% were black, and 8.1%
ards were not proportional across time), the time
were of other racial or ethnic groups. More than
at risk was divided into empirically (visually) de-
half the patients (54.5%) had received at least one
rived intervals, and separate log-rank tests were
previous kidney transplant, and 14.7% had under-
performed within these strata. We used time- gone at least two transplantations (Table 1). Pa-
dependent Cox models to test these inferences for
tients had received renal-replacement therapy for
sensitivity. All statistical analyses were performed
a mean (SD) of 8.87.7 years before desensitiza-
with the use of Stata/MP, version 11.0. tion. The average calculated cytotoxic panel-reac-
tive antibody level was 8223%, and 32.7% of pa-
R e sult s tients had a level of 98% or more.
For desensitization, patients were treated with
Patients and Matched Control Subjects a mean of 44 plasmapheresis sessions before
Of the 211 patients who underwent desensitiza- transplantation and 54 sessions after transplan-
tion and transplantation, 66.8% were women; tation. All patients received at least two plasma-

n engl j med 365;4 nejm.org july 28, 2011 321

The New England Journal of Medicine
Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 2. Comparison of HLA-Incompatible Kidney-Transplant Recipients and Matched Control Subjects.*

HLA-Incompatible
Recipients
Characteristic (N=211) Matched Control Subjects
Dialysis or
Dialysis Only Transplantation Therapy
(N=1050) (N=1040)
Age (yr) 4413 4512 4612
Blood type (% of patients)
O 47.4 51.8 49.4
A 37.4 32.9 35.2
B 12.3 12.7 12.9
AB 2.8 2.7 2.5
Female sex (% of patients) 66.8 65.2 64.8
Black race (% of patients) 13.7 13.2 13.8
Panel-reactive antibody level (%) 8223 8224 8224
No. of previous transplants (% of patients)
0 45.5 42.6 42.2
1 39.8 41.1 41.5
2 12.8 13.3 13.6
3 1.9 3.0 2.7
Years of renal-replacement therapy 8.87.7 8.77.0 8.66.8
Percent of years of renal-replacement therapy with 3033 2935 2831
afunctioning allograft
Diabetes (% of patients) 19.9 19.5 19.7

* Plusminus values are means SD. The results of pairwise comparisons between HLA-incompatible recipients and either
set of matched control subjects were not significant in any category.
Renal-replacement therapy includes both dialysis and previous renal transplantation.

pheresis sessions after transplantation, according
to the protocol. Donor-specific anti-HLA anti-
body was identified in all patients. Before the
initiation of plasmapheresis, the specificities of
donor-specific anti-HLA antibody were as fol-
lows: class I only, 41.2%; class II only, 25.6%; and
both class I and class II, 33.1%. Seventy-four pa-
tients had a positive cross-match on antihuman
globulinenhanced cytotoxicity assay before de-
sensitization, and 80.1% of patients had a positive
cross-match on either flow cytometry or comple-
ment-dependent cytotoxicity assay. There were no
significant differences in any of the matched vari-
ables between the treatment group and the two
control groups (Table 2).
Survival Benefit
During the overall study period, desensitization
was associated with a significant increase in the
rate of patient survival, as compared with the rates
in both the dialysis-only group and the dialysis-or-
transplantation group (Fig. 1). In the treatment
group, KaplanMeier estimates of the rates of sur-
vival were 90.6% at 1 year, 85.7% at 3 years, 80.6%
at 5 years, and 80.6% at 8 years, as compared
with rates of 91.1%, 67.2%, 51.5%, and 30.5%,
respectively, in the dialysis-only group and with
rates of 93.1%, 77.0%, 65.6%, and 49.1%, respec-
tively, in the dialysis-or-transplantation group
(P<0.001 for both comparisons). However, at the
end of the first year, there were no significant dif-
ferences in the rates of survival among the three
groups (P>0.20 for both comparisons).
Survival Benefit and Cross-Match Strength
A survival benefit was associated with desensitiza-
tion in all preselected categories of donor-specific
anti-HLA antibody level. Patients with positive
results on the bead assay (19.9% of the cohort)
had the shortest follow-up, with a survival rate of
90.8% at 48 months in the treatment group, as
compared with 57.4% for control subjects in the

