1212 • JID 2009:200 (15 October) • BRIEF REPORT
B R I E F R E P O R T
Serum Immune Activation Markers
Are Persistently Increased in Patients
with HIV Infection after 6 Years
of Antiretroviral Therapy
despite Suppression of Viral Replication
and Reconstitution of CD4+T Cells
Martyn A. French,1Martin S. King,2Jean M. Tschampa,2
Barbara A. da Silva,2and Alan L. Landay3
1School of Pathology and Laboratory Medicine, University of Western Australia,
and Department of Clinical Immunology and Immunogenetics, Royal Perth
Hospital and PathWest Laboratory Medicine, Perth, Australia;
Pharmaceutical Research and Development, Abbott Laboratories, Abbott Park,
Center, Chicago, Illinois
3Department of Immunology and Microbiology, Rush University Medical
The effect of long-term antiretroviral therapy on serum im-
mune activation markers was assessed in a cohort of 63 pa-
tients before and after 6 years of boosted lopinavir–based
antiretroviral therapy. High levels of most markers were as-
sociated with lower CD4+T cell counts at baseline and at
year 6, with the exception of soluble cytotoxic T lymphocyte
antigen-4 (sCTLA-4); high levels of sCTLA-4 were associated
with higher CD4+T cell counts at year 6. Abnormalities of
serum immune activation markers persisted after 6 years of
ART but probably had different causes. Furtherinvestigation
of the clinical usefulness of assaying immunoglobulin A,
neopterin, and sCTLA-4 levels to assess the effectiveness of
treatments for human immunodeficiency virus (HIV) dis-
ease are warranted.
Immune activation induced by human immunodeficiency vi-
rus (HIV) infection has several postulated causes, including
activation of plasmacytoid dendritic cells (pDCs) resulting in
production of interferon a (IFN-a)  and depletion of mu-
Received 30 December 2008; accepted 4 April 2009; electronically published 3 September
Potential conflicts of interest: M.S.K., J.M.T., and B.A.D. are employed by the study sponsor.
M.A.F. and A.L.L.: no conflicts.
Financial support: Abbott Laboratories.
Presented in part: The 4th International AIDS Society Conference on HIV Pathogenesis,
Treatment, and Prevention, Sydney, Australia, 22–25 July 2007 (abstract TUPEA035).
Reprints or correspondence: Prof. Martyn French, Dept. of Clinical Immunology and
Immunogenetics, Royal Perth Hospital, GPO Box X2213, Perth 6847, Australia (martyn
The Journal of Infectious Diseases2009;200:1212–15
? 2009 by the Infectious Diseases Society of America. All rights reserved.
cosal CD4+T cells resulting in the translocation of bacterial
products from the gut . This results in the expression of
activation markers, including CD38 and human leukocyte an-
tigen (HLA)–DR, on the surface of T cells and increased levels
of serum or plasma proteins, including neopterin, soluble tu-
mor necrosis factor receptor (sTNFR) II and immunoglobulin
(Ig) A [3, 4]. Assaying serum immune activationmarkersmight
provide additional information to the enumeration of activated
T cells, because they are derived from different cell types (eg,
neopterin from monocytes and macrophages and IgA from
plasma cells) and reflect different aspects of HIV-induced im-
Treatment of HIV infection with combination antiretroviral
therapy (ART) reduces immune activation, but immune acti-
vation may not return to normal. Ongoing immune activation
is associated with poorer recovery of CD4+T cells during early
 and long-term ART  and might contribute to the path-
ogenesis of “non–AIDS-related HIV diseases,” such as ather-
osclerotic vascular disease and non–AIDS-related cancers .
Monitoring immune activation, in addition to monitoring
CD4+T cell counts, may therefore be clinicallyusefulintreating
patients who receive ART.
