Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D.,...
-
Upload
coleen-dean -
Category
Documents
-
view
219 -
download
0
Transcript of Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D.,...
![Page 1: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/1.jpg)
Maximizing Kidney Paired Donation
CSGF Fellows Conference 2005
Sommer Gentry, MIT
Dorry Segev, M.D.,
Robert A. Montgomery, M.D., Ph.D.
Johns Hopkins School of Medicine
![Page 2: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/2.jpg)
ESRD and kidney donation
• End-stage-renal-disease (kidney failure) strikes tens of thousands of people– Kidneys can no longer purify the blood
• Treatments include– Dialysis, artificial kidney support, is
extremely expensive ($60,000 per year) and debilitating
– Kidney transplantation generally restores a patient to full health
![Page 3: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/3.jpg)
Deceased and live donors
• The demand for deceased donor kidneys far outstrips the supply (UNOS)– About 10,000 deceased donor kidneys per year
are transplanted– Over 60,000 patients are on the waiting list– Deceased donation is level
• Live donation is an attractive option– Operation is very safe, can be laparoscopic– About 6,000 live donations in 2003– Spouses, friends, siblings, parents, children
![Page 4: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/4.jpg)
Donor Incompatibilities
• Many willing live donors are disqualified– Bloodtypes: O(46%), A(34%), B(16%), AB(4%)
• O recipients can accept only an O kidney• A recipients can accept only O, A kidneys• B recipients can accept only O, B kidneys• AB recipients can accept any kidney
– Positive crossmatch (XM+) predicts rejection
• Even blood compatible donors might have XM+• Crossmatch is difficult to predict• Highly sensitized patients almost always XM+
![Page 5: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/5.jpg)
Kidney Paired Donation (KPD)
• Donations must be simultaneous
Donor Recipient
A
B A
B
ABO Incompatible
ABO Incompatible
![Page 6: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/6.jpg)
Kidney Paired Donation (KPD)
Donor Recipient
A
O A
O
ABO Incompatible
Positive Crossmatch
![Page 7: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/7.jpg)
Simulation and Algorithms
• Because it is critically important not to waste offers of live donor kidneys, the matching algorithm must be analyzed in advance of any national implementation
• Simulated patients and virtual exchanges can answer questions about the outcomes of a matching algorithm
![Page 8: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/8.jpg)
Donor / Recipient bloodtypes
• ABO distribution(Zenios, Woodle, Ross, Transplantation 72:4, 2001)(Zenios, Management Science 48:3, 2002)
• Each recipient has up to four potential donors– Mother, father, spouse, sibling
• Bloodtypes distributed as in population– Parent has two alleles
• AA, AO, BB, BO, OO, or AB
– A child receives one allele from each parent
![Page 9: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/9.jpg)
Decision tree model of family
Potential donors
Medical workup (pass 56% or 75%),
crossmatch tests (11%), bloodtyping
Incompatibledonor/recipient
pairsDirect donation
(Zenios, Woodle, Ross, Transplantation 72:4, 2001)
3584 pairs annually
No willing, medically
clear donor
Simulate until reach# of real live donors
![Page 10: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/10.jpg)
Graph showing recipient/donor pairs
Donor: BRecipient: A
Donor: ARecipient: B
Donor: ORecipient: A
XM+
Donor: ARecipient: O
Pair 2
Pair 1 Pair 4
Pair 3
Node isan incompatible donor / recipient
pair
Edge connects two
pairs if an exchange is
possible
![Page 11: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/11.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39
![Page 12: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/12.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39
Node 22: (13 edges)Donor Type ORecipient Type AWilling to travel
![Page 13: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/13.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39
Node 5: (3 edges)Donor Type ARecipient Type OWilling to travel
![Page 14: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/14.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39Node 35: (1 edge)Donor Type ARecipient Type OUnwilling to travel
Node 6: (1 edge)Donor Type ARecipient Type OUnwilling to trade
with older donors
![Page 15: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/15.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39
![Page 16: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/16.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
23
24
25
26
2728
29 30 3132
33
34
35
36
37
39
![Page 17: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/17.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39
Maximal matching:Accept matches at random until no edges remain.
