Juliet N. Barker, MBBS (Hons), FRACP

1
Optimising Cord Blood Unit Selection
-3
-2
-1
-4
-7
-6
-5
30
100
0
CSA/ MMF
Juliet N. Barker, MBBS (Hons), FRACP Associate
Attending Director Cord Blood Transplant
Program Memorial Sloan-Kettering Cancer Center
2
Acknowledgements
MSKCC Staff of Adult and Pediatric
Transplant Search Courtney Byam, Rosanna
Ferrante Debbie Wells, Kathleen Doshi, Sinda
Lee Cytotherapy Lab esp Allison Schaible CB
Research Staff Marissa Lubin Anne Marie Gonzales
, Katie Evans Cellular Immunology Lab Kathy
Smith Malcolm Moore Machi Scaradavou Nancy Kernan
Richard OReilly Doris Ponce Marcel van den
Brink Sergio Giralt
U of Minnesota John E. Wagner NYBC Pablo
Rubinstein Cladd Stevens Machi Scaradavou
3
What have we achieved?
4
Retrospective studies suggest improved
engraftment GVL.
5
MSKCC Donor Algorithm
Sibling typing ? simultaneous URD CB search
Suitable Sibling (match/ donor health) Suitable
URD (match/ availability)
Suitable CB Graft (match/ dose) 4-6/6 A,B
antigen, DRB1 allele 2 units each gt 2 x 107 NC/kg
Hi Dose TCD 9-10/10 donor
Midi/ Mini 10/10 donor
Hi Dose Prep Midi or Mini
(Unmodified) Children (Young adults)
Donors identified for gt 95 patients.
6
3 intensities, mainly Cy-Flu-TBI based, no ATG,
no steroids.
7
Neutrophil Engraftment after DCBT (n
108) Median 41 yrs (range 6-69), high risk
heme malignancies
Early auto recovery switched to sustained
donor engraftment
High rates of sustained donor engraftment.
Dahi, P., ASBMT 2012
8
(No Transcript)
9
Comparison of Donor-Recipient HLA-Match CB (n
75, 150 units) vs URD (n 184)
P lt 0.001
Ponce, BBMT 2011
CB grafts marked HLA-disparity. CD34 cell
dose also much lower RD 7.9, URD 6.0, CB 0.09 (
p lt 0.001).
10
DCBT if Acute Leukemia MDS/MPD 2-yr DFS
Children (n 23, median 9 yrs, range 0.9-15)
78
Low incidence of relapse (9 children, 6
adults) translates to relatively high survival
rates.
Adults (n 52, median 41 yrs, range 16-69)
64
Disease-Free Survival
Inf. TNC 3.3 2.6 2.7 1.9
Time Post-Transplant (Months)
Barker et al, ASH 2011
11
DCBT if Acute Leukemia MDS/MPD 2-yr DFS
Children 0-15 yrs (n 23) 78 (Europeans
86, Non-Europeans 75)
No difference between European non-European pati
ents. In multivariate analysis only CMV
serostatus was significant.
Adults 16-69 yrs (n 52) 64 (Europeans
62, Non-Europeans 66 )
Disease-Free Survival
Time Post-Transplant (Months)
Barker et al, ASH 2011
12
Why are these results important?
13
Best Matched URD Best CB if Combined Search by
Patient Ancestry (n 525)
Best Donors Europeans (n 341) Non-Europeans (n 184) p
Best URD
10/10 (n 218) 180 (53) 38 (21) lt0.001
9/10 (n 148) 99 (29) 49 (27)
lt8/10 (n 159) 62 (18) 97 (53)
Best CB
5-6/6 (n 401) 270 (79) 131 (71)
4/6 (n 90) 56 (16) 34 (18)
No CB (n 34) 15 (4) 19 (10)
Volunteer unrelated donors poor HSC source for
non-Europeans.
Barker et al 2010, BBMT
14
CB Extends Transplant Access to
Minorities URD vs CB vs No Graft by Ancestry
(n 385)
Barker et al 2010, BBMT
15
Updated Data, MSKCC 2012 (n 597)
URD (n426)
CB (n137)
No Graft (n34)
25 53
76
  NW Europe   Asian

  Eastern Europe   African

  Southern Europe   White Hispanic

  Europe Mix   Middle Eastern

  Non-Europe Mix
Greater than 50 of CBTs had non-European ancestry
16
Variables that Determine Outcome
Transplant Related Factors
CB Dose, match, quality

• Conditioning
• High, Midi , Mini

-7
28
100
180
1 year
0

• Immunosuppression rejection/ GVHD
• Supportive care infection, bleeding,
• nutrition

• Patient Related Factors
• Biology of Malignancy determines need for hi
  dose prep vs reliance on GVL
• Patient Characteristics age, extent of prior Rx,
  co-morbidities.

