Changing Paradigms in Matched Platelet Support

1
Changing Paradigms inMatched Platelet Support

• Ralph R. Vassallo, MD, FACP
• Chief Medical Officer,
• ARCBS Heritage Division,
• Adjunct Associate Professor of Medicine,
• University of Pennsylvania School of Medicine

2
Trends in Platelet Utilization

• Annually, 1 tfxn. for every 168 people in the US
  3-4 doses per episode in 2001
• Annual risk of platelet tfxn. in the US

Cobain TJ, Vamvakas EC, Wells A, Titlestad K. A
survey of demographics of blood use. Transfus
Med 2007171-15.

• 52-74 of US transfusions are prophylactic

1.39M doses of SDPs 1.54M RDPs transfused
(50 - 6, 23 - 5)
Brecher ME. The platelet prophylactic
transfusion trigger when expectations meet
reality. Transfusion 200747188-91.
3
Trends in Platelet Utiliz. (contd)

• Canadian experience with transition to universal
  leukoreduction
• Alloimmunization was significantly reduced
• 19 to 7 (p lt.001)
• Alloimmune platelet refractoriness was similarly
  reduced
• 14 to 4 (p lt.001)
• Annual utilization growth rate from NBDRC data is
  6

Seftel MD, Growe GH, Petraszko T, et al.
Universal prestorage leukoreduction in Canada
decreases platelet alloimmunization and
refractoriness. Blood 2004103333-9.
4
Consequences of Failed Plt. Tfxn.

• Unlike red cell recipients, platelet recipients
  have to earn an immunologic workup by
  experiencing two or more consecutive therapeutic
  failures
• Presumably since stakes for clinical failures are
  lower
• The real economics of failure
• Expensive intervention without any therapeutic
  efficacy
• Additional donor exposures and chance for
  reactions or further alloimmunization
• With active bleeding, fewer platelets delivered
  to bleeding sites continued hemorrhage
• For prophylaxis, patient remains at risk for
  bleeding

5
Bleeding without Prophylaxis
Total days at risk of bleeding during study for
each of 8 different levels of platelet counts
1956-9 92 leukemics ASA used to treat fever
I All bleeding II Skin nosebleeds censored
out (incl. oropharyngeal, vaginal, hematomas,
invasive sites) III Gross hemorrhage
(hematuria, hematemesis, hemopytsis,
hematochezia/melena)
Gaydos LA, Freireich EJ, Mantel N. The
quantitative relation between platelet count and
hemorrhage in patients with acute leukemia. NEJM
1962266905-9.
6
Platelet Refractoriness

• Inappropriately low platelet count increments
  after 2 or more consecutive transfusions
• Appropriateness of increment dependent upon
  product content, blood vol. splenic pooling
• PPR 30 1hr, 20 after 18-24hr ()
• CCI 7,500 1hr, 5,000 after 18-24hr
  (m2/µL/1011plts.)
• Predictive value dependent upon definition
• Likelihood of next failure after 2 - 4
  consecutive failures 2 - 46 3 - 68
  4 - 75 (CCI lt7,000 _at_ 1o)
• In reality, this is related to the breadth of
  recipients alloimmunization
• Likelihood of alloimmunization related to
  underlying disease prior exposure through tfxn.
  or pregnancy

7
Platelet Refractoriness (contd)

• Up to 90 of unsuccessful transfusions occur in
  setting of one or more nonimmune causes of
  platelet loss, with or without concurrent immune
  causes
• Splenomegaly, fever/infection, DIC, s/p BMT/PSCT,
  SOS/VOD, amphotericin B use, bleeding
• Of immune etiologies, only alloimmunization
  addressable through matched products others are
• Drug-induced Ab autoAb circulating immune
  complexes (e.g., plasma protein Ab, HIV Ag-Ab)
• Makes sense to screen for alloimmune
  refractoriness to avoid product failure
• ABO-mismatched transfusion can result in
  suboptimal increments (major 0-90 decrement,
  minor 0-18) also see higher likelihood of /
  earlier alloimmunization

8
Alloimmunization

• Alloimmunization against Class I Human Leukocyte
  Antigen (HLA) locus A and B proteins, with
  occasional Human Platelet Antigen (HPA)
  co-immunization (6-8), or rarely (lt1-2),
  HPA-only alloimmunization
• Ab can form w/in 4-10d (usually w/in 2 - 4 wks.)
• Majority of Ab directed at HLA public private
  (70-90) vs. private epitopes alone
• 10 - 15 pts. broaden Ab specificity over time,
  on the upswing of Ab development or throughout
  Rx.
• 30 decr. Ab reactivity (cessation of Ag
  stimuli and/or anti-id Ab) transiently or
  permanently regain responsiveness
• PRA (panel or percent reactive antibody)
  describes the breadth of alloimmunization (what
  of donors are expected to be incompatible)