322 n engl j med 365;4 nejm.org july 28, 2011

The New England Journal of Medicine

Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
 Desensitization in HLA-Incompatible Kidney Recipients

dialysis-only group and 67.3% for those in the

dialysis-or-transplantation group. Survival curves 100
overlapped throughout the first 12 months for 90
Desensitization treatment
80
patients in the treatment group and all control
70 Dialysis or
subjects (P=0.90 by the log-rank test). After 12

Survival (%)
60 transplantation
months, the rate of survival was higher in the 50
treatment group than in both the dialysis-only 40
group (P=0.008) and the dialysis-or-transplanta- 30
Dialysis only
tion group (P=0.02) (Fig. 2A). 20
Among patients in the treatment group who 10
had a positive flow-cytometric cross-match (45.0% 0
0 12 24 36 48 60 72 84 96
of the cohort), rates of survival were 92.0% at
Months
1 year, 85.5% at 3 years, 79.7% at 5 years, and
No. at Risk
79.7% at 8 years, as compared with rates of Desensitization 210 170 143 110 75 58 42 28 14
90.3%, 67.1%, 53.2%, and 33.4%, respectively, in treatment
Dual therapy 1027 854 688 497 321 230 157 96 41
the dialysis-only group and rates of 92.2%, 75.4%, Dialysis only 1012 822 626 419 250 159 93 54 17
64.8%, and 49.3%, respectively, in the dialysis-or-
transplantation group. The survival curve in the Figure 1. Survival Benefit of Desensitization in HLA-Incompatible Kidney
treatment group crossed at 12 months with the Recipients.
curve for the dialysis-only group and at 18 months KaplanMeier estimates of patient survival are shown for patients who un-
with the curve for the dialysis-or-transplantation derwent desensitization treatment before kidney transplantation (treatment
group. During these months, the survival rate in group), as compared with two matched control groups of patients on a kid-
ney waiting list who continued to receive dialysis (dialysis-only group) or
the treatment group was similar to the rates in who either continued to undergo dialysis or underwent HLA-compatible
both the dialysis-only group (P=0.75) and the dial- transplantation (dialysis-or-transplantation, or dual therapy, group). There
ysis-or-transplantation group (P=0.85). Thereafter, were 35 deaths in the treatment group. The most common cause of death
survival rates were higher in the treatment group was cardiac disease (ischemic or valvular), which accounted for 15 deaths
than in both the dialysis-only group (P<0.001) and (43%). Infections, primarily opportunistic pneumonias, were responsible
for 6 deaths (17%). The other causes of death were hemorrhage in 3 patients,
the dialysis-or-transplantation group (P=0.004) bowel perforation or loss of vascular access in 2 patients each, and pancre-
(Fig. 2B). atitis, cancer, motor vehicle accident, hypoglycemia, pulmonary embolism,
Among patients in the treatment group who pulmonary venous occlusion, and airway obstruction in 1 patient each.
had positive cross-matches on the complement-
dependent cytotoxicity assay, survival rates were
87.7% at 1 year, 82.0% at 3 years, 78.0% at 5 years, all rate of minor reactions to plasmapheresis (rash,
and 78.0% at 8 years, as compared with rates of itching, flushing, tachycardia, headache, nausea,
92.2%, 67.0%, 49.7%, and 27.1%, respectively, in shortness of breath, paresthesias, and hypotension
the dialysis-only group and with rates of 94.4%, or hypertension) was 10.9%. Major events that were
77.7%, 65.3%, and 44.6%, respectively, in the dial- associated with plasmapheresis, including anaphy-
ysis-or-transplantation group. The survival curves laxis with hypotension and airway edema, occurred
for the treatment group crossed at 18 months with in three patients (1.4%). Since plasmapheresis de-
the curve for the dialysis-only group and at 30 pletes coagulation factors, some of the bleeding
months with the curve for the dialysis-or-trans- complications may have been associated with the
plantation group. During these months, survival protocol. The rate of surgical-site bleeding that
rates in the treatment group were similar to those required a return to the operating room was 5.2%.
in the dialysis-only group (P=0.62) and in the Kidney biopsies in three patients (1.4%) were as-
dialysis-or-transplantation group (P=0.76). There- sociated with bleeding that required reoperation,
after, survival rates were higher in the treatment and graft loss occurred in one patient.
group than in both the dialysis-only group (P<0.001)
and the dialysis-or-transplantation group (P=0.004) Discussion
(Fig. 2C).
In this study, we evaluated rates of survival after
Adverse Events live-donor renal transplantation in 211 consecu-
Table 3 shows the frequency of adverse events as- tive patients with donor-specific anti-HLA anti-
sociated with desensitization treatment. The over- body who underwent desensitization with a stan-