We have previously described patients with HIV infection
who received an effective lopinavir and low-dose ritonavir–
based ART regimen for 6 years, after which time, the propor-
tions of activated (CD38+, HLA-DR+) CD4+and CD8+T cells
were significantly lower than they were in untreated patients
with HIV infection and comparable to what they were in HIV-
negative subjects . It is unknown what effect receiving ART
for this length of time has on levels of serumimmuneactivation
markers or what relationship exists between serum immune
activation markers and CD4+T cell counts and T cellactivation.
We therefore assayed levels of 10 immune activation markers
in serum samples obtained from these patients.
Patients and methods.
Serum samples from 63 patients
enrolled into Abbott study 720 , which were stored at base-
line and years 3 and 6 of ART, were obtained for evaluation.
All patients were ART naive when therapy was initiated with
lopinavir and low-dose ritonavir with stavudine and lamivu-
dine. Change of nucleoside analogues was allowed for drug
toxicity. At year 6, all but one patient had a plasma HIV RNA
level !50 copies/mL. Data on CD4+T cell counts are presented
in Table 1, and clinical data are presented elsewhere .
Enzyme-linked immunosorbent assays were used to evaluate
serum levels of soluble cytotoxic T lymphocyte antigen-4
(sCTLA-4), IFN-inducible protein 10 (IP-10), monocyte che-
by guest on February 2, 2016
BRIEF REPORT • JID 2009:200 (15 October) • 1213
at Baseline, Year 3, and Year 6
Clinical and Laboratory Values for Human Immunodeficiency Virus (HIV)–Negative Subjects and for HIV-Infected CasePatients
HIV-infected case patients treated
with lopinavir-ritonavir plus 2 NRTIs
Baseline Year 3Year 6
CD4+T cell count, cells/mm3
HIV-1 RNA level, log10copies/mL
8.0 (5.9–11.4)a, c
7.2 (5.1–9.8)a, c
NRTI, nucleoside reverse transcriptase inhibitor; sCTLA-4, soluble cytotoxic T lymphocyte antigen 4; sTNF , soluble tumor necrosis factor receptor; sTRAIL,soluble
tumor necrosis factor–related induced ligand.
aStatistically significant change from baseline value () .P !.05
bStatistically significantly different from reference value in HIV-negative subjects (
cData not available for HIV-negative control subjects because the enzyme-linked immunosorbent assay used for patient serum samples was not available at
the time of testing.
Data are mean value (interquartile range). IFN, interferon; Ig, immunoglobulin; IP-10, inducible protein 10; MCP-1, monocyte chemotactic protein 1;
motactic protein 1 (MCP-1), and sTNFR-II (Biosource; Invi-
trogen Life Science); IFN-a and soluble TNF-related induced
ligand (sTRAIL) (Diaclone; Tepnel Research Products and Ser-
vices); and neopterin (Immuno-Biological Laboratories). Se-
rum levels of IgG, IgA, and IgM were assayed by immunotur-
bidimetry (Abbott Architect 8200 Chemical Analyzer). Total
and activated CD4+and CD8+T cells were enumerated by flow
cytometry, as described elsewhere .
Serum levels of immune activation markers were summa-
rized at baseline and years 3 and 6 using the median values
and first and third quartiles. Changes from baseline to year 3
and 6 were assessed using a paired Student’s t test. Levels of
immune activation markers were compared at each timepoint,
using a 1-way analysis of variance, with levels of immune acti-
vation markers obtained for a group of HIV-negative subjects.
The relationship between baseline serum immune activation
markers and CD4+T cell counts at baseline and at year 6 and
the relationship between year 6 serum immune activation mark-
was assessed using linear regression. A multivariable linear re-
baseline variables, including baseline CD4+T cell count and se-
rum levels of immune activation markers, and CD4+Tcellcount
at year 6, using a stepwise selection process with a P value of
?.05 required to enter and remain in the model.