9
Matched Platelet Strategies

• Unit selection based upon
• Knowledge of HLA type
• Search of unrelated donors for identical HLA-A
  B type
• gt35 have no perfect match in most donor bases
  gt80 no match in inventory other matches are
  best guess
• Crossmatching
• React patient serum with platelets from
  inventoried units
• Few antibodies, few units screened many
  antibodies, many screened (, test TAT unit
  availability issues)
• Knowledge of HLA antibodies HLA type
• Provide units lacking specified HLA determinants
• 1-2 orders of magnitude more donor / product
  matches

30-75 successes (55)
50-75 successes (60)
60-75 successes (70)
10
HLA Serological CREGs
HLA-B
HLA-A
11
HLA-Based Selection

• Quality of the HLA-A -B locus match
  determines outcome RECIPIENT A1 A2 B7 B8
• A HLA identical all 4 loci A1 A2 B7 B8
• BU All loci identical, only 3 Ags detected A1
  B7 B8 B2U All loci identical, only 2 Ags
  detected A1 B8
• BX 3 loci identical, 4th cross-reactive A1 A11
  B7 B8 BUX 3 Ags, 2 identical, 3rd X-reactive
  A1 A11 B8
• B2X 2 loci identical, 2 X-reactive A1 A11 B7
  B60
• C one locus mismatched A1 A32 B7 B8
• D two or more loci mismatched A1 A32 B7 B60

12
HLA-Based Selection (contd)

• Statistically poor Intra-CREG Matches BX
• B5 vs. B15, B17 5C
• A1, A11 vs. A3 1C (bidirectional)
• Statistically poor Out-of-CREG Matches C
• 8C (B8, B14) vs. 12C (B12, B21) (bidirectional)
• 5C (B17, B21) vs. 7C (all)
• 7C (B27) vs. 5C (all)

13
Computer-Assisted HLA Selection

• Each HLA antigen represents an array of
  immunogenic 3 amino acid epitopes (eplets)
• Patients cannot produce alloantibodies against
  epitopes on their own HLA-A, -B -C molecules
• Any donor without mismatched eplets is
  potentially fully compatible (A/BU-like)
• HLA type ( Ab screen results) can further
  identify antigens with the fewest mismatched
  eplets (i.e., BX/C matches more likely to succeed)

Nambiar A, Duquesnoy RJ, Adams S, et al.
HLAMatchmaker-driven analysis of responses to
HLA-typed platelet transfusions in alloimmunized
thrombocytopenic patients. Blood
20061071680-7.
14
Platelet Crossmatching

• Immucor Capture-P Solid Phase Red Cell Adherence
  (SPRCA) assay most freq. used
• Detects HLA, HPA and ABO Abs
• 2/3 of Os have anti-A or -B mediated pos. tests
• General correlation with the PRA, but misses some
  significant IgG all IgM Ab
• Up to 17 of sig. AHG-LCT pos. may be SPRCA neg.
• Only 17 of 2800 XMs were compatible in one study

Petz LD. The HLA system and platelet
transfusion. In Murphy S, ed. The HLA System
Basic Biology and Clinical Applications,
Bethesda MD AABB Press, pp. 133-175, 1999.
15
Solid Phase Red Cell Adherence
Platelet Monolayer
Positive
Negative
Anti-Platelet Antibody
16
Antigen Negative Approach

• Antibody Specificity Prediction (1621 tfxns. in
  114 pts.)
• Post-tfxn. PPR ASP - 24.13 XM - 23.38 HLA -
  20.77 Rnd - 14.87
• PPRs adjusted for ABO, post-count timing, PRA,
  fever ampho B use
• Markedly expands of potential donors
• (n29 recipients vs. 7247 donors)

A matches BU matches ASP matches Range 0 - 60 0
- 117 11 - 4638 Median/Mean 1 / 6 20 / 33 1365 /
1426
Petz LD, Garratty G, Calhoun L, et al. Selecting
donors of platelets for refractory patients on
the basis of HLA antibody specificity.
Transfusion 2000401446-56.
17
Antigen Negative Approach (contd)

• Some HLA antigens w/ low platelet expression may
  yield good increments despite the presence of
  specific Ab
• B8 B12 (44,45)
• Possibly B13 decr. Bw4 expression B14
  (64,65) decr. Bw6 expression