n engl j med 365;4 nejm.org july 28, 2011 323

The New England Journal of Medicine
Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Positive Multiplex Bead Assay, Negative Flow-Cytometric Assay Figure 2. Survival Benefit of Desensitization According to
100 Strength of Cross-Matching for Donor-Specific Anti-HLA
Desensitization treatment Antibody.
90
Dialysis or transplantation KaplanMeier survival estimates are shown for kidney
80
70 recipients in the treatment group, as compared with
Survival (%)

60 the two matched control groups of patients on a kidney

Dialysis only
50 waiting list who continued to receive dialysis (dialysis-
only group) or who either continued to undergo dialysis
40
or underwent HLA-compatible transplantation (dialysis-
30
or-transplantation, or dual therapy, group), according to
20
whether donor-specific anti-HLA antibody was detected
10
on the multiplex bead assay (Panel A), flow-cytometric
0
assay (Panel B), or positive complement-dependent
0 6 12 18 24 30 36 42
cytotoxicity assay (Panel C).
Months
No. at Risk
Desensitization
treatment
38 30 26 23 18 15 14 9 dardized protocol that included plasmapheresis
Dual therapy 187 152 129 110 84 66 59 31 and the administration of low-dose intravenous
Dialysis only 182 147 123 101 74 57 50 27 immune globulin. During the 11-year study pe-
riod, the patients had a significant survival ben-
B Positive Flow-Cytometric Assay, Negative Cytotoxic Cross-Match efit, as compared with matched control subjects
100
who were on a kidney-transplant waiting list.
90
80
Desensitization treatment Sensitization to HLA is a serious public health
70 Dialysis or transplantation
problem inasmuch as it is observed in 30% of
Survival (%)

60 patients on the kidney-transplant waiting list. Of

50 these patients, 7908 are highly sensitized.5 De-
40 spite being given priority in the current organ-
30 Dialysis only allocation algorithm, highly sensitized patients
20
(those with a panel-reactive antibody level >80%)
10
0
have annual transplantation rates as low as 6.5%.18
0 12 24 36 48 60 72 84 96 In a study by Vo et al.,18 16 of 20 patients (80%)
Months who underwent desensitization with intravenous
No. at Risk immune globulin and anti-CD20 received a renal
Desensitization 88 71 58 39 27 23 17 16 10 transplant within a mean of 12 months after treat-
treatment
Dual therapy 425 346 276 176 114 94 67 56 27 ment. In our study, 211 of 215 of patients (98%)
Dialysis only 420 335 260 154 88 68 43 35 12 who began treatment with plasmapheresis pro-
gressed to transplantation. Desensitization has
C Positive Cytotoxic Cross-Match
the potential to greatly increase access to trans-
100
90
plantation for this disproportionately disadvan-
80 Desensitization treatment taged population of candidates for transplanta-
70 Dialysis or tion and to increase their longevity.
Survival (%)

