1. science
2. medicine
3. surgery
4. transplants

“Belt & Suspenders”:
A Matter of
Canadian Precautionary Pride,
or Fashion-Backwards
for Risk Mitigation of Transfusion-Transmitted CMV?
Day 1: Thursday April 11th 2013,
Session 2B: Modern Transfusion Medicine in BMT, 10:45 am – 12:15 pm
(Sub-Session III: 11:45 am – 12:15 pm)
Fort Garry Hotel, Gateway/Tache, Mezzanine Level
Christine Cserti-Gazdewich, MD FRCPC
Assistant Professor, University of Toronto
Laboratory Hematology (Pathology) & Clinical Hematology (Medicine)
Blood Transfusion Laboratory, University Health Network
Toronto, Ontario, Canada
[email protected]
Conflicts –
Princess
Margaret
Hospital
University Health Network
Sunnybrook
Hospital
Holland
Ortho &
Arthritic
Centre
Women’s
College
Hospital
Sunnybrook Health Sciences Centre
UHN LMP Northern Network
SU(H)N-Network
Toronto
Western
Hospital
Toronto
General
Hospital
St Michael’s Hospital
Hospital for Sick Children
Nil Commercial
Canadian Blood
Services
national blood provider
PMH Cell Therapy
Program /
Philip S Orsino Facility
local cell therapy dispensary
Tanis Steward
polling
encouragement &
support
Trillium Gift of
Life Network
provincial organ network
Ontario Regional
Blood Coordinators
session invitation
Mount Sinai Hospital
TRANSFUSION MEDICINE COLLABORATIVE
University of Toronto teaching hospitals:
Hospital Transfusion Services
provincial blood management
Ontario
Transfusion
Coordinators
provincial blood conservation
Goal Sequence
1. Canadian Consensus Conference Pre-History
on TT-CMV: The Case for Conservative
Action (or the “Belt-and-Suspenders” PostKrever Precautionary Principle on Ultra-RiskMitigated Products for Most Vulnerable BMT
Patients)
2. Perceptions & Practices… and a Pre-Meeting
Poll
3. New Scientific Evidence from the Last (½)
Decade
4. Shifts in the Current Views & Future
Inventory
Special Attributes  Product-Specific!
All blood in Canada is now CMV-reduced-risk via
Pre-Storage LeukoReduction (PSLR) since 8/1999
ABO
RhD
+/-
XRT Special
Antigen
Type
RBC
Plt
pRBC
FP
CMV wash
IgGor
(SN) PVR
HLA since
’80’s
or
HPA
IgA-
S/DT
PI
(B2 +
UVB)
CMV
• cyto-megalo-virus, aka HHV-5
• 100 nm virion
• 120 – 230 kb ds DNA
• community seroconversion rates:
– HCWs: 0.6 – 3.3% / y
– between pregnancies: 2-6%
– adolescents: 13% / y
• 50 – 80% of population is
seropositive (SP)
• 40% of population is
seronegative (SN)
Definitions & Processes at CBS Since 2007
Donor (1st timers or SN on last donation)
CMV IgG/IgM testing
53% of CBS donors
not seronegative (SP)
• not labelled as such
• never re-tested on future donations (U)
• assumed to be a latently immune
seronegative (SN)
• labelled as such
• re-tested on next donation
40% of donor samples undergo testing with PK® CMV-PA 7300 ($3/test) at $810K/y
(favour B2 WB [destined for pRBC / BCP / FP] & SDAP)
Will CMV-SN Be With Us Forever?