18
XM 66 (A/BU vs. BX 74 vs. 62) XM
35 (ABO identical vs. nonidentical 40 vs.
13)
A/BU/BX 60
XM 51 (ABO identical vs. nonidentical
63 vs. 36) XM 30 (ABO identical vs.
nonidentical 40 vs. 26)
C 45
A/BU 74 BX 62 C 51
XM 57
A/BU/BX 35 C 30
XM 33
Heal JM, Blumberg N, Masel D. An Evaluation of
Crossmatching, HLA, and ABO Matching for Platelet
Transfusions to Refractory Patients. Blood
19877023-30.
19
Matching Hierarchy

• HLA-identical platelet selection (A/BU
  matches when available selective recruitment for
  pts. w/ high PRAs)
• Ag-neg. crossmatched (XM) products (dubious
  Ab ID results or suspected HPA)
• Ag-negative units
• Platelet crossmatches
• HLA type-driven BX C matches
• ABO Identical RDP (or SDP) when HLA Ag/Ab w/u
  is pending crossmatches unavailable

20
Why Not Just Crossmatch?

• Crossmatch availability issues
• Limited or no night weekend availability
• High PRA pts. unlikely to have many compatible
  units among subset of inventory tested
• 4 hour test TAT (vs. immediate release when Ab
  ID / HLA typing results are used to choose units)
• Ties up distribution of 10-20 units during
  testing
• The few compatible units may be released during
  testing to meet patient needs in other hospitals
  ? support failure!
• Results may be misleading
• ABO Ab may prevent ID of truly compatible unit
• Some clinically sig. HLA antibodies missed
  (15-20)
• Less expensive only for very brief support

21
Unexplained MatchedProduct Failures

• Broadening of HLA antibody repertoire
• Repeat antibody screen
• Development of an HPA antibody
• Perform HPA screen and/or ID procedure
• i.e., crossmatch panel against group O platelets
  for low PRA pts. or HPA ELISA (e.g., GTI
  PakPlus)
• Clinical deterioration
• Review record for new / recrudescent causes of
  nonimmune or immune refractoriness

22
Pearls for Managing Plt. Refractory Patients

• 1hr CCIs or PPRs very important in high PRA pts.
• Identifies good donor-recipient matches
• Clue to broadening of alloimmunity in sick pts.
• Establishment of a transfusion schedule is
  critical for recruitment of A/BU matches
• Does the pt. really need CMV neg. units or will
  LR do? (prophylaxis vs. bleeding consider PRA!)
• HLA-selected / crossmatched products should be
  irradiated to avoid TA-GvHD
• Matching is usually not helpful for pts. w/o
  demonstrable HLA (or HPA) antibodies
• Consider brief support if IgM HLA Ab suspected

23
Pearls for Managing Plt. Refractory Patients

• Persistently unresponsive, highly refractory pts.
  w/o available A/BU donors and no increments from
  prophylactic transfusions
• Consider high-dose RDPs for bleeding only (and
  rVIIa)
• Prophylactic use of Amicar
• Consider heroic therapy aimed at suppression of
  alloimmunity
• IVIg
• Protein A Column / Plasma Exchange
• Massive Platelet Transfusion
• Immunosuppression e.g., CyA (? role of
  rituximab)
• VCR / VLB

24
Heritage Division Stats

• 8,000 9,000 HLA-typed donors (potential
  gt13,000)
• 90 PlA1-negative donors
• gt95,000 apheresis products collected annually
• HLA-selection charge equals XM panel charge
• XM charge is for panel of 10 higher PRAs require
  more XMs and a correspondingly greater charge
• Initial 500-600 investment for HLA Ab ID and
  typing (only 100 if Ab screen is negative) paid
  in full if failure of a single SDP is avoided

25
One Regions Stats

• Platelet refractory individuals CY 2006 audit
• We support 160 platelet-refractory patients
  annually
• 19 of products were A/BU matches
• 90 of products were chosen to avoid known
  recipient HPA and/or HLA antibodies
• 20 of requests are urgent
• 81 A/BU Ag-neg success rate (n154 1o CCI
  gt7,500)
• 0/2 XM successful
• 1/7 BX/C/D successful (14)
• Average of 12 HLA-selected units per patient

26
PlA1 (HPA-1a) Negative Units

• Platelet antigen in 98 of donors
• 2 (PlA2 homozygotes) can become alloimmunized
  (assoc. with HLA-DRB30101 DR52a RR 24.9 and
  HLA DQB10201 RR 39.7)
• PlA1 antibodies responsible for
• 5-10 of HPA-mediated platelet refractoriness
• 70-85 of cases of post-transfusion purpura
• 75-80 of neonatal alloimmune thrombocytopenia,
  which affects 1800-1000 newborns
• We support 15 PlA1-negative patients per year
• Average of 3 PlA1-negative products (NAIT and
  PTP) per patient