Assessment of immunovirological features in HIV related non-Hodgkin lymphoma patients and their impact on outcome.

Abstract

Despite the era of highly active antiretroviral therapy, non-Hodgkin lymphoma (NHL) remains one of the main causes of death in HIV-infected patients, with a wide variation on the outcome.
We investigated immunological status and EBV, HHV8, HIV viral load in a group of HIV-infected patients at diagnosis of NHL to evaluate their prognostic significance.
Eighty-one consecutive HIV+ NHL patients were studied. CD4 and CD8 cell counts, HHV8 DNA, EBV DNA, HIV RNA and HIV DNA were assessed at diagnosis and at 3 months after chemotherapy initiation. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) of disease free survival (DFS) and overall survival (OS) were computed according to CD4 and CD8 cell counts, EBV DNA, HIV RNA and HIV DNA. HRs were, thereafter, computed also for continuous variation of CD4, CD8 cell counts and EBV DNA.
In the multivariate analysis, CD4<160 and CD8<590 cell/μl and EBV DNA≥300 c/ml were independently associated to DFS (HR=2.98; 95%CI: 1.26-7.03; HR=2.65, 95%CI: 1.13-6.19; HR=4.01; 95%CI: 1.81-8.91) and OS (HR=3.32; 95%CI: 1.41-7.83; HR=4.62, 95%CI: 1.91-11.19; HR=3.11, 95%CI: 1.42-6.80). HRs for DFS and OS decreased continuously with increasing CD4 and CD8 cell counts, while they increased continuously with increasing EBV DNA levels.
The association with survival of low CD4 and CD8 cell counts and detectable EBV viremia, measured at lymphoma's diagnosis, identified three independent prognostic biomarkers that might help in the management of NHL HIV+ patients, offering complementary information in the ascertainment of their outcome.

Full text

Page 1

Journal
of
Clinical
Virology
53 (2012) 297–
301
Contents
lists
available
at
SciVerse
ScienceDirect
Journal
of
Clinical
Virology
j ourna
l ho
mepage:
www.elsevier.com/locate/jcv
Assessment
lymphoma
of
immunovirological
features
in
HIV
related
non-Hodgkin
patients
and
their
impact
on
outcome
Rosamaria
Emanuela
Tedeschia,∗,
Maria
Teresa
Bortolina, Ettore
Bidolib,
Stefania
Zanussia,
Chiara
Pratesia,
Vaccherc,
Umberto
Tirellic,
Paolo
De
Paolid
aMicrobiology-Immunology
bEpidemiology
cMedical
dScientific
and
Virology
Unit,
Centro
di Riferimento
Oncologico,
IRCCS,
via
F. Gallini
2, 33081
Aviano,
Italy
Unit,
Centro
di
Riferimento
Oncologico,
IRCCS,
via
F. Gallini
2, 33081
Aviano,
Italy
Oncology
Department,
Centro
di Riferimento
Oncologico,
IRCCS,
via
F. Gallini
2, 33081
Aviano,
Italy
Directorate,
Centro
di
Riferimento
Oncologico,
IRCCS,
via
F. Gallini
2,
33081
Aviano,
Italy
a
r
t
i
c
l
e