60 transplantation Concerns about the durability of allograft func-

50 tion, patient safety, and the financial burden on the
40 health care system associated with desensitization
30
Dialysis only
protocols have persisted in the absence of studies
20
reporting outcomes beyond 2 years. Data from
10
0
studies to date have been difficult to interpret or
0 12 24 36 48 60 72 84 96 compare because of heterogeneity among desensi-
Months tization strategies, histocompatibility testing tech-
No. at Risk niques, donor-specific anti-HLA antibody levels,
Desensitization 74 63 57 48 38 31 23 11 3 and demographic and clinical characteristics of
treatment
Dual therapy 365 330 281 221 163 115 79 35 9 donor and recipient populations. In our study, early
Dialysis only 360 317 252 184 132 81 48 19 5 rates of death after transplantation were higher
than expected for live-donor transplantation,25 per-

324 n engl j med 365;4 nejm.org july 28, 2011

The New England Journal of Medicine

Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
 Desensitization in HLA-Incompatible Kidney Recipients
haps in part reflecting the excess burden of co-
existing illnesses associated with sensitization. As
a result, direct comparisons of outcomes with those
of compatible live-donor transplantations may be
misleading. Notably, the baseline characteristics
of our study patients were skewed toward high
degrees of sensitization and prolonged periods of
renal-replacement therapy.
In our study, most of the adverse events associ-
ated with desensitization treatment were mild,
but more serious reactions, including anaphylaxis
and bleeding, did occur. In addition, there were six
deaths from infection, and although a direct link
to desensitization could not be established, such a
connection merits further investigation.
To determine the expected survival rate among
patients with similar phenotypes on the waiting
list, we identified two matched control groups: one
in which patients continued to receive dialysis and
one in which some of the patients receiving dialy-
sis underwent transplantation of an HLA-compat-
ible kidney during the study period. In the first
12 months, there was no significant difference in
the survival rate between the treatment group and
either control group. However, after the first year,
the rate of survival in the treatment group was
significantly higher. Furthermore, the survival
benefit in the treatment group was preserved in
all three subgroups defined on the basis of the
donor-specific anti-HLA antibody level.
We and other investigators have identified an
increase in the donor-specific anti-HLA antibody
level as an important predictor of reduced graft
survival, and increased antibody levels require
more intensive plasmapheresis before transplan-
tation.26 In our study, however, the rate of survival
in the treatment group was superior to that in ei-
ther matched control group, even among patients
with the highest cross-match levels.
For patients who have a willing but incompat-
ible kidney donor, the alternative to desensitiza-
tion is kidney paired donation, in which patients
are entered into a pool and matching algorithms
are used to identify compatible pairings.27,28 On
the surface, this would appear to be the least
costly and most effective method for treating pa-
tients who have a positive cross-match with their
donor, but since match rates for highly sensitized
patients are very low, kidney paired donation is not
a good solution for most sensitized patients.4,29
Plasmapheresis does not result in a durable re-
duction in HLA antibody unless the patient under-
n engl j med 365;4 nejm.org july 28, 2011
The New England Journal of Medicine
Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
Table 3. Frequency of Adverse Events in 211 Patients
during Desensitization Treatment.*
Adverse Event Frequency
no. of patients (%)
Related to plasmapheresis
Minor event 23 (10.9)
Major event 3 (1.4)
Bleeding after biopsy 3 (1.4)
Reoperation for bleeding 11 (5.2)
* Minor events included rash, itching, flushing, tachycar-
dia, headache, nausea, shortness of breath, paresthesias,
and hypotension. Major events included anaphylaxis with
hypotension and airway edema.
goes transplantation within several days after the
last treatment. This factor accounts for the paucity
of reports of protocols that use plasmapheresis to
desensitize patients who are on the waiting list for
a transplant from a deceased donor. The imple-
mentation of this type of protocol would require
that the patient receive some sort of special status
to hasten the availability of an organ from a de-
ceased donor. Studies of desensitization with the
use of high-dose intravenous immune globulin in
patients without live donors have shown increased
rates of transplantation and good outcomes.18
Our study has certain limitations. These in-
clude the single-center design and follow-up for
less than 3 years for half the patients in the
cohort. Thus, our findings will need to be con-
firmed in multicenter trials among nonreferral
populations.
With the elucidation of donorrecipient phe-
notypes that will benefit the most from desensi-
tization, long-term outcomes associated with the
procedure are likely to improve. Also, earlier refer-
ral for desensitization treatment, which would re-
sult in fewer years of renal-replacement therapy,
would be predicted to reduce mortality. Outcomes
could be improved and costs could be lowered if
patients were not required to have a completely
negative cross-match for a kidney paired donation
and if highly sensitized patients could be matched
with a donor for whom HLA antibody cross-match-
ing was weaker than that for their original donor.30
This study shows that sensitized patients, who
have had limited access to transplantation for 50
years, can derive a survival benefit from desensi-
tization and receipt of a kidney transplant from
a live donor.
325
 Desensitization in HLA-Incompatible Kidney Recipients