Word on the Street, 2012…
CSA Standard Z902-04
Blood & Blood
Components
Blood Advisory Committee (OBAC)
PSLR in Canada
• since 1999, universal pre-storage leukodepletion
= 3-4 log reduction in white blood cells (WBC)
109 (bill)  106 to 105 (<mill)
– benefits:
• ↓ cytomegalovirus (CMV) transmission
• ↓ immune disturbance to host
– ↓ FNHTR
– ↓ HLA sensitization
– ↓ TRIM  ↓ TA-GVHD
8
CMV Compartmentalization
• peripheral blood leukocytes (PBL) & stem cells (HPC) =
target cells of infection (latency to replication)
• among healthy SP donors:
–
–
–
–
0.01 – 0.001 % of myeloid progenitors
0.01 – 0.12% of PBMCs
2 – 13 viral genomes per infected cell
1 – 25 monocytes for every 106 PB WBCs
(5% [50,000] of which are PBMCs, <<<1% of which are affected)
• monocyte differentiation  latent infectivity to active replication
• free plasma virus: unstable & transient (< 4mo);
? independent pathogenicity ?
CMV “Antigen-emia”,
or CMV DNA levels,
aka viral load
• correlates with severity of viral disease
(and/or level of immune compromise, all arms of which
count)
– virtually “un-culture-able” from healthy donors
(2 in every 1500 buffy coat samples)
– 105.1 genomes/mL = 50% CMV disease risk
– 105.5 genomes/mL = 90% CMV disease risk
– 108 genomes/mL = typical for immune compromised patients
CMV Host Immune Response
preserologic
(isolated
cell-free
virus)
risk? st
IgM alone: 8%
transmission risk
+ anti-CMV CTL
1 mo: 60% >1010/105 WBC
2nd mo: <3.3% >101/105 WBC
plasma
CMV
DNA
3 mo
2 mo
WBC
CMV
DNA
window phase: 1-4 mo
IgG alone: 0.3%
transmission risk
CMV Infection: Subclinical
Seroconversion to Severe Disease
• asymptomatic
• cytopenias
• tissue-invasive disease
– CMV hepatitis
– CMV retinitis
– CMV interstitial pneumonitis
– CMV encephalitis
– CMV gastroenteritis, esophagitis
disease severity
infection
• fever, splenomegaly, atypical lymphs, liver enzymopathies
• for primary disease, within 3 – 8 weeks of inoculation
Who Is At Risk?
ie- who needs CMV-reduced risk blood?
•
A: harm proven; benefits of CMV-avoidance proven:
– LBW infants (<1500g) of SN mothers
– SN recipients of SN BMTs
– SN recipients of SN solid organ transplants (except renal & cardiac)
•
B: harm (lesser degree) proven; benefits of CMV-avoidance not yet proven
but assumed reasonable:
–
–
–
–
–
–
–
•
C: harm low but not negligible; benefits of CMV-avoidance assumed:
–
–
–
–
–
•
SN pregnancy needing maternal or intrauterine transfusion
SP pregnancy needing intrauterine transfusion
LBW infants or SN immunosuppressed patients needing granulocyte transfusion
SN HIV+ patients
children of HIV+ mothers
LBW infants of SP mothers
SN patients before BMT
N birthweight babies of SN mothers
heavily transfused or ECMO infants
SN recipients of SN kidneys or hearts
SN patients on chemotherapy
SN patients after major trauma or splenectomy
D: no risk
– all birthweight babies of SP mothers
– all others
CMV Disease Presentations vs
Significant TT-CMV
SN
SP
IC
ID
IC
ID
CMV:
Immune Competence Context
SP
• NOT an issue in
“already SP”
recipients
– suffer (less severe)
non-primary postlatency infections
(reactivations) rather
than new 2nd-strain
infections
SN
• in the
immunocompetent
“still SN” recipients,
TT-CMV is no more
(actually less!) frequent
or significant than
community-acquired
CMV
– <1.2% transfusion risk vs
1.7%/patient year
– even at transmission
rates of 30%, CMV
disease is rare
SP
CMV:
Immune Impairment Context
SN
ID
ID
• TT-CMV IS clinically important for the immune-suppressed SN
recipient
– 13 – 37% risk with unmodified transfusions
• for infants:
– 13.5% if <50cc
– 24% if >50cc
» 50% risk of serious disease for the <1.2kg (VLBW)
• for R-SN/D-SN HPCT:
– 20 – 60% risk
» 1/3rd of these develop pneumonitis, with CFR 50%
• for fetuses of SN gravids,
– 35 – 50% risk
» 5-15% of these have adverse outcomes
• “reduced risk” products
– risk falls to 1-4% in R-/D- HPCT with LR or SN transfusion restriction
If at risk, what can you offer?