i
n
f
o
Article
Received
Received
13
Accepted
history:
24 August
2011
in revised
form
December
2011
19 December
2011
Keywords:
NHL
HIV
Viral
CD4
CD8
EBV
load
a
b
s
t
r
a
c
t
Background:
remains
Objectives:
patients
Study
DNA,
initiation.
(DFS)
and
EBV
Results:
independently
95%CI:
1.42–6.80).
they
Conclusions:
measured
in
their
Despite
the
era
of
highly
active
antiretroviral
therapy,
non-Hodgkin
lymphoma
(NHL)
one
of
the
main
causes
of
death
in
HIV-infected
patients,
with
a
wide
variation
on
the
outcome.
We
investigated
immunological
status
and
EBV,
HHV8,
HIV
viral
load
in
a
group
of
HIV-infected
at
diagnosis
of NHL
to
evaluate
their
prognostic
significance.
design:
Eighty-one
consecutive
HIV+
NHL
patients
were
studied.
CD4
and
CD8
cell
counts,
HHV8
EBV
DNA,
HIV
RNA
and
HIV
DNA
were
assessed
at
diagnosis
and
at
3
months
after
chemotherapy
Hazard
ratios
(HRs)
and
corresponding
95%
confidence
intervals
(CIs)
of
disease
free
survival
and
overall
survival
(OS)
were
computed
according
to
CD4
and
CD8
cell
counts,
EBV
DNA,
HIV
RNA
HIV
DNA. HRs
were,
thereafter,
computed
also
for
continuous
variation
of CD4,
CD8
cell
counts
and
DNA.
In
the
multivariate
analysis,
CD4
<
160
and
CD8
<
590
cell/?l
and
EBV
DNA
≥
300
c/ml
were
associated
to
DFS
(HR
=
2.98;
95%CI:
1.26–7.03;
HR
=
2.65,
95%CI:
1.13–6.19;
HR
=
4.01;
1.81–8.91)
and
OS
(HR
=
3.32;
95%CI:
1.41–7.83;
HR
=
4.62,
95%CI:
1.91–11.19;
HR
=
3.11, 95%CI:
HRs
for
DFS
and
OS
decreased
continuously
with
increasing
CD4
and
CD8
cell
counts,
while
increased
continuously
with
increasing
EBV
DNA
levels.
The
association
with
survival
of
low
CD4
and
CD8
cell
counts
and
detectable
EBV
viremia,
at
lymphoma’s
diagnosis,
identified
three
independent
prognostic
biomarkers
that
might
help
the
management
of
NHL
HIV+
patients,
offering
complementary
information
in
the
ascertainment
of
outcome.
© 2012 Elsevier B.V. All rights reserved.
1.
Background
Non-Hodgkin
from
cal
immune
stantial
the
at
stratification
who
lymphoma
(NHL)
in HIV-infected
subjects
differs
NHL
in the
HIV-seronegative
population
in several
clini-
and
immunohistological
aspects.1Although
improvements
in
functions,
due
to HAART
introduction,
have
led
to sub-
decrease
in overall
NHL
risk,
this
disease
remains
one
of
main
causes
of
death.
The
acquisition
of
prognostic
parameters
initial
diagnosis
may
contribute
to implementation
of risk
based
of
therapy
and
may
facilitate
identification
of
those
may
benefit
from
early
intensive
therapy.
Abbreviations:
Human
∗Corresponding
E-mail
NHL,
non-Hodgkin
lymphoma;
EBV,
Epstein
Barr
virus;
HHV8,
herpes
virus
8; PBMCs,
peripheral
blood
mononuclear
cells.
author.
Tel.:
+39
0434
659658;
fax:
+39
0434
659402.
address:
rtedeschi@cro.it
(R.
Tedeschi).
Nevertheless,
the
degree
of
immune
deficiency
has
been
proved
to
disease.2In addition,
latent
tant
is detected
owing
EBV
the
phoma
absolute
for
high
EBV
gate
relationship
beyond
lymphoma.10HIV
be
one
strong
risk
factor
for
NHL
during
the
progressive
HIV
the
loss
of
cell
mediated
immune
control
of
herpesviruses
infections
was
often
observed
as
an
impor-
contribute
to virus
reactivation
with
severe
complications.
EBV
in 40–90%
of
AIDS
related
lymphomas
(ARL),
probably
to a lack
of
EBV
control,3and
transformation
of
B cells
by
is likely
to play
a
role
in some
forms
of
NHL.4The
impact
of
tumour-cell
EBV
status
on
the
prognosis
of
patients
with
lym-
has
been
studied
with
controversial
results,5,6whereas
the
EBV
DNA
load
had
poor
diagnostic
and
prognostic
values
AIDS–NHL,
particularly
because
of
its
lack
of
specificity
and
its
fluctuation
over
time.7,8However,
a recent
study
indicated
quantification
in HIV
patients
treated
for
NHL
as
a surro-
marker
of
clinical
outcome
of
AIDS–NHL.9Recently,
a
broader
between
HHV8
infection
and
ARL
has
been
proposed,
that
one
known
for
clinically
distinctive
primary
effusion
contribution
to lymphomagenesis
as
pro-virus
1386-6532/$
doi:10.1016/j.jcv.2011.12.021
– see
front
matter ©