17
How to Select Units?
18
TRM by Combined TNC Dose A,B Antigen, DRB1
Allele-Match 1061 NYBC Single Unit Myeloablative
CBT 1993-2006
Very high TRM if mismatch low TNC
Lowest TRM 6/6 match
Lowest TRM best HLA-match, not highest dose.
Barker et al, Blood 2010
19
Implications for Unit Selection (applies to
single unit CBT, may also apply to double)

• Biggest cell dose not necessarily the best.
• 6/6 units highly attractive (?cell dose
  threshold).
• Sliding scale more mismatch, greater required
• cell dose. Converse also true match can
  compensate
• for low dose.
• Implies
• Above a cell dose threshold best matched unit
  the best.
• New measures needed if best unit is mismatched.

Barker, Blood 2010
20
Additional factors to consider in unit selection
- revealed in investigation of double unit
biology
21
Engraftment in 44 Double Unit CBTs Engrafting
with a Single Unit.
CD34 Cell Viability Engrafting Unit (N44) Non-Engrafting Unit (N44)
lt75 (N16) 1 15
75 (N72) 43 29
Using CD34 viability threshold of 75
(mean-2SD), all but one (43/44) engrafting units
had CD34 viability gt75 (p0.0006) OR Only 1/16
poor viability units engrafted. Poor CD34
viability correlated with lower CFUs (p0.02).
Scaradavou, BBMT 2010
22
Unit Quality Schema of CD34s of 2 CB Units
Total CD34 Cells in 2 Units
BAD UNIT GOOD
GOOD UNIT
50 viable
90 viable
Unit 1 Unit 2
Units similar infused viable CD34 doses-but very
different. In part, double unit CBT effective
as increases chance of transplanting at least one
good quality unit.
Scaradavou, BBMT 2010
23
Implications

• Unit quality varies from unit to unit, bank to
  bank.
• Not all banks are the same.
• Factors that dictate unit quality need to be
  determined
• eg collection standards, processing
  methodology, red
• cell content, cryo volume, age.
• Methods to test unit quality prior to thaw
  should
• be priority eg testing the segment.

24
Do the principles of single unit CBT also apply
to double unit CBT?
25
Sustained Neutrophil Engraftment After
Myeloablative DCBT by CD34 Cell Dose of
Engrafting Unit (n 61)
1.0-2.0 (n13) 100 _at_ 20 days
gt2.0 (n10) 100 _at_ 16.5 days
lt1.0 (n38) 89 _at_ 27.5 days
P lt 0.001
High rate sustained engraftment directly
dependent on infused CD34 of winner if low can
be very slow.
Avery, Blood 2011
26
Total Graft Cell Dose DCB Engraftment (n 61)
1.0
1.0
gt4.3 x107/kg 100
gt1.8 x105/kg 97
0.8
0.8
lt4.3 x107/kg 87
lt1.8 x105/kg 90
0.6
0.6
0.4
0.4
TNC
CD34
0.2
0.2
p 0.10
p 0.0007
Sustained Neutrophil Engraftment
0.0
0.0
0 10 20
30 40 50
0 10 20
30 40 50
gt6.2 x104/kg 97
gt7.8 x106/kg 97
lt6.2 x104/kg 90
lt7.8 x106/kg 90
CD3
CFU
0 10 20
30 40
50
0 10 20
30 40
50
Time Post Transplant (Days)
Total TNC CD3 dose of graft also have an
effect.
Avery, Blood 2011
27
Grade III-IV aGVHD by Engrafting Unit-Recipient
10 Allele HLA-Match (n 115)
100
Recipient-Unit Match HR P 2-7/10 (n
88) Reference 8-9/10 (n 27) 0.37 0.105