Only 10 of 40 patients get a transplant
![Page 18: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/18.jpg)
0
1
2
3
4
5
6
78
9101112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2728
29 30 3132
33
34
35
36
37
38
39
Maximum cardinality matching:Paths, Trees, and Flowers, Edmonds (1965)
14 of 40 get transplants
![Page 19: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/19.jpg)
Maximizing # of transplants
• First-Accept method finds maximal matching– When an incompatible pair arrives, search the
entire current list for a feasible trade, and if one is found, perform that exchange
• Maximum cardinality matching method– Over the entire pool, periodically choose the
exchanges that result in the greatest benefit– Combinatorial explosion: for 100 donor/recipient
pairs, there are millions of possible matchings, for 1000, there are 10250
– Edmonds’ algorithm finds, efficiently, the exchanges that yield the maximum number of transplants
![Page 20: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/20.jpg)
Flexibility: edge weights• All exchanges are equal, but some
exchanges are more equal than others– Bonus points for disadvantaged groups, like O
patients, the highly sensitized, or those who have been waiting a long time
– Points to reward clinical predictors of good outcomes: HLA antigen mismatch
– Patient priorities: donor age versus different transplant center
• Maximum edge weight matching problem: find the matching that maximizes the sum of the included edge weights (Edmonds, 1965)
![Page 21: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/21.jpg)
Advantage of Optimization
• First-Accept and Optimized matching algorithms tested– 30 simulated pools of 4000 patients– First-Accept averages
• 1673 transplants
– Optimized averages• 1891 transplants
(Segev, Gentry, et al., Journal of the American Medical Association, 2005)
![Page 22: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/22.jpg)
PKE pool size advantage
0
5
10
15
20
25
30
35
40
45
50
100 500 1000 5000 100 500 1000 5000
# Donor/Recipient Pairs
% M
atch
ed
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
HL
A M
ism
atch
First-Accept Optimized
![Page 23: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/23.jpg)
Minimizing travel outside region
0
10
20
30
40
50
60
Fir
st-A
Op
timiz
ed
Fir
st-A
Op
timiz
ed
Fir
st-A
Op
timiz
ed
Fir
st-A
Op
timiz
ed
Fir
st-A
Op
timiz
ed
None Sens Only Sens+10% Sens+25% All Travel
%Must Travel
%No Travel
%Must Travel
%No Travel
42.7% 42.0%
![Page 24: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/24.jpg)
Financial benefit to mathematical literacy
• Long wait times on dialysis are extremely expensive
• $290 million saved over dialysis using any form of paired kidney donation
• $50 million additional saved over dialysis using maximum edge weight matching for paired kidney donation
(Segev, Gentry, et al., Journal of the American Medical Association, 2005)
![Page 25: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/25.jpg)
• Maximum edge weight does not necessarily yield a maximum cardinality matching
• We use a 100 point bonus for all edges, then differentiate between edges in the range of 0-10 points – within .4% of the size of maximum cardinality
matching, with vastly reduced travel requirements and HLA antigen mismatch
Edge weight points system
100
10 10
![Page 26: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/26.jpg)
Alternative optimal matchings
• Roth et al. have suggested using maximum node weight matchings– Advantage: all maximum node weight matchings
are also maximum cardinality matchings– Disadvantage: node weights can not capture
priorities we wish to place on particular exchanges, like a very good compatibility match, or an exchange between patients at the same hospital that does not require travel
(Roth, Sonmez, Ünver, Pairwise Kidney Exchange, working paper, 2005)
10
10 20
20
![Page 27: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/27.jpg)
Waiting times and accumulation
• Pairs that do not match will accumulate– Waiting list like that for deceased donors– How will the waiting list evolve? How long
will patients wait?
• Iterated optimization rather than static– As the waiting list grows, so do
computation times for the optimal solution
• Currently purchasing a Linux cluster for embarassingly parallel Monte Carlo simulations
![Page 28: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/28.jpg)
Unmatched after paired donation: 52%# Pairs dO dA dB dABrO 214 849 293 50rA 26 155 23 90rB 14 22 39 62rAB 1 7 5 6
(Gentry, Segev, Montgomery, Am. J. Transplantation, 2005)
# Pairs dO dA dB dABrO 657 1066 354 51rA 257 429 174 107rB 83 175 91 72rAB 12 27 17 11
All incompatible donor / recipient pairs
![Page 29: Maximizing Kidney Paired Donation CSGF Fellows Conference 2005 Sommer Gentry, MIT Dorry Segev, M.D., Robert A. Montgomery, M.D., Ph.D. Johns Hopkins School.](https://reader035.fdocuments.us/reader035/viewer/2022062408/56649e9e5503460f94b9f4b5/html5/thumbnails/29.jpg)
Deceased donation requires a simpler optimization than KPD
Deceaseddonorkidney
Recipients
12
8
7
10
4
10
6
0
1
2
3
45
67
8910111213
1415
16
17
18
19
20
21
22
23
2425
2627
28 29 30 31 3233
3435
36
37
38
39
10 10
21
10
10
10
10
10
32
32
32
32
32
32
32
21
21
21
21
21
21
2121
21
21
21
21
6
6
6
6
6