2012 Elsevier B.V. All rights reserved.

Page 2

298
R.
Tedeschi
et al.
/ Journal
of
Clinical
Virology
53 (2012) 297–
301
was
and/or
replication,
CD4+
viral
of
recently
transplantation.13
Improvements
HIV-related
response
lation.
prognostic
these
tion
decisions.
also
suggested,
through
deregulation
of
oncogenes
expression
transcription
factors.11Despite
the
HAART
control
on
virus
HIV
persists
at low
levels
mainly
in infected,
resting
T cells,
the
stable
viral
reservoir.
Independently
of
HIV
RNA
load,
HIV-DNA
evaluation
may
have
a clinical
value
as
a marker
viral
burden,
disease
progression
and
death,12as
we
have
also
reported
in the
setting
of
HIV
lymphoma
and
stem
cells
in outcome
and
survival
of
patients
treated
for
NHL
have
been
reported
in the
post-HAART
era,
with
rates
and
survival
approaching
those
of
the
general
popu-
The
availability
of
more
appropriate
and
reliable
laboratory
markers
might
help
clinicians
in the
management
of
high
risk
patients,
and
it might
offer
complementary
informa-
in the
ascertainment
of
NHL
outcome,
thus,
facilitating
clinical
2.
Objectives
Our
study
investigated
EBV,
HHV8
and
HIV
viral
load
as
well
as
immunological
diagnosis.
free
cal
patients’
status
in a group
of
HIV-infected
patients
at
NHL
The
laboratory
findings
were
correlated
with
disease
and
overall
survival
to assess
whether
any
of
the
virologi-
or
immunological
parameters
could
have
a
predictive
value
on
outcome.
3.
Study
design
3.1.
Patients
and
samples
Eighty
one
HIV-infected
patients
who
developed
NHL,
followed
up
and
in
chemotherapy
reported.14–17Of
due
range:
Serum
sis
history,
ples
previously
obtained
during
1996–2006,
were
recruited
in the
study.
Their
clinical
histological
characteristics
at NHL
diagnosis
are
summarized
Table
1.
All
patients
received
treatment
with
combination
regimens
as
part
of
a phase
2 or
3 study,
as
the
81 patients,
48.1%
were
alive
and
51.9%
died
to progressive
disease.
The
median
follow
up was
14 months,
0–121
months.
and/or
PBMCs
samples
were
collected
at time
of
diagno-
and
after
3 months
of
chemotherapy;
however,
due
to medical
sampling,
and/or
technical
gaps,
serum
and/or
PBMCs
sam-
were
not
always
available
for
all
patients.
An
informed
consent,
approved
by the
Internal
Review
Board
of
Ethics,
was
from
all
patients.
3.2.
Laboratory
analyses
Samples
PBMCs
saline
kit
instructions.
EBV
Man
System
and
EPICS
by
detect
phenol
primers
the
aliquots
of
200
?l of
serum
and
of
about
2 × 106
thawed
and
resuspended
in 200
?l of
phosphate-buffered
(PBS)
were
processed
for
DNA
extraction
by
QiAmp
Blood
(Qiagen
Gmbh,
Hilden,
Germany),
following
manufacturer’s
DNA
and
HHV8
DNA
were
measured
by
a real
time
Taq-
PCR,
by using
the
ABI
PRISM
7900
HT
Sequence
Detection
(Applied
Biosystems),
as
previously
described.18,19CD4
CD8
lymphocyte
counts
were
analyzed
on
whole
blood
by
an
XL
flow
cytometer
(Beckman–Coulter).
HIV
RNA
was
assessed
Quantiplex
HIV
bDNA
assay
(3.0v,
Siemens).
To quantitatively
HIV1
DNA,
genomic
DNA
was
extracted
from
PBMCs
by
chloroform
technique
and
a real
time
PCR
was
used
with
and
TaqMan
probe
detecting
a 121-bp
DNA
fragment
in
viral
LTR
region.13,20
Table
Baseline
NHL
1
clinical
and
immunovirological
characteristics
of
the
81
HIV+
patients
at
diagnosis.
VariableNo.
(%)Median
(range)
Age
Gender
Male
Female
Clinical
Diffuse
Burkitt
Immunoblastic
Plasmoblastic
Unknown
Risk
Injection-drug
Homosexual
Heterosexual
Unknown
Performance
0
1
2
3–4
Grade
Intermediate
High
Stage
I
II
III
IV
Age-IPI
0
1
2
3
HAART
Yes
No
LDH
VES
Albumin
WBC
Hb
Platelet
Neutrophil
Lymphocyte
EBV
EBV
HIV
HIV
CD4%
CD4
CD8%
CD8
at diagnosis
(yr)

–
40 (26–63)

64
17
(79%)

–
–
(21%)

subtype
large
B cell

38
20(24.79%)
10
7 (8.6%)
6 (7.4%)
(46.9%)

–
–
–
–
–

(12.3%)



group
use

30
19(23.5%)
24
8
(37%)

–
–
–
–

(29.6%)

(9.9%)

status
(WHO)
21(25.9%)
40
14
6 (7.4%)

–
–
–
–
(49.4%)

(17.3%)


10(12.3%)
71

–
–
(87.7%)


7 (8.6%)
7 (8.6%)
14
53

–
–
–
–

(17.3%)

(65.4%)


12
14
39
16
(14.8%)

–
–
–
–
(17.3%)

(48.2%)

(19.8%)

pre-CT

53
28(34.6%)
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
(65.4%)

–
–
642
48
4 (2–5)
5970
12
240
3477
1200
301
4145
2856
779
13
161
56
592

(?l)
(161–16,300)
(2–125)
(g/dl)

(×103/?l)
(1250–26,100)
(g/dl)
(7–16)
count
(×103/?l)
(25–633)
(×103/?l)
(650–21,000)
(×103/?l)
(140–23,000)
DNA
c/ml
(0–5,317,283)
DNA
c/106PBMCs
(30–589,559)
RNA
c/ml
(49–500,001)
DNA
c/106PBMCs
(5–26,791)
(1–47)
cells/?l
(3–1247)
(17–82)
cells/?l
(22–4168)
3.3.
Statistical
analysis
The
Spearman
correlation
coefficient
(r)
was
used
to analyze
the
correlations
variables.
pare
disease
through
was
OS
The
OS
ard
were
median
variation
multivariate
between
the
studied
immunological
and
virological
The
non-parametric
Wilcoxon
test
was
used
to com-
viral
load
before
and
after
chemotherapy.
Information
on
free
survival
(DFS)
and
overall
survival
(OS)
was
obtained
an
active
follow-up
to verify
patients’
vital
status:
DFS
measured
from
the
date
of
diagnosis
(baseline)
to relapse;
was
measured
from
the
date
of
diagnosis
(baseline)
to death.
impact
of
immunological
and
virological
factors
on
DFS
and
was
estimated
with
the
Cox
proportional
hazards
model.
Haz-
ratios
(HRs)
and
corresponding
95%
confidence
intervals
(CIs)
calculated
by
dichotomizing
each
variable
by means
of
their
value.
HRs
were
subsequently
computed
for
continuous
of
the
variables
resulting
associated
to DFS
and
OS
at
analysis.
All
the
analyses
were
adjusted
for
aa-IPI

