Identification of acceptable HLA mismatches for highly
Download
Report
Transcript Identification of acceptable HLA mismatches for highly
Predicting the
immunogenicity of structural
HLA class I epitopes
Reyna Goodman
Introduction
Between 5-10% of patients waiting for a renal
transplant are classified as highly sensitised
(Panel Reactive Activity ≥85% IgG)
Highly sensitised renal dialysis patients wait
longer for a suitable crossmatch negative donor
compared to non-sensitised patients
Identification of acceptable HLA mismatches
increases the likelihood of transplantation
HLA specific antibody screening using
single antigen beads
Beads are coated with single HLA specificities
Each bead population has a different ratio of
two dyes which allows identification of up to
100 individual HLA specificities
Analysed using a Luminex platform
HLAMatchmaker
A computer algorithm that determines HLA
compatibility at a structural level by
comparing differences between polymorphic
amino acid triplets in the antibody accessible
region of the HLA molecule
HLAMatchmaker performs inter-locus and
intra-locus comparisons between the patient’s
HLA type and each mismatched HLA
specificity and calculates the number of
amino acid triplet mismatches
Aim
To determine whether HLAMatchmaker could be
used to predict the immunogenicity of structural
epitopes
comparing acceptable HLA mismatches identified
using single antigen HLA specific antibody detection
beads with those predicted using the HLAMatchmaker
computer algorithm
To identify acceptable HLA mismatches for highly
sensitised patients
Patient cohort
Of 406 patients on the Addenbrooke’s Hospital renal
transplant waiting list, 24 were identified as highly sensitised
and selected for study
Male : Female
10:14
Pregnancies (females)
9/14
(64%, range 1 –5)
Previous transplants
17/24
(71% range 1 – 3)
Blood transfusions
23/24
(96% range 2 – >70)
Time waiting
Mean 7.5 years
(range 3 – 21 years)
Age
Mean 43 years
(range 27-60 years)
Highest IgG PRA
Mean 96%
(range 87% – 100%)
Data analysis
Single antigen beads detect antibody to 65 individual
HLA-A and -B specificities of which 64 are represented
in the HLAMatchmaker algorithm
Patient HLA class I types and each mismatched HLA
specificity represented on the single antigen beads were
entered into the HLAMatchmaker program to determine
the number of triplet mismatches
Logistic regression analysis was used to determine the
relationship between the detection of antibody using
single antigen beads and the number of amino acid
triplet mismatches for each HLA specificity determined
by HLAMatchmaker
HLA specific antibody screening using
single antigen beads
The serum sample with the highest PRA was selected
for study
Of 1,451 mismatched HLA specificities, 972 (67%) were
antibody positive
A mean of 19 (range 1 - 41) acceptable mismatches
were identified for each patient
For 19 patients there were 10 or more acceptable
mismatches identified including some common antigens
Presence of antibody for each mismatched HLA specificity
categorised by triplet mismatch
Number of amino acid
triplet mismatches
Total number of
HLA specificities
Antibody positive
HLA specificities
0
46
4 (9%)
1
98
40 (41%)
2
145
61 (42%)
3
205
121 (59%)
4
214
136 (64%)
5
197
155 (79%)
6
178
141 (79%)
7
141
116 (82%)
8
115
101 (88%)
9
58
53 (91%)
10
30
27 (90%)
11
15
12 (80%)
12
7
5 (71%)
13
2
0 (0%)
Total
1,451
972 (67%)
Logistic regression analysis of antibody
binding to single HLA specificities
stratified by number of triplet mismatches
% antibody positive
4
0
2
0
0
P<0.001
Number of amino acid triplet mismatches
Proportion antibody positive
Individual logistic regression analyses of antibody binding to
single HLA specificities stratified by number of triplet
mismatches
Patient 1
Patient 2
Patient 3
Patient 4
Patient 5
Patient 6
Patient 7
Patient 8
Patient 9
Patient 10
Patient 11
Patient 12
Patient 13
Patient 14
Patient 15
Patient 16
Patient 17
Patient 18
Patient 19
Patient 20
Patient 21
Patient 22
Patient 23
Patient 24
Number of triplet mismatches
Conclusions
HLAMatchmaker is an effective tool for predicting the
immunogenicity of a particular HLA mismatch
The number and nature of triplet mismatches for a
given HLA type correlates with the risk of humoral
sensitisation
The presence of a single triplet amino acid mismatch is
often sufficient to invoke a strong antibody response
In combination with single antigen beads, acceptable
mismatches were identified for highly sensitised
patients, increasing access to the donor pool
• Goodman et al Transplantation 2006;81:1331-1336
Triplets versus eplets
Examine the extent to which the structural
information provided by triplet and eplet epitope
analysis of HLA compatibility enables the
prediction of:
Acceptable mismatches
Magnitude of the antibody response
Data analysis
Sera obtained from 34 highly sensitised patients were
screened using single antigen beads, to determine the
presence and magnitude of HLA alloantibody
Patient HLA class I types and each mismatched HLA
specificity represented on the single antigen beads were
entered into the HLAMatchmaker program to determine the
number of triplet and eplet mismatches
A total of 85 sera were screened (median 2 sera per patient,
range 1 to 6) and 2,088 mismatched combinations examined
The qualitative and quantitative relationship between
alloantibody levels to each HLA specificity and the number of
triplet and eplet mismatches was determined
Number of triplet & eplet mismatches
present within mismatched HLA
specificities
Frequency
Frequency
0
0
50
50
100
100
150
150
200
200
250
250
300
300
P<0.001
0
1
2
3
4
5
6
Triplets
7
8
9
10
11
12
13
0
1 2
3
4
5
6
7 8
9 10 11 12 13 14 15 16 17 18
Eplets
Relationship between triplet & eplet
mismatches and alloantibody
production
Proportion Antibody
Positive
Proportion Antibody
Positive
P<0.001
P<0.001
P<0.001
Triplets
P<0.001
Eplets
Relationship between triplet & eplet
mismatches and alloantibody levels
Median Fluorescence
Intensity
Triplets
P>0.001
Median Fluorescence
Intensity
Eplets
P>0.001
Conclusions
The number of triplet and eplet mismatches
between an alloantigen and the recipient HLA
type correlates closely with both the
development and strength of an alloantibody
response
Eplets offer additional epitope discrimination but
do not improve HLA immunogenicity prediction
Self triplets and eplets may form immunogenic
epitopes when expressed in a different
conformation on mismatched HLA alloantigens
Acknowledgements
H&I department
Cambridge University Hospitals
NHS Foundation Trust,
Addenbrooke’s Hospital
Dr Craig Taylor
Miss Cheryl O’Rourke
Mr Tim Key
Centre for Applied medical statistics,
University of Cambridge
Mr Andrew Lynch
Dr Linda Sharples
Department of Surgery,
University of Cambridge,
Addenbrooke’s Hospital
Prof J Andrew Bradley
Mr Vasilis Kosmoliaptsis
Patient follow-up
5/24 patients transplanted
Renal
Number of Number of
transplant mismatched mismatched
recipient
triplets
eplets
1
2
3
4
5
0
0
4
6
10
0
0
5
8
11
HLA
mismatch
grade
Graft
survival
(years)
0.0.0
0.0.0
0.1.0
1.0.1
1.0.1
1
2.3
0.9
3
1.2