Prion
infections, blood and transfusions
Adriano Aguzzi* and Markus Glatzel
Prion infections lead to invariably fatal diseases of the CNS, including
Creutzfeldt–Jakob disease (CJD) in humans, bovine spongiform
encephalopathy (BSE), and scrapie in sheep. There have been hundreds
of instances in which prions have been transmitted iatrogenically among
humans, usually through neurosurgical procedures or administration of
pituitary tissue extracts. Prions have not generally been regarded as
bloodborne
infectious agents, and case–control studies have failed to identify
CJD in transfusion recipients. Previous understanding was, however,
questioned by reports of prion infections in three recipients of blood
donated by individuals who subsequently developed variant CJD. On
reflection, hematogenic prion transmission does not come as a surprise, as
involvement of extracerebral compartments such as lymphoid organs and
skeletal muscle is common in most prion infections, and prions have been
recovered from the blood of rodents and sheep. Novel diagnostic strategies,
which might include the use of surrogate markers of prion infection, along
with prion removal strategies, might help to control the risk of iatrogenic
prion spread through blood transfusions. ...
snip...
INTRODUCTION
Prion diseases, also termed transmissible
SPONGIFORM ENCEPHALOPATHIES, constitute
a group of neurodegenerative conditions that
are transmissible within and between mammalian
species. A characteristic of these diseases is
the accumulation of a misfolded prion protein,
PrPSc, which is a post-translationally modified
form of the host-encoded prion protein (PrPC).
The processes underlying PrPSc formation
remain enigmatic, but there is little doubt that
a conformer of PrPC, which might exist in an
oligomeric form,1 is identical to the infectious
entity.2 Prions damage the brain by transmitting
toxic signals to cells expressing PrPC.3
Although genetic evidence has been taken
to indicate that human prion diseases have
been with us since prehistoric times,4 the first
documented cases of Creutzfeldt–Jakob disease
(CJD) date back only 85 years.5–7 Since then, it
has become obvious that human prion diseases
have three distinct etiologies: they can arise in the
absence of any documented exposure to infectious
prions as sporadic CJD (sCJD), as an autosomal
dominantly inherited disease in the case
of genetic, or familial, CJD (gCJD/fCJD), or as
an acquired condition in the case of IATROGENIC
and variant CJD (iCJD, vCJD), or kuru, which
resulted from cannibalism.8
Some prion diseases that occur in animals
might have been recognized several centuries
ago, as suggested by early descriptions of sheep
diseases that seem to correspond to scrapie.
Most prion diseases affecting animals, however,
were discovered relatively recently.6 A transmissible
spongiform encephalopathy affecting
cattle (bovine spongiform encephalopathy,
or BSE) has caused a massive epidemic in
European countries, affecting around 2 million
animals.9 Epidemiological, biochemical, neuropathological
and transmission studies have
substantiated the concern that BSE prions might
have crossed the species barrier between cattle
and humans, resulting in a novel form of human
prion disease, vCJD.10–13 During 1996–2001, the
incidence of vCJD in the UK rose year upon year,
evoking fears of a large upcoming epidemic.
Since 2001, however, the incidence of vCJD in
the UK appears to have been stabilizing, indicating
that the extent of the epidemic might be
limited.14 As might be expected for in frequent
stochastic events, the numbers of new cases of
vCJD fluctuate from year to year. For example,
data available on the web page of the National
CJD Surveillance Unit15 show that the number
of onsets of vCJD was higher in 2004 than it was
in 2003, but this is not necessarily indicative of
an upward trend.
It must be assumed that a number of asymptomatic
carriers of vCJD exist in human populations
that have been exposed to BSE. The
existence of such a chronic carrier state is a
logical and unavoidable consequence of the
long incubation time of prion diseases, which
is typically in the order of several years and—
in the case of oral exposure to prions—can
reach several decades. Consequently, anybody
who has contracted the infection but has not
developed clinical signs and symptoms might
be consider ed a carrier. Some of these carriers
are likely be ‘preclinical’, and will proceed,
in due course, to the development of disease.