Page 3

R. Tedeschi
et
al.
/
Journal
of
Clinical
Virology
53 (2012) 297–
301
299
Table 2
Hazard
ratios
(HR)
and
corresponding
95%
confidence
intervals
(CI)
for
disease
free
and
overall
survival
evaluated
by immunovirological
parameters
in 81
HIV+
NHL
patients.
Variable

Mean
in
survival
(standard
error)
months
HRa
95%CI

p-Value

HRb
(95%CI)

p-Value
Disease
CD4
≥13
<13
CD4
≥160
<160
CD8
≥55
<55
CD8
≥590
<590
Serum
<300
≥300
EBV
<4000
≥4000
HIV1
<3000
≥3000
HIV1
<800
≥800
Overall
CD4
≥13
<13
CD4
≥160
<160
CD8
≥55
<55
CD8
≥590
<590
Serum
<300
≥300
EBV
<4000
≥4000
HIV1
<3000
≥3000
HIV1
<800
≥800
aAdjusted
bAdjusted
free
survival
(%)

12.3 (1.1)
11.3 (1.0)

1
1.54
–
–
(0.71–3.33)

0.27

–
–
(cells/?l)

13.4 (1.0)
8.9 (0.8)

1
2.98
1
2.11

(1.26–7.03)

0.01

(0.81–5.47)

0.13
(%)

11.7 (1.0)
12.0 (1.1)

1
0.96
–
–
(0.49–1.89)

0.92

–
–
(cells/?l)

13.2 (1.0)
9.8

1
2.65
1
2.11
(0.9)


(1.13–6.19)

0.02

(0.84–5.31)

0.11
EBV
DNA
(c/ml)
13.4
10.4 (1.0)
(1.0)

1
4.01
1
4.35

(1.81–8.91)

0.0006

(1.91–9.94)

0.0005
DNA
(c/106PBMCs)

12.5 (0.9)
10.2 (1.0)

1
1.00
–
–
(0.46–2.16)

0.99

–
–
RNA
(c/ml)

12.7 (1.1)
11.0 (1.0)

1
1.86
–
–

(0.83–4.17)

0.13

–
–
DNA
(c/106PBMCs)
12.4
11.0 (1.2)
(0.9)

1
1.11
–
–
(0.49–2.49)

0.81

–
–
survival
(%)

14.1 (1.4)
13.4

1
1.71
–
–
(1.0)


(0.79–3.69)

0.17

–
–
(cells/?l)
14.6
12.8 (1.3)
(0.9)

1
3.32
1
2.56

(1.41–7.83)

0.006

(1.01–6.50)

0.04
(%)

15.7 (1.3)
12.3 (1.1)

1
1.39
–
–
(0.71–2.72)

0.34

–
–
(cells/?l)

17.6 (1.2)
11.0

1
4.62
1
2.00
(1.0)

(1.91–11.19)

0.0007

(0.91–4.38)

0.08
EBV
DNA
(c/ml)

16.8 (1.3)
11.5 (1.0)

1
3.11
1
1.48

(1.42–6.80)

0.005

(0.66–3.28)

0.34
DNA
(c/106PBMCs)
16.1
11.8 (1.2)
(1.1)

1
1.21
–
–
(0.55–2.66)

0.63

–
–
RNA
(c/ml)

16.2 (1.3)
12.2 (1.0)

1
2.04
–
–
(0.91–4.60)

0.08

–
–
DNA
(c/106PBMCs)

11.3 (0.7)
14.5 (1.5)

1
1.11
–
–
(0.49–2.49)

0.81

–
–
for
age,
age-IPI,
risk
group,
HAART
pre-diagnosis,
NHL
grade
and
histotype,
type
of
therapy
and
HAART
generation.
for
all
previous
variables,
plus
CD4
and
CD8
cell
counts
and
serum
EBV
DNA.
(age-adjusted
mance
risk
type
age
SAS
used
p-values
IPI
score,
calculated
using
three
variables,
perfor-
status
ECOG
> 2,
high
LDH
levels
and
clinical
stage
3 or 4),
HIV
factors,
HAART
use
before
diagnosis,
NHL
grade
and
histotype,
of
chemotherapy
and
HAART;
considering
that
patients’
mean
was
below
sixty
years,
age
was
also
included
for
adjustment.
version
9.20
(SAS
Institute
Inc.,
Cary,
NC,
USA,
1999–2001)
was
for
statistical
analyses.
All
statistical
tests
were
two-sided,
and
<0.05
were
considered
statistically
significant.
4.
Results
4.1.
Virological
evaluations
of
NHL
HIV+
patients
At NHL
(median
PBMCs
333,895
diagnosis,
HHV8
DNA
was
detected
in both
serum
value:
(median
14,937,
value:
range:
undetectable,
undetectable–783,332
range:
c/ml)
and