Supported by a grant (RC1 DK086731) from the National In-
stitute of Diabetes and Digestive and Kidney Diseases and by a
grant from the Charles T. Bauer Foundation (to Drs. Montgom-
ery, Warren, and Segev).
Dr. Montgomery reports that his institution, John Hopkins
University, has received grant support from Alexion, Genzyme,
and Viropharma related to HLA-incompatible transplantation
and that he has received lecture fees from educational grants to
sponsoring academic institutions from Astellas and Genzyme;
Dr. Zachary, receiving lecture fees from Genzyme and having an
equity interest in Luminex; and Dr. Segev, receiving consulting
and lecture fees from Genzyme. No other potential conflict of
interest relevant to this article was reported.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
We thank Vanessa Collins for providing data-management
services and Julie Houp for her technical assistance with immu-
nogenetics procedures.

References

1. Port FK, Wolfe RA, Mauger EA, Ber-
ling DP, Jiang K. Comparison of survival
probabilities for dialysis patients vs ca-
daveric renal transplant recipients. JAMA
1993;270:1339-43.
2. Wolfe RA, Ashby VB, Milford EL, et al.
Comparison of mortality in all patients
on dialysis, patients on dialysis awaiting
transplantation, and recipients of a first
cadaveric transplant. N Engl J Med 1999;
341:1725-30.
3. Lee AJ, Morgan CL, Conway P, Currie
CJ. Characterisation and comparison of
health-related quality of life for patients
with renal failure. Curr Med Res Opin 2005;
21:1777-83.
4. Segev DL, Gentry SE, Warren DS,
Reeb B, Montgomery RA. Kidney paired
donation and optimizing the use of live
donor organs. JAMA 2005;293:1883-90.
5. United Network for Organ Sharing
home page. (http://www.unos.org.)
6. Patel R, Terasaki PI. Significance of the
positive crossmatch test in kidney trans-
plantation. N Engl J Med 1969;280:735-9.
7. Williams GM, DePlanque B, Lower R,
Hume D. Antibodies and human transplant
rejection. Ann Surg 1969;170:603-16.
8. Starzl TE, Marchioro TL, Holmes JH,
et al. Renal homografts in patients with
major donor-recipient blood group in-
compatibilities. Surgery 1964;55:195-200.
9. Lee PC, Terasaki PI, Takemoto SK, et
al. All chronic rejection failures of kidney
transplants were preceded by the develop-
ment of HLA antibodies. Transplantation
2002;74:1192-4.
10. Glotz D, Antoine C, Julia P, et al. De-
sensitization and subsequent kidney trans-
plantation of patients using intravenous
immunoglobulins (IVIg). Am J Transplant
2002;2:758-60.
11. Tyan DB, Li VA, Czer L, Trento A, Jor-
dan SC. Intravenous immunoglobulin sup-
pression of HLA alloantibody in highly
sensitized transplant candidates and trans-
plantation with a histoincompatible organ.
Transplantation 1994;57:553-62.
12. Montgomery RA, Zachary AA, Racu-
sen LC, et al. Plasmapheresis and intrave-
nous immune globulin provides effective
rescue therapy for refractory humoral re-
jection and allows kidneys to be success-
fully transplanted into cross-match-posi-
tive recipients. Transplantation 2000;70:
887-95.
13. Issa N, Cosio FG, Gloor JM, et al.
Transplant glomerulopathy: risk and prog-
nosis related to anti-human leukocyte an-
tigen class II antibody levels. Transplan-
tation 2008;86:681-5.