“CMV safe blood”
= any type of plasma,
or
cellular products with either of the following
modifications:
– seronegative (SN)
OR
– leukoreduced (LR)
OR
– frozen-deglycerolized
American
Association of
Blood
Banks
1997 Bulletin
#97-2 
Standards
• bedside LR vs SN RCT of 502 SN BMT patients
• 1º endpoint of CMV infection or disease for LR vs SN = NS:
• 2.4 vs 1.3 % (P=1.0) for infection
• 2.4 vs 0 % (P=1.0) for disease
Not fair;
• LR is now pre-storage, which is more effective
• there were several “filtration-miss” errors in the study
For day 21-100 post-BMT analysis:
• LR: 2 deaths / 3 infections (all disease-manifesting)
• SN: 0 deaths / 2 infections (non-disease-manifesting)
2/252 SN developed infection+disease 6mo post
But what is the safest measure, for
those at highest risk?
• The combination of
SN AND LR
(aka “belt and suspenders”)
is technically also an option….
Canada: Conservative
“Belt and Suspenders” Fashion
• groups of interest:
7 (LR + SN) : 2 (fetal risk mitigation) : 1 (LR enough)
– both LR + SN blood for SN pregnant
women, intrauterine fetal transfusions,
and R-SN/D-SN allogeneic marrow
transplant recipients
– SN blood may be indicated for RSN/D-SN solid organ transplantatees,
in SN patients with conditions likely to
require allogeneic HPC
transplantation, and SN HIV+ patients
• CBS agrees to provide this
inventory
1. Blajchman MA, Goldman M, Freedman JJ, Sher GD. Proceedings of a consensus conference: prevention of posttransfusion CMV
in the era of universal leukoreduction. Transfus Med Rev 2001;15:1-20.
2. Laupacis A, Brown J, Costello B, et al. Prevention of posttransfusion CMV in the era of universal WBC reduction: a consensus
statement. Transfusion 2001; 41:560-569.
• prospective cohort (n=807 SN SCT patients) over 2 periods
• SN and/or LR (’94-’96): 1.7%
• LR alone (’96-’00: more CMV+ donors): 4%
• LR pRBC exerted 32% greater risk for TT-CMV, but
• LR SDAP were not inferior
• ganciclovir for CMV antigenemia prevented disease
LR Alone
• benefits: (ie- arguments to not bother with SN status)
– not all SP blood is infective
– lower viremia in SPs than window period SNs
– SP donors convey potentially neutralizing anti-CMV
passive immunity
• drawbacks: (ie- arguments to demand SN status)
– LR not 100% effective, as residual latently infected WBC
or plasma virions (shed during asymptomatic
reactivations) may be conveyed
– 1.2 – 2.4% rate of transmission with possibly worsened
clinical sequelae in breakthrough transmissions
– fears of cell-free CMV release from filtration WBC lysis
1.
2.
3.
4.
5.
Lipson SM et al. CMV infectivity in whole blood following LR by filtration. Am J Clin Pathol 2001; 116: 52.
Dumont LJ et al. The effect of LR method on the amount of human CMV in blood products: a comparison of apheresis and filtration methods. Blood
2001; 97: 3640-7.
Bowden RA et al. Use of leukocyte-depleted platelets and cytomegalovirus-seronegative red blood cells for prevention of primary cytomegalovirus
infection after marrow transplant. Blood 1991; 78:246-50.
Drew WL et al. Frequency and duration of plasma CMV viremia in seroconverting blood donors and recipients. Transfusion 2003; 43: 309-313.
Nichols WG et al. TT-CMV infection after receipt of leukoreduced blood products. Blood 2003; 101: 4195-4200.
• 829 recipients of SN products (11 studies) vs
878 recipients of LR products (12 studies)
• In BMT, SN vs LR TT-CMV risk
1.63% (11/674) vs 3.01% (21/679)
• 1 of 3 studies suggested superiority of SN over LR
– 58% RR (summary OR 0.42, 95% CI 0.22 – 0.79, P<0.05)
What About Giving PSLR a Chance?