undetectable–
c/106PBMCs)
of
6/81
patients
(7.4%).
HHV8
viral
load
could
for
patients
Serum
a
PBMCs-associated
value
a statistically
samples
2–11,563
HIV
ent,
range
cally
viremia
485,
chemotherapy).
not
be
evaluated
after
chemotherapy,
due
to missing
samples
these
patients;
HHV8
DNA
was
always
undetectable
among
who
were
negative
at diagnosis.
EBV
DNA
was
detected
in 55/81
patients
(68%)
with
median
value
of
301,
range
undetectable–5,317,283
c/ml,
and
EBV
DNA
in 71/71
patients
(100%)
with
a median
of
4145,
range
30–589,559
c/106cells.
After
chemotherapy,
significant
decrease
of
EBV
load
in both
biological
was
measured,
with
undetectable
median
values
(range
c/ml
and
9–22,270
c/106PBMCs;
p < 0.001).
RNA
levels
before
and
after
chemotherapy
were
also
differ-
with
a median
value
of
2856,
range
49–500,001
c/ml,
and
49,
49–437,385
c/ml,
respectively
(p < 0.01);
whereas
no
statisti-
significant
difference
was
detected
when
considering
HIV
DNA
(median
value:
779,
range:
5–26,791
and
median
value:
range:
81–3039
c/106PBMCs,
respectively
before
and
after

Page 4

300
R.
Tedeschi
et al.
/ Journal
of
Clinical
Virology
53 (2012) 297–
301
Fig.
counts,
1. Smoothing
spline
plot
of
unadjusted
hazard
ratios
(HRs)
for
disease
free
survival
(DFS)
and
overall
mortality
(OS)
in HIV+
NHL
patients
according
to CD4
and
CD8
cell
and
serum
EBV
DNA
evaluated
at
diagnosis.
Hatched
lines:
95%
confidence
intervals.
Baseline
serum
or
PBMCs
EBV
DNA
levels
were
not
associated
to
CD8
tively);
(r
CD4
(r = −0.11,
p = 0.33
and
r = −0.15,
p = 0.21,
respectively)
or
cell
counts
(r = −0.03,
p = 0.80
and
r = −0.03,
p = 0.82,
respec-
neither
baseline
serum
EBV
DNA
was
associated
to HIV
RNA
= 0.10,
p = 0.37).
4.2.
Immunovirological
evaluations
and
patients’
survival
We estimated
logical
NHL
multivariate
ease
95%CI:
1.13–6.19;
predictors
vs
590+,
ciations
for
and
p = 0.005,
HIV
Subsequently,
associated
counts
means
tinuously,
CD4
constantly
were
Finally,
adjusted
associations
ciation
1.91–9.94;
for
worth
for
the
HRs
of
DFS
and
OS
according
to viro-
and
immunological
parameters
assessed
at
time
of
diagnosis,
before
starting
chemotherapy
(Table
2).
In the
proportional-hazards
model,
an
increased
risk
of
dis-
progression
was
associated
to CD4
< 160
cell/?l (HR
= 2.98,
1.26–7.03;
p = 0.01)
and
CD8
< 590
cell/?l (HR
= 2.65,
95%CI:
p = 0.02).
Both
CD4
and
CD8
cell
counts
remained
strong
of
an increased
risk
of death
(HR
= 3.32
for
CD4
< 160
160+,
95%CI:
1.41–7.83;
p = 0.006
and
HR
= 4.62
for
CD8
< 590
vs
95%CI:
1.91–11.19;
p = 0.0007).
Statistically
significant
asso-
of serum
EBV
DNA
to DFS
and
OS
were
found
(HR
= 4.01
EBV
DNA
≥ 300
vs
<300
c/ml,
95%CI:
1.81–8.91;
p = 0.0006
HR
= 3.11
for
EBV
DNA
≥ 300
vs <300
c/ml,
95%CI:
1.42–6.80;
respectively).
PBMCs-associated
EBV
DNA,
HIV
RNA
or
DNA
were
not
associated
to DFS
or
OS.
to further
support
the
associations,
variables
to DFS
and
OS
at the
multivariate
analysis
(CD4,
CD8
cell
and
serum
EBV
DNA)
were
evaluated
for
continuous
risk
by
of
Cox
model
(Fig.
1).
The
HRs
for
DFS
and
OS
decreased
con-
without
a precise
identified
threshold
value,
either
with
or CD8
cell
count
decreases
(with
the
upper
limits
of
the
95%CI
below
1.0);
whereas,
increasing
serum
EBV
DNA
levels
associated
to a continuous
increased
risk
of
DFS
and
OS.
when
in the
multivariate
model,
the
analyses
were
also
for
CD4
and
CD8
cell
counts
and
serum
EBV
DNA,
the
were
confirmed;
in particular
serum
EBV
DNA
asso-
to DFS
(HR
= 4.35
for
EBV
DNA
≥ 300
vs
<300
c/ml,
95%CI:
p < 0.001)
and
CD4
cell
count
association
to OS
(HR
= 2.56
CD4
< 160
vs
160+,
95%CI:
1.01–6.50;
p = 0.04)
(Table
2).
It is
noting
that
the
mean
times
of
DFS
and
OS
were
very
close
the
three
significant
variables.
5.
Discussion
We