14. Higgins RM, Bevan DJ, Carey BS, et
al. Prevention of hyperacute rejection by
removal of antibodies to HLA immedi-
ately before renal transplantation. Lancet
1996;348:1208-11.
15. Stegall MD, Gloor J, Winters JL,
Moore SB, Degoey S. A comparison of
plasmapheresis versus high-dose IVIG de-
sensitization in renal allograft recipients
with high levels of donor specific alloan-
tibody. Am J Transplant 2006;6:346-51.
16. Higgins R, Hathaway M, Lowe D, et
al. Blood levels of donor-specific human
leukocyte antigen antibodies after renal
transplantation: resolution of rejection in
the presence of circulating donor-specific
antibody. Transplantation 2007;84:876-84.
17. West-Thielke P, Herren H, Thielke J, et
al. Results of positive cross-match trans-
plantation in African American renal
transplant recipients. Am J Transplant
2008;8:348-54.
18. Vo AA, Lukovsky M, Toyoda M, et al.
Rituximab and intravenous immune glob-
ulin for desensitization during renal trans-
plantation. N Engl J Med 2008;359:242-51.
19. Haririan A, Nogueira J, Kukuruga D,
et al. Positive cross-match living donor
kidney transplantation: longer-term out-
comes. Am J Transplant 2009;9:536-42.
20. Segev DL, Simpkins CE, Warren DS, et
al. ABO incompatible high-titer renal trans-
plantation without splenectomy or anti-
CD20 treatment. Am J Transplant 2005;
5:2570-5.
21. Montgomery RA, Zachary AA. Trans-
planting patients with a positive donor-
specific crossmatch: a single centers per-
spective. Pediatr Transplant 2004;8:535-
42.
22. Zachary AA, Montgomery RA, Ratner
LE, et al. Specific and durable elimination
of antibody to donor HLA antigens in re-
nal-transplant patients. Transplantation
2003;76:1519-25.
23. Montgomery RA, Locke JE, King KE,
et al. ABO incompatible renal transplan-
tation: a paradigm ready for broad im
plementation. Transplantation 2009;87:
1246-55.
24. Segev DL, Muzaale AD, Caffo BS, et
al. Perioperative mortality and long-term
survival following live kidney donation.
JAMA 2010;303:959-66.
25. Organ Procurement and Transplanta-
tion Network. Scientific registry of trans-
plant recipients. (http://optn.transplant
.hrsa.gov.)
26. Gloor JM, Winters JL, Cornell LD, et
al. Baseline donor-specific antibody levels
and outcomes in positive crossmatch kid-
ney transplantation. Am J Transplant
2010;10:582-9.
27. Montgomery RA, Zachary AA, Ratner
LE, et al. Clinical results from transplant-
ing incompatible live kidney donor/recipi-
ent pairs using kidney paired donation.
JAMA 2005;294:1655-63.
28. Montgomery RA, Gentry SE, Marks
WH, et al. Domino paired kidney dona-
tion: a strategy to make best use of live
non-directed donation. Lancet 2006;368:
419-21.
29. Gentry SE, Segev DL, Montgomery RA.
A comparison of populations served by
kidney paired donation and list paired do-
nation. Am J Transplant 2005;5:1914-21.
30. Montgomery RA. Renal transplanta-
tion across HLA and ABO antibody barri-
ers: integrating paired donation into de-
sensitization protocols. Am J Transplant
2010;10:449-57.
Copyright 2011 Massachusetts Medical Society.

326 n engl j med 365;4 nejm.org july 28, 2011

The New England Journal of Medicine

Downloaded from nejm.org by ALOK GUPTA on April 1, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.