• 4/04 – 12/06, prospective recipient & donor sample study to
ascertain TT-CMV incidence in PSLR, CMV-U products (now
that GMP-standard technology is 100X more effective than Bowden’s materials)
• 120 recruited  107 testable samples14-180d post-exposure
(24 with SCT)
 46 CMV SN recipients (vs 57% being SP)
– 37% of donors to these patients found to be SP
– 19 cases bore CMV marker dynamism for investigation
– 3 true seroconversions identified (albeit “possible” donor
attributability), remainder passive or indeterminate
= 6.5% (95% CI 1 – 18%) of recipients or
0.23% (95% CI 0.06 – 0.62%) of products
CMV Seronegativity As A Surrogate of
“Cleanliness” – Elephant in the Room
IF CMV SP rates are directly associated
with:
– age
– sexual activity (nil vs hetero vs MSM)
THEN
is CMV seronegativity not a surrogate of
those with the epidemiologically “least
promiscuous” microbiomes &
microviromes?
In the era of pre-storage leukoreduction,
are CMV-seronegative blood products still
necessary in HSCT?
Does your centre still use CMV-negative products?
If so, for what indications? [CBMTG Conference Q&A]
Respondents
N = 16
+5
anonymous
respondents
7 days of polling
before CBMTG
Edmonton: 1
Saskatoon: 1
Winnipeg: 2
Hamilton: 1
Toronto: 3
Ottawa: 2
Montréal: 1
1
3
7
•(no response)
5
2
1
13
1
•(no response) 2
•sometimes, but generally if we have trouble with
blood products it is an availability of the product,
or the space to give the product more than blood
transfusion services themselves
13
4
11
1
•(no response) 4
1
11
•Do not know
•Not sure
•SOPs are in place, & decisions are based on that and Blood Bank
would/could not act alone
•(no response)
0
4
2
7
3
•FACT requires frequent formal interactions of all
sorts between BMT and BB, so monthly
•They attend our quality assurance meeting monthly
3
4
2
7
3
11
2
Ottawa, Edmonton, (London)
0
•(no response) 3
•No, but exception might be neonates Ottawa
Hamilton
•HaploBMT only
2
Non-Toronto
Ontario
11
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
Pre-BMT
?ever
?auto
?allo
Patient/Recipient
CMV+
unknown
CMV-
Allo:
AutoSCT
Transplantation Exposure Type
CMV+
Donor
CMV-
1
everyone
2
2
0
4
•(no response)
2
4
no one
Post (-/U)R (-/U)D only, in 1
Pre (-/U)R , or Post (-/U)R (-/U)D, in 3
Generally (still) complying with Canadian Consensus
Conference recommendations, some more or less.
•(no response) x 6
•Not aware of the
research to make an
informed decision
•(no response) x 4
3
3
1
3
2
7
pro
7
1
con
3
2
8
1
0
0
7
7
•Need evidence x 2
•BMT program would decide, and did decide years ago
8
1
2
1
•(no response)
•CMV- is not provided; not a consideration
10
2
1
1
2
1
10
2
2
1
•(no response)
•Do not know enough about the
platelet antibodies’ significance to
decide
13
0
1
2
13
HLA Similarity (DR) α Post-Transplant CMV
Mañez et al, 1993, 1067
Donaldson et al, 1987, 945
Waltzer et al, 1987, 4077
May et al, 1978, 110
In Exotic Realm of BMT:
“2 negs buys you both measures”
Clinical scenario:
CMV IgGrecipient
CMV IgGdonor
R- / D“neg neg”
Dispensary rule:
PSLR + IgG CMVLR / SN
“belt and suspenders”
• easier to assume everyone is vulnerable / not test / not await results
• easier to not stratify dispensary rules or invigilate eligibility & compliance
• tempting to assume that a “best” product is generally preferred for all
Fairness (Safety & Cost-Effectiveness)
• until PI/PR technologies are available and precautionary
principle is invoked (or the SN inventory gets taken away!),
screen recipients to establish SN status before committing to
SN products
– cuts down on orders for SN blood
• eg –
– 40% (not 100%!) of gestations, HIV, and pre-transplant patients
– (0.4)(0.4)= 16% (not 100%!) of alloBMT
• surveys still show variability in perceptions & practices, re:
– blanket provisions of SN+LR products to at-risk patient
categories despite serostatus
– SN+LR products to those with unknown serostatus
(incomplete qualification) and/OR verified vulnerability
– abandoning SN inventory/matching efforts altogether
1.