retrospectively

analyzed

viral

load

for

HIV,

HHV8
and
lected
identifying
vival.
CD4
emerged
OS.
Several
and
the
were
lymphoma-related
used.17,23The
is
in
cell
HRs
IPI-score
quently
within
impact
phoma
addition,
EBV-unrelated
CD4+
features.
nificantly
median
and
death.
the
considerable
observed,
EBV
ciation
reflect
EBV
and
immunological
status
in 81
consecutive,
unse-
HIV+
patients
at
NHL
diagnosis,
with
the
objective
of
laboratory
prognostic
biomarkers
of
patient’s
sur-
Cox
proportional
hazards
analyses
were
performed
and
< 160
cell/?l,
CD8
< 590
cell/?l and
serum
EBV
DNA
>
300
c/ml
as
factors
independently
associated
to decreased
DFS
and
prognostic
determinants
were
described
in HIV+
NHL
subdivided
into
HIV-
and
lymphoma-related
factors.
Among
HIV-related
ones,
low
CD4+
counts
and
the
use
of
HAART
frequently
associated
with
survival,21,22and
among
the
ones,
IPI
score
is probably
the
most
widely
issue
of
which
ones
are
the
stronger
prognosticators
still
in debate;
it is
likely
that
both
factors
play
a role
and
interact
HIV+
lymphoma.
In our
group
of
patients,
CD4
but
also
CD8
low
counts
were
associated
to worse
survival,
further
supported
by
evaluation
in continuous,
while
no significant
association
for
was
found.
Malignancies
in
HIV
infected
patients
are
fre-
associated
to EBV
co-infection,24with
viral
DNA
present
the
malignant
cells
in about
half
of
affected
patients.
The
of
the
tumour-cell
EBV
status
on
the
prognosis
of
lym-
patients
has
been
studied
with
controversial
results.5,6In
some
of
these
tumours
may
belong
to the
AIDS-related
lymphomas,25where
a decrease
in the
number
of
lymphocytes
and
persistently
high
EBV
viremia
are
consistent
In our
study,
serum
EBV
load
after
chemotherapy
was
sig-
lowered;
moreover,
EBV
viremia
at
diagnosis
above
the
value
of
300
c/ml
was
associated
to relapse/progression
worse
survival
with
an observed
time
very
close
to that
of
Although
EBV
viremia
seems
a suboptimal
marker
to predict
occurrence
of
NHL
in HIV,
as
high
viral
loads
are
present
with
fluctuations
over
time
among
HIV+,26a
recent
study
in agreement
with
our
findings,
a
significant
decrease
of
load
in patients
treated
with
chemotherapy
and
a strong
asso-
with
NHL
outcome.9The
nature
of
serum
EBV
DNA
may
a release
of EBV
fragments
from
infected
tumour
cells
rather