80
60
C.I. Grade III-IV aGVHD
40
2-7/10 HLA Match
20
8-9/10 HLA Match
0
0 1 2
3 4 5
6
Months Post-Transplant
P 0.07 on multivariate HLA-match likely
critically important
Ponce, D., ASBMT 2012
28
Evaluate search for units 4-6/6 gt 2.0 x
107/kg.
Review info bank for each unit. Obtain missing
info, CT units of interest. Prepare CB Search
Summary Report.
Review CTs, update Search Summary
Rank units by A,-B antigen, -DRB1 allele
match Hi to low TNC within each match grade
(correct for RBC).
1st
2nd
3rd
6/6 units Choose largest.
5/6 units Choose largest.
4/6 units Choose largest.
Make final selection of unit(s) (1a 1b
if double).
Prepare domestic back-up unit(s).
Plan shipment(s)
Ignore unit-unit match in double unit CBT
29
Require att. segment for identity testing
complete IDMs. Select on bank, dose, match,
other (RBC content).
30
What about higher resolution match?
31
COBLT Single CBT OS in Pediatric Malignancies
A, B, DRB1 allele match lt 5/6 allele
match associated with higher severe
aGVHD. Trend toward improved OS with better
match.
Kurtzberg, J. et al, Blood 2008
32
Effect of C A,B,C Antigen, DRB1 Allele
N 803, median 10 yrs (lt1 62), leukemia/ MDS

• Inferior neut engraftment with hi degree MM (lt
  5/8).
• Worse GVHD if lt 5/8 including HLA-A MM.
• Relapse lower if any MM vs match (but no
  advantage to
• multiple mismatches.
• TRM significantly worse if lt 6/8 (trend for
  7/8).
• 3 year TRM 8/8 9 7/8 (non-C) 19 7/8 (C)
  26
• 6/8 (C other) 31.
• Significance lost in overall mortality except
  for 6/8 (C
• other). Contributed to by rel. high TNC of
  group?

C is important-but how to trade off against cell
dose? What is new lower limit of acceptable match?
Eapen, M. et al, Lancet, 2011
33
New Easy to Implement
34
Incorporating Vector of HLA-Match 1202 Single
Unit CBT, NYBC
CI of Neutrophil Engraftment
Significant advantage to both 0 GVHD vector
only mismatches
Stevens C E et al. Blood 2011
35
HLA-Match Vector 1202 Single Unit CBT
CI of 3 Year TRM
In heme maligs GVH only mismatch equal to 0
mismatch.
Stevens C E et al. Blood
20111183969
36
New But More Difficult to Implement
37
NIMA-Match 1121 Single Unit CBT, NYBC
3 Year TRM in Patients gt 10 Years Old
If 1 MM, advantage if this is a NIMA match
(predom. due to better neutrophil engraftment).

van Rood J et al. PNAS 2009
38
Relapse by Shared IPA 845 Singles (AML/ALL)
1.0
1-3 HLA MM, No Shared IPA Reference 1-3 HLA MM,
Shared IPA 0.4 lt0.001 0 HLA MM 0.3
0.012
0.8
0.6
Cox Regression Multivariate
No Shared IPA (n49)
C.I. of Relapse
0.4
0 HLA Mismatch (n45)
0.2
Shared IPA (n751)
0.0
0
1
2
3
Years Post-Transplant
Patient shares IPA reduced relapse. ??Indirect
evidence that maternal T-cells mediate GVL?
39
Implications for Unit Selection

• CB banks should report maternal HLA type.
• Should
• Select for NIMA match expands no. of well
  matched units.
• Avoid No Shared IPA grafts in leukemics.

40
MSKCC Strategy for Unit Selection
1) TNC/ HLA-match Above 2.0 x 107/kg prioritize
match Within match grade choose largest. Consider
vector C. 2) Also consider bank of origin
(speed, reliability, quality). 3) For malignancy
use 2 Increase chance of transplanting at least
one unit of good quality PLUS unit vs unit
effects may augment engraftment reduce
relapse. 4) For doubles same rules apply to
selecting units 1 2. Ignore unit-unit
HLA-match. 5) Consider hi res match if
possible-esp in children. 6) Unresolved issues
selecting based on CD34 dose, red cell content,
testing of segment, high res match vs dose,
incorporation of NIMA IPA.
Barker, Blood 2011 -How I Treat