Data on serum immune activation markers in
HIV-positive patients at baseline and years 3 and 6 of ART and
comparison of values at these time points with those for non–
HIV-infected subjects are presented in Table 1. At baseline, levels
of all serum immune activation markers were increased in HIV-
jects, with the exception of IgM and sCTLA-4.However,analyses
of sCTLA-4 were complicated by the observation that 67% of
patients had undetectable levels of sCTLA-4. At year 6, serum
levels of IgG, IP-10, sTNFRII, neopterin, and sTRAILwerelower
than baseline, but sTNFRII and sTRAIL levels remained higher
in HIV-infected subjects than in non–HIV-infected subjects. Se-
rum neopterin levels in HIV-infected patients could not becom-
pared with levels in non–HIV-infectedsubjects,becausetheassay
kits were not available.
Serum levels of IgA, IFN-a, and MCP-1 at year 6 were not
statistically significantly lower than serum levels at baseline.
Serum sCTLA-4 levels were !0.20 ng/mL in all 18 non–HIV-
infected subjects, whereas serum levels were 0.20–2.5 ng/mL in
19 (33%) of 58 patients at baseline and in 17 (28%) of 61
patients at year 6, including 9 patients who had levels 10.2 ng/
mL at both time points.
At baseline, levels of all serum immune activation markers
except IgG, IgM, IFN-a, and TRAIL were correlated with CD4+
T cell counts (). All statistically significant correlationsP ! .05
were negative, with the exception of sCTLA-4 level. At year 6,
CD4+T cell counts correlated negatively with baseline serum
levels of IgA (;R p ?0.366 P p .004
), sTNFRII (P p .024R p ?0.295 P p .023
(;). In contrast, baseline serum levels ofR p ?0.267 P p .043
sCTLA-4 correlated positively with year 6 CD4+T cell counts
), MCP-1 ( ;R p ?0.293
) and neopterin;
by guest on February 2, 2016
1214 • JID 2009:200 (15 October) • BRIEF REPORT
antigen 4 (sCTLA-4) correlated positively with year 6 CD4+T cell counts.
B, Serum immunoglobulin (Ig) A levels correlated negatively with CD4+T
cell counts after 6 years of effective antiretroviral therapy (ART).C,Serum
neopterin levels correlated positively with activated CD4+and CD8+T
cells after 6 years of effective ART.
A, Baseline serum levels of soluble cytotoxic T lymphocyte
undetectable in 67% of patients (Figure 1A). In multivariable
analyses, adjusting for the impact of baseline CD4+T cellcount,
the association between baseline serum levels of immune ac-
tivation markers and CD4+T cell counts at year 6 or change
in CD4+T cell count from baseline to year 6 were no longer
Plasma HIV RNA levels and CD4+T cell counts correlated
with serum immune activation markers to a similar degree at
baseline (data not shown), with the exception of neopterin,
which correlated more strongly with HIV RNA levels than did
CD4+T cell count (;
R p 0.55 P ! .001
At year 6, theserumIgAlevelcorrelatednegativelywithCD4+
T cell count (;
R p ?0.338 P p .008
not correlate with the proportion of activated CD4+or CD8+
T cells. In contrast, the serum neopterin level correlated pos-
itively with the proportion of activated CD8+(
) and CD4+T cells (
P p .005R p 0.289 P p .026
but not with CD4+T cell counts.Furthermore,serumneopterin
levels were higher in patients with activated CD4+T cells 18%
and/or activated CD8+T cells 119% (ie, 12 standard deviations
above mean levels) than in patients with activated CD4+T cells
!8% and activated CD8+T cells !19% (8.1 nmol/L vs 17.0
nmol/L; ). There was also a weak correlation between
P p .004
serum sTNFR-II levels and the proportion of activated CD4+
T cells (;).
R p 0.263 P p .04
Increasing evidence suggests that treatments for
HIV infection should suppress immune activation as well as
increase CD4+T cell counts. Assessment of immune activation
by measuring proportions of activatedCD4+and/orCD8+Tcells
may not provide information about all aspects of immune ac-
tivation. Here, we have assayed levels of several serum immune
activation markers in patients who received effective ART for 6
useful, as well as provide information about the characteristics
of ongoing immune activation.