Page 5

R. Tedeschi
et
al.
/
Journal
of
Clinical
Virology
53 (2012) 297–
301
301
than
ruled
EBV
replication,
even
if lytic
reactivation
has
not
been
formally
out
in immunocompromised
When
prevalence
without
study,2814%
Lin
(22%)
associated
among
consider
Our
nature
short
the
disease
standing
impact
defined
and,
on
deeper
not
this,
EBV
HIV+
ate
might
NHL
aging
Funding: The
zata
Competing
Ethical
patients.27
compared
to previous
studies,
lower
HHV8
infection
(7.4%)
and
no virus
reactivation
emerged
in our
patients
a history
of
virus-related
disease.
In a nested
case
control
of
NHL
HIV
patients
were
HHV8
DNA
positive
and
et
al.10detected
HHV8
DNA
in both
PBMCs
(19%)
and
plasma
declining
with
chemotherapy.
Since
the
spectrum
of
HHV8-
lymphomas
is
in expansion
and
the
virus
quite
prevalent
HIV+,
viremia
evaluation
could
be
an auxiliary
biomarker
to
in this
context,
as
suggested.10
study
carries
several
weaknesses,
such
as
the
retrospective
of
the
investigation,
the
small
number
of
patients,
and
the
longitudinal
time
of
follow-up,
without
a timeline
to describe
kinetic
of
EBV
DNA
and
immunological
changes,
as
related
to
evolution.
In order
to contribute
further
clues
in under-
the
pathogenic
mechanisms
of
HIV
related
NHL,
the
study
should
be
validated
in a larger
number
of
patients
with
well-
timing.
Furthermore,
tumour-cell
EBV
status
was
unknown
considering
the
reported
different
impact
of
EBV
infection
survival
by different
age
groups
and
histological
defined
type,
analyses
based
on
both
age
and
type
of
lymphoma
could
be
conducted
due
to the
limited
number
of
our
cases.
Despite
our
results
suggest
that
CD4
and
CD8
cell
counts
and
serum
viral
load
are
independently
associated
to poor
survival
among
NHL
patients.
We
believe
that
the
availability
of
appropri-
additional
and
reliable
prognostic
markers
from
the
laboratory
offer
complementary
information
in the
ascertainment
of
outcome
with
considerable
clinical
implications
useful
in man-
these
high
risk
patients.
study
was
in part
supported
from
Ricerca
Finaliz-
and
Fondazione
CRUP.
interests: None
declared.
approval: Not
required.
Acknowledgements
The
authors
wish
to thank
the
patients
and
their
families
and
manuscript.
Mrs
Luigina
Mei
for
her
expert
assistance
in revising
the
References
1. Silvestris
non-Hodgkin’s
strategies.
2.
affecting
human
Oncol
3.
ular pathogenesis
4.
coma herpesvirus/human
Research
5. Park
Barr
2007;110(3):972–8.
6.
EBV-LMP1
2008;32(9):1424–30.
N,
Crucitta
E,
Lorusso
V,
Gamucci
T,
De
Lena
M. AIDS-related
lymphoma:
clinico-pathological
characteristics
and
therapeutic
Int
J Oncol
2002;20(3):611–5.
Pluda
JM,
Venzon
DJ,
Tosato
G,
Lietzau
J, Wyvill
K, Nelson
DL,
et al.
Parameters
the
development
of
non-Hodgkin’s
lymphoma
in patients
with
severe
immunodeficiency
virus
infection
receiving
antiretroviral
therapy.
J Clin
1993;11(6):1099–107.
Carbone
A.
AIDS-related
non-Hodgkin’s
lymphomas:
from
pathology
and
molec-
to treatment.
Hum
Pathol
2002;33(4):392–404.
International
Agency
for
Research
on
Cancer.
Epstein
Barr
virus
and
Kaposi’s
sar-
herpesvirus
8.
Lyon,
France:
International
Agency
for
on
Cancer;
1997.
S,
Lee
J, Ko YH,
Han
A,
Jun
HJ,
Lee
SC,
et al.
The
impact
of
Epstein-
virus
status
on clinical
outcome
in diffuse
large
B-cell
lymphoma.
Blood
Paydas
S, Ergin
M,
Erdogan
S,
Seydaoglu
G.
Prognostic
significance
of
and
VEGF-A
expressions
in non-Hodgkin’s
lymphomas.
Leuk
Res
7.
Dehee
Quantification
immunodeficiency
2001;65(3):543–52.
A,
Asselot
C,
Piolot
T, Jacomet
C,
Rozenbaum
W,
Vidaud
M,
et
al.
of
Epstein-Barr
virus
load
in peripheral
blood
of
human
virus-infected
patients
using
real-time
PCR.
J Med
Virol
8.
Fan
load as a
H,
Kim
SC,
Chima
CO,
Israel
BF,
Lawless
KM,
Eagan
PA,
et al.
Epstein-Barr
viral
marker
of
lymphoma
in AIDS
patients.
J Med
Virol
2005;75(1):59–69.
9.
Bonnet
gitudinal
non-Hodgkin
10. Lin
chemotherapy
coma herpesvirus
11.
of
Infect
12. Kostrikis
Karafoulidou
forms
progression
13. Bortolin
tive
Patients
transplantation
14.
logical,
highly
15.
et
nisone
with human
2001;91(1):155–63.
16. Spina
Rituximab
HIV-associated
Blood
17. Mounier
related
adapted
18.
value
patients
2006;233(2):247–54.
19.
herpesvirus
patients
20.
5
count
21. Lim ST,
HIV-related
antiretroviral
22. Simcock
ment