2.
International forum. Prevention of post-transfusion cytomegalovirus: leucoreduction or screening? Vox Sanguinis 2002; 83: 7287.
Smith et al. Survey of current practice for prevention of transfusion-transmitted CMV in the US: LR vs SN. Vox Sanguinis 2010;
98:29-36.
Research Hypotheses & Strategies
Needed to Overturn LR+SN Dogma
• Show that what we think
about SN may be wrong,
and that it’s not as good
as we think it is…
• Show that LR alone is not
disadvantageous (not
incrementally [more]
harmful)
•
CMV DNA detectability in
seroconverting donors
•
2000 – 2004: 82
seroconversions / 12,800
donations (0.8%/y)
seroconverters had
highest CMV DNA rates
(including 3% of the last
recovered SN samples)
•
of 598 donors SP for
> 1y: 0 % DNA+
•
CMV DNA prevalence due
to 1° infection in donors:
0.13%
More conservative, evidence-based,
biologically-intuitive expert consensus
points out the following to temper any
haste:
• CMV biology is too complex to be modelled so simply
• even though the window-period SN or seroconverting donor
is the most appealing scapegoat, this donor type (ie- cellfree virus) still hasn’t ever been conclusively implicated, on
a molecular-proof basis, in TT-CMV
risk of SN donors being FP or
seroconverting?
risk of LR failures for the latently SP?
• prospective cohort (n=23 SN allo-SCT patients, ’99-’09)
received LR products alone (1847 from 3180 donors)
• 0% developed CMV DNA detectability or IgM
• 17/23 developed CMV IgG (passive)
• mid’06 policy∆ to LR-only after nil TT-CMV from 14/27
CMVneg/neg getting non-SN products
• 18mo pre/post (1/05-12/07) 100 CMVneg/neg : 6465 products
• pre: 5 exclusively SN recipients; only 11-15% of inventory SN
• 2 IgG “ transient seroconverts” (IgM & NAT-) in SN period, 0 in U
= 2%/pt, 0.3%/product, 0% confirmed TT-CMV
What do these other alloBMT centres
have in common?
The London Health Sciences Centre
(Cyrus Hsia and Ian Chi-Yee: 2012)
The Ottawa Hospital
(2006)
Hospitals of the
Harvard Joint
Program in
Transfusion
Medicine: DFCI,
BWH, MGH,
BIDMC, CHB
Fred Hutchinson
Cancer Center
& Puget Sound
Blood Centers,
Seattle,
Wash U
Finland, Sweden, Germany
We’re Not the Only Ones Still Asking…
64 days ago…
No responses yet.
•Are we pessimistic about
capacity of evidence-thus-far to
arbitrate?
•Do we trust our surveillance &
pre-emptive treatment
successes more than ever?
•Are we waiting for other
advances?
•Adoptive immunotherapies
•CMV DNA PCR
•Pathogen inactivation
•External forces
Concluding Thoughts
• Post-Krever hyper-precautionary era (1997)
vs Post-Great Recession (2007) hyper-cost-conscious era
• Canadian Consensus Conference (2000)  13y later,
intercurrent evidence suggests:
– SN not all that great (window period donor risk)
– LR-SP/U not all that bad (>8000 products to >100 neg/neg BMT)
– slippery slope for prodigal use, perceptions & practices
• Let’s vote again:
In the PSLR era, for the most vulnerable R-/DBMTs, do you consider CMV-SN cellular
components to still be
NECESSARY? nevertheless PREFERABLE?