Serum IgA levels at baseline predicted CD4+T cell counts
at year 6 more strongly than any other serum immune acti-
vation marker and did not decrease after 6 years of ART, at
which time they correlated negatively with CD4+T cell counts.
We have previously shown that serum IgA levels are negatively
correlated with effector-memory CD4+T cell responses in pa-
tients with increased CD4+T cell counts who receive ART .
In contrast, we did not demonstrate a relationship between
serum IgA levels and activated T cells after 6 years of ART.
with higher serum neopterin levels. Because serum neopterin
levels are considered to reflect activation of monocytes and
macrophages , activation of T cells and monocytes and mac-
rophages after 6 years of ART may have a similar cause but
one that is different from the cause of high serum IgA levels.
;), although as noted above, sCTLA-4 was
R p 0.355 P p .007
) (Figure 1B), but it did
R p 0.362
) (Figure 1C);
Increased IgA production in patients withHIVinfectiondur-
ing ART might reflect IgA antibody responses to lipopolysac-
charides of gastrointestinal bacteria  as a consequence of
failure to fully regenerate gut-associated lymphoidtissue(GALT)
after long-term ART or increased production of cytokines that
are switch factors for IgA-positive B cells, such as transforming
growth factor b , which is produced by regulatory T cells
that are increased in the blood of patients who are receiving
ART . An increased proportion of regulatory T cells and/
or failure to regenerate lymphoid tissue, represented by GALT,
by guest on February 2, 2016
BRIEF REPORT • JID 2009:200 (15 October) • 1215 Download full-text
might also impair the recovery of total and effector-memory
We show here that higher serum levels of IgA, MCP-1,
sTNFR-II, and neopterin at baseline predicted lower CD4+
T cell counts after 6 years of ART. However, in multivaria-
ble analyses, none of these serum immune activation markers
was as effective as the baseline CD4+T cell count in predicting
low CD4+T cell counts at year 6. Therefore, assaying serum
immune activation markers provides no additional informa-
tion to the CD4+T cell count in predicting recovery of CD4+
T cells on ART.
Serum sCTLA-4 levels before and after 6 years of ART were
higher in 28%–33% of HIV-infected patients than they were
in non–HIV-infected subjects. Similar findings have been re-
ported for patients with autoimmune thyroid disease , in
whom increased serum sCTLA-4 levels appear to reflect pro-
duction of a distinct transcript of the CTLA-4 gene. Interest-
ingly, we found a positive correlation between serum levels
of sCTLA-4 at baseline and CD4+T cell counts at baseline
and at year 6. CTLA-4 is an immunoregulatory receptor on
B and T cells, and ligand engagement reduces T cell activa-
tion. sCTLA-4 may bind to ligands of CTLA-4, thereby pre-
venting their engagement with cell-bound CTLA-4. This might
suppress the inhibitory effect of CTLA-4 on T cell proliferation
and result in increased CD4+T cell proliferation.However,high
serum sCTLA-4 levels might also have deleterious effects by
increasing susceptibility to autoimmune disease, particularly
autoimmune thyroid disease , which is one of several im-
mune reconstitution disorders that may affect patients with
HIV infection who receive ART .
The findings of our study add to the growing body of knowl-
edge on the role of IFN-a in HIV-induced immune activation
and suggest that pDC activation and/or dysfunction persist in
patients who receive ART. As lymph node pDCs appear to be
the major source of IFN-a in patients with HIV infection ,
assaying serum IFN-a levels might be the most convenient way
of assessing this.
There are limitations to our study that should be considered.
First, there may be serum immune activation markers not eval-
uated here that are more relevant than those that we investi-
gated. However, we evaluated markers that appeared to bemost
informative, as determinedbyareviewofpreviouspublications.
Second, the HIV-negative subjects were selected post hoc and
were not matched with the HIV-positive patients by sex, age,
or any other factor.