antiretroviral
2007;12(6):931–9.
23.
de
of
and
factors.
24. Wood
2005;15:947–52.
25.
26. Van
F, et
virus
J
27.
ization
lymphoma
28.
O,
associated
Syndr
F,
Jouvencel
AC,
Parrens
M,
Leon
MJ,
Cotto
E, Garrigue
I, et
al.
A
lon-
and
prospective
study
of
Epstein-Barr
virus
load
in AIDS-related
lymphoma.
J Clin
Virol
2006;36(4):258–63.
L, Lee
JY,
Kaplan
LD,
Dezube
BJ,
Noy
A, Krown
SE,
et
al.
Effects
of
in AIDS-associated
non-Hodgkin’s
lymphoma
on
Kaposi’s
sar-
DNA
in blood.
J Clin
Oncol
2009;27(15):2496–502.
Katano
H,
Sato
Y,
Hoshino
S,
Tachikawa
N,
Oka
S,
Morishita
Y,
et
al.
Integration
HIV-1
caused
STAT3-associated
B cell
lymphoma
in an
AIDS
patient.
Microbes
2007;9(14–15):1581–9.
LG,
Touloumi
G,
Karanicolas
R,
Pantazis
N,
Anastassopoulou
C,
A,
et
al.
Quantitation
of
human
immunodeficiency
virus
type
1
DNA
with
the
second
template
switch
in peripheral
blood
cells
predicts
disease
independently
of
plasma
RNA
load.
J Virol
2002;76(20):10099–108.
MT,
Zanussi
S,
Talamini
R,
Simonelli
C,
Pratesi
C, Tedeschi
R,
et
al.
Predic-
value
of
HIV
type
1 DNA
levels
on overall
survival
in HIV-related
lymphoma
treated
with
high-dose
chemotherapy
(HDC)
plus
autologous
stem
cell
(ASCT).
AIDS
Res
Hum
Retroviruses
2010;26(2):245–51.
Tirelli
U,
Spina
M,
Gaidano
G,
Vaccher
E, Franceschi
S,
Carbone
A.
Epidemio-
biological
and
clinical
features
of
HIV-related
lymphomas
in the
era
of
active
antiretroviral
therapy.
AIDS
2000;14:1675.
Vaccher
E,
Spina
M,
di
Gennaro
G,
Talamini
R,
Nasti
G,
Schioppa
O,
al.
Concomitant
cyclophosphamide,
doxorubicin,
vincristine,
and
pred-
chemotherapy
plus
highly
active
antiretroviral
therapy
in patients
immunodeficiency
virus-related,
non-Hodgkin
lymphoma.
Cancer
M,
Jager
U,
Sparano
JA,
Talamini
R,
Simonelli
C,
Michieli
M,
et
al.
plus
infusional
cyclophosphamide,
doxorubicin,
and
etoposide
in
non-Hodgkin
lymphoma:
pooled
results
from
3 phase
2 trials.
2005;105(5):1891–7.
N,
Spina
M,
Gabarre
J, Raphael
M,
Rizzardini
G,
Golfier
JB,
et
al.
AIDS-
non-Hodgkin
lymphoma:
final
analysis
of
485
patients
treated
with
risk-
intensive
chemotherapy.
Blood
2006;107(10):3832–40.
Bortolin
MT,
Pratesi
C,
Dolcetti
R,
Bidoli
E,
Vaccher
E,
Zanussi
S,
et
al.
Clinical
of
Epstein-Barr
virus
DNA
levels
in peripheral
blood
samples
of
Italian
with
undifferentiated
carcinoma
of
nasopharyngeal
type.
Cancer
Lett
Tedeschi
R,
Enbom
M,
Bidoli
E,
Linde
A,
De
Paoli
P,
Dillner
J. Viral
load
of human
8 in peripheral
blood
of
human
immunodeficiency
virus-infected
with
Kaposi’s
sarcoma.
J Clin
Microbiol
2001;39(12):4269–73.
Viard
JP,
Burgard
M,
Hubert
JB,
Aaron
L,
Rabian
C,
Pertuiset
N,
et
al.
Impact
of
years
of
maximally
successful
highly
active
antiretroviral
therapy
on CD4
cell
and
HIV-1
DNA
level.
AIDS
2004;18(1):45–9.
Karim
R,
Tulpule
A, Nathwani
BN,
Levine
AM.
Prognostic
factors
in
diffuse
large-cell
lymphoma:
before
versus
after
highly
active
therapy.
J Clin
Oncol
2005;23(33):8477–82.
M,
Blasko
M,
Karrer
U,
Bertisch
B,
Pless
M,
Blumer
L, et
al.
Treat-
and
prognosis
of
AIDS-related
lymphoma
in the
era
of
highly
active
therapy:
findings
from
the
Swiss
HIV
Cohort
Study.
Antivir
Ther
Miralles
P, Berenguer
J, Ribera
JM,
Rubio
R,
Mahillo
B, Téllez
MJ,
et
al.
Grupo
Estudio
del
SIDA
Register
of
Systemic
AIDS-Related
Lymphomas,
Prognosis
AIDS-related
systemic
non-Hodgkin
lymphoma
treated
with
chemotherapy
highly
active
antiretroviral
therapy
depends
exclusively
on
tumor-related
J Acquir
Immune
Defic
Syndr
2007;44(2):167–73.
C,
Harrington
WJ.
AIDS
and
associated
malignancies.
Cell
Res
Boshoff
C,
Weiss
R.
AIDS-related
malignancies.
Nat
Rev
Cancer
2002;2:373–82.
Baarle
D,
Wolthers
KC,
Hovenkamp
E,
Niesters
HG,
Osterhaus
AD,
Miedema
al.
Absolute
level
of
Epstein-Barr
virus
DNA
in human
immunodeficiency
type
1 infection
is
not
predictive
of
AIDS-related
non-Hodgkin
lymphoma.
Infect
Dis
2002;186(3):405–9.
Chan
KC,
Zhang
J, Chan
AT,
Lei
KI,
Leung
SF,
Chan
LY,
et al.
Molecular
character-
of
circulating
EBV
DNA
in the
plasma
of
nasopharyngeal
carcinoma
and
patients.
Cancer
Res
2003;63(9):2028–32.
Beachler
DC,
Gellert
LL,
Jacobson
LP,
Ambinder
RF,
Breen
EC,
Martínez-Maza
et al.
Kaposi
sarcoma-associated
herpesvirus
serum
DNA
and
antibodies
not
with
subsequent
non-Hodgkin
lymphoma
risk.
J Acquir
Immune
Defic
2011;56(2):188–92.