Alternatively, it is conceivable that the carrier
state can persist for an indefinite period of
time, in which case infected individuals could
be regarded as ‘permanent asymptomatic
(sub clinical) carriers’. Studies performed in
rodents indicate that the permanent subclinical
carrier state might be a common phenomenon,
such as occurs when immune deficient mice
are exposed to prions.16 Unlinked anonymous
screens for hallmarks of prion infection in
archival tissues have suggested that the prevalence
of individuals with sub clinical vCJD might
be higher than previously antici pated, and could
reach 237 cases per million individuals.17
The recent discovery of transmission of vCJD
via blood in three individuals indicates with
near certainty that blood-borne prion transmission,
in conjunction with an unknown
prevalence of vCJD-infected carriers, leads
to secondary transmission of host-adapted
prions.18 Consequently, the vCJD epidemic
might be prolonged, or, in the worst-case
scenario, vCJD be rendered endemic and selfsustained.
In this article, we review how prions
could act as blood-borne infectious agents, and
consider strategies aimed at minimizing the risk
of secondary trans mission of prion diseases.
TRANSMISSION OF PRION DISEASES
IN HUMANS
The cause of most human prion diseases is
unknown. In the case of sCJD, the term ‘sporadic’
is used as a euphemism, meaning that we have
no idea about the origin of this form of CJD. By
contrast, gCJD always segregates within families
with mutations in the gene encoding the prion
protein (PRNP), suggesting that these mutations
are causally involved in disease pathogenesis. As
no families have been described in which gCJD
segregates with mutations in genes other than
PRNP, it has been difficult to use human genetics
to understand the pathogenesis of prion diseases.
The discovery of PRNP mutations in gCJD has
led to the proposal that at least some cases of
sCJD might be due to somatic PRNP mutations
analogous to those found in the germline of
gCJD patients. It is equally possible, however,
that some of the cases of alleged sCJD derive
from hitherto unrecognized infectious causes.
In apparent agreement with the ‘intrinsic’
origin of sCJD, which accounts for more than
90% of all human prion diseases, epidemiological
studies were not able to identify a
conclusive link between this form of CJD and
external risk factors.19 This fact is reflected in
the pathological and biochemical features of
these diseases. Although low levels of PrPSc and
prion infectivity can be demonstrated in peripheral
sites such as lymphoid organs or skeletal
muscle,20,21 the highest levels of PrPSc and prion
infectivity appear to occur in the CNS. These
facts might account, at least in part, for the lack
of evidence of sCJD transmission by labile or
stable blood products.22 Indeed, several retrospective
studies have failed to identify blood
transfusion or exposure to plasma products as
risk factors for the development of sCJD,19 and
prion diseases appear to be exceedingly rare
in hemophiliacs, a group of patients that is at
particularly high risk of contracting emerging
blood-borne infectious diseases. Although these
studies cannot exclude the possibility that transmission
of sCJD might have occurred through
blood transfusions in rare cases, and despite
the fact that the etiology of sCJD is unclear,
it would appear that transmission of sCJD by
trans fusion of blood or blood products does
not play a major role in the pathogenesis of this
disease entity.
In the case of acquired prion diseases, however,
the situation is very different. For vCJD, high
levels of prion infectivity and of PrPSc have
been detected in lymphoid organs such as the
appendix and tonsils (Figure 1).23,24 For this
reason, it has been speculated for almost a decade
that vCJD might be associated with a higher risk
of blood-borne transmission than sCJD. It is
important to be cautious, however, as the differences
in the organ tropism of sCJD and vCJD
might be quantitative rather than qualitative, and
PrPSc has been detected in the lymphoid organs
of both vCJD and sCJD patients.21 Initial studies
have failed to detect PrPSc and prion infectivity
in the blood of patients with vCJD, but these
negative data are likely to be attributable to
the lack of sensitivity of the assays available at
the time.23
The recent identification of three indiv iduals
with probable transmission of vCJD via blood
transfusion has provided tragic evidence that vCJD
prions can indeed be transmitted through blood
(Figure 2). On the basis of the epi demiological
and pathogenetic considerations discussed above,
it can only be a matter of time before further
cases of blood-transfusion-associated cases of
vCJD will ensue (Figure 3).