In summary, higher baseline serum levels of IgA, MCP-1,
sTNFR-II, and neopterin were associated with lower CD4+T
cell counts in patients with untreated HIV infection and in
patients who had received 6 years of effective ART. In contrast,
higher serum levels of sCTLA-4 before ART initiation was as-
sociated with higher CD4+T cell counts at year 6. After 6 years
of effective ART, levels of most serum immuneactivationmark-
ers remained higher in HIV-infected patients thaninnon–HIV-
infected subjects. At this time, IgA levels were negatively cor-
related with CD4+T cell counts, and high serum neopterin
levels were associated with T cell activation. We conclude that
abnormalities of serum immune activationmarkerspersistafter
long-term, effective ART, but they probably have different
causes. Further investigation of the clinical usefulness of as-
saying serum levels of IgA, neopterin, and sCTLA-4 to assess
the effectiveness of ART and other treatments for HIV disease
1. Lehmann C, Harper JM, Taubert D, et al. Increased interferon alpha
expression in circulatingplasmacytoiddendriticcellsofHIV-1–infected
patients. J Acquir Immune Defic Syndr 2008;48:522–30.
2. Brenchley JM, Price DA, Schacker TW, et al. Microbial translocation
as a cause of systemic immune activation in chronic HIV infection.
Nat Med 2006;12:1365–71.
3. Fahey JL, Taylor JM, Detels R, et al. The prognostic value of cellular
type 1. N Engl J Med 1990;322:166–72.
4. Mildvan D, Spritzler J, Grossberg S, et al. Serum neopterin,animmune
activation marker, independently predicts disease progression in ad-
vanced HIV-1 infection. Clin Infect Dis 2005;40:853–8.
5. Hunt PW, Martin JN, Sinclair E, et al. T cell activation is associat-
ed with lower CD4+ T cell gains in human immunodeficiency vi-
rus–infected patients with sustained viral suppression during antiret-
roviral therapy. J Infect Dis 2003;187:1534–43.
6. Fernandez S, Price P, McKinnon EJ, Nolan RC, French MA. Low CD4+
T-cell counts in HIV patients receiving effective antiretroviraltherapyare
effector memory T-cell function. Clin Immunol 2006;120:163–70.
7. Phillips AN, Neaton J, Lundgren JD. The role of HIV in seriousdiseases
other than AIDS. AIDS 2008;22:2409–18.
8. Landay A, da Silva BA, King MS, et al. Evidence of ongoing immune
reconstitution in subjects with sustained viral suppression following 6
years of lopinavir-ritonavir treatment. Clin Infect Dis 2007;44:749–54.
9. Murphy RL, da Silva BA, Hicks CB, et al. Seven-year efficacy of a
lopinavir/ritonavir based regimen in antiretroviral-naı ¨ve HIV-1–in-
fected patients. HIV Clin Trials 2008;9:1–10.
10. French M, Keane N, McKinnon E, Phung S, Price P. Susceptibility to
opportunistic infections in HIV-infected patients with increased CD4
T-cell counts on antiretroviral therapy may be predicted by markers
of dysfunctional effector memory CD4 T cells and B cells. HIV Med
11. Fernandez MI, Pedron T, Tournebize R, Olivo-Marin JC, Sansonetti
PJ, Phalipon A. Anti-inflammatory role for intracellular dimeric im-
munoglobulin A by neutralisation of lipopolysaccharide in epithelial
cells. Immunity 2003;18:739–49.
12. Cerutti A. The regulation of IgA class switching. Nat Rev Immunol
13. Lim A, Tan D, Price P, et al. Proportions of circulating T cells with a
regulatory cell phenotype increase with HIV-associated immune acti-
vation and remain high on antiretroviral therapy. AIDS 2007;21:
thyroid diseases: relationship with clinical status and possible role in the
immune response dysregulation. Clin Immunol 2007;123:190–8.
15. French MA. Immune reconstitution inflammatory syndrome: a re-
appraisal. Clin Infect Dis 2009;48:101–7.
by guest on February 2, 2016