In the first of the cases reported, a patient
received a single unit of non-leukodepleted
erythrocyte concentrate from an individual who
went on to develop vCJD 3.5 years later, and
was therefore likely to have been subclinically
prion-infected at the time of the donation. The
recipient developed vCJD 6.5 years following
the transfusion.25
In the second case, transmission of prion
disease occurred again via a single unit of nonleukodepleted
red-blood-cell concentrate.
The donor developed vCJD 2 years following
blood donation, again raising the possibility
of pre clinical infection at the time of the donation.
18 The recipient died of causes unrelated
to the prion infection 5 years after the transfusion.
Although this individual did not display
overt signs of vCJD, PrPSc could be detected
in lymphoid organs, enforcing the concept of
subclinical prion disease in this individual.
Recently, a third case of blood-borne prion
transmission has been reported.26 In this case,
the incubation time in the recipient was 8 years,
whereas the donor showed vCJD symptoms
20 months following his blood donation.
Until now, sequencing of the PRNP gene
in all individuals who succumbed to vCJD
revealed homozygosity for the sequence ‘ATG’,
which encodes methionine, at codon 129. In
the general population, only 33% of people are
homozygous for ATG at this codon of PRNP, so
this particular genetic trait, known as the MM
genotype, has been regarded as a risk factor for
vCJD.8 The second identified recipient of prioninfected
blood, however, was heterozygous for
methionine/valine at codon 129 (MV genotype).
The MV genotype is underrepresented in
sporadic and acquired CJD, and has therefore
been considered a protective genetic trait. The
fact that this individual died of a cause unrelated
to prion disease raises the question of whether
MV heterozygotes might develop a permanent
carrier status, in which the prion replicates
within their body but clinical signs are absent
for an indeterminate period of time.
Of course, it would be imprudent to draw
far-reaching conclusions on the basis of three
cases of blood-borne prion infection. We deem
it justified, however, to highlight a number of
surprising details that have become clear on
analysis of these cases.
First, vCJD prions can indeed propagate using
blood as a vector. In the past, this idea has often
been regarded as ‘worst-case scenario’, ‘highly
specula tive’, and ‘barely a theoretical possibility’.
The wishful thinking of many physicians
involved in blood transfusion has often conjured
up a sense of safety, which, as we regrettably now
know, is unwarranted.
Second, a single unit of vCJD-prion-infected
blood is sufficient to cause transmission of the
disease. This fact is particularly unsettling, as it
can only be taken to signify that the concentration
of ID50 units in blood is relatively high.
One ID50 unit is defined as the infectious
dose sufficient to establish infection in 50% of
recipients; animal experiments indicate that the
amount of prion infectivity needed to reach
one ID50 unit is much higher when prions are
administered intravenously than when they
are inoculated intracerebrally.
Third, blood from preclinically vCJD-infected
patients can be infectious. Although not
un expected, this aspect is particularly worrisome,
as it suggests that preclinical donors
might subjectively not consider themselves at
risk. Consequently, the only way to identify such
donors would be to subject the donation to a
prion screen of satisfactory sensitivity, which is
currently unavailable.
Last, despite all epidemiological evidence to
the contrary, patients who are methionine/valine
heterozygous at codon 129 of the PRNP gene are
susceptible to infection with vCJD prions, which
raises several important questions. Is the virulence
of BSE prions enhanced when passaged
from human to human, as opposed to the
original bovine to human situation? Passaging
experiments of scrapie infectivity between mice
and hamsters indicate that this scenario is highly
plausible.6 Even more importantly, can vCJD
infection of heterozygous individuals establish
a permanent subclinical carrier state? Although
this situation might constitute a best-case
scenario for the infected individuals, it could be
disastrous from an epidemiological viewpoint,
as it might lead to an unrecognized and possibly
self-sustaining epidemic. ...
snip...
JUNE 2006 VOL 2 NO 6 AGUZZI AND GLATZEL NATURE CLINICAL PRACTICE NEUROLOGY
329