M A J O R A R T I C L E
Pegylated Interferon Alfa-2a Monotherapy
Results in Suppression of HIV Type 1
Replication and Decreased Cell-Associated HIV
Livio Azzoni,1Andrea S. Foulkes,2Emmanouil Papasavvas,1Angela M. Mexas,3Kenneth M. Lynn,1,4Karam Mounzer,5
Pablo Tebas,4Jeffrey M. Jacobson,6Ian Frank,4Michael P. Busch,7,7aSteven G. Deeks,8Mary Carrington,9,9a
Una O’Doherty,3Jay Kostman,4and Luis J. Montaner1
1HIV-1 Immunopathogenesis Laboratory, The Wistar Institute,2Division of Biostatistics and Epidemiology, University of Massachusetts, Amherst,
3Department of Pathology and Laboratory Medicine and4Department of Medicine, School of Medicine, University of Pennsylvania,5Jonathan Lax
Treatment Center, Philadelphia FIGHT, and6Department of Medicine, Drexel University, Philadelphia, Pennsylvania;7Blood Systems Research Institute
and7aDepartment of Laboratory Medicine and8Department of Medicine, University of California–San Francisco, San Francisco, California; and
9Laboratory of Experimental Immunology, AIC Frederick, NCI Frederick, Frederick, Maryland and9aRagon Institute of MGH, MIT and Harvard, Boston,
(See the editorial commentary by McNamara and Collins, on pages 201–3.)
the immune system to spontaneously control human immunodeficiency virus (HIV) replication.
Methods.A total of 23 HIV type 1 (HIV-1)–infected, virologically suppressed subjects receiving ART (CD4+
T-cell count, >450 cells/μL) were randomly assigned to have 180 μg/week (for arm A) or 90 μg/week (for arm B)
of pegylated (Peg) interferon alfa-2a added to their current ART regimen. After 5 weeks, ART was interrupted,
and Peg–interferon alfa-2a was continued for up to 12 weeks (the primary end point), with an option to continue
to 24 weeks. End points included virologic failure (viral load, ≥400 copies/mL) and adverse events. Residual viral
load and HIV-1 DNA integration were also assessed.
Results.At week 12 of Peg–interferon alfa-2a monotherapy, viral suppression was observed in 9 of 20 subjects
(45%), a significantly greater proportion than expected (arm A, P=.0088; arm B, P= .0010; combined arms,
P<.0001). Over 24 weeks, both arms had lower proportions of subjects who had viral load, compared with the
proportion of subjects in a historical control group (arm A, P=.0046; arm B, P= .0011). Subjects who had a
sustained viral load of <400 copies/mL had decreased levels of integrated HIV DNA (P= .0313) but increased
residual viral loads (P=.0078), compared with subjects who experienced end-point failure.
Conclusions.Peg–interferon alfa-2a immunotherapy resulted in control of HIV replication and decreased
HIV-1 integration, supporting a role for immunomediated approaches in HIV suppression and/or eradication.
Clinical Trials Registration. NCT00594880.
Antiretroviral therapy (ART)–mediated immune reconstitution fails to restore the capacity of
Keywords.HIV-1; interferon-alpha; viral integration; immunotherapy.
The quest to effect long-term control of human immu-
nodeficiency virus type 1 (HIV-1) in the absence of
antiretroviral therapy (ART) has led to numerous ther-
apeutic approaches aimed at increasing host-mediated
control of HIV and/or clearance of latent virus reser-
voirs , while maintaining the beneficial effects of
immune reconstitution. Pilot strategies currently under
investigation include gene therapy , therapeutic vac-
cines , cytokines , and chemotherapy . Despite
Received 25 April 2012; accepted 30 July 2012; electronically published 26
Presented in part: 19th Conference on Retroviruses and Opportunistic Infec-
tions, Seattle, Washington, 5–8 March 2012. Abstract 631.
Correspondence: Luis J. Montaner, DVM, DPhil, MSc, The Wistar Institute,
3601 Spruce St, Philadelphia, PA (firstname.lastname@example.org).
The Journal of Infectious Diseases2013;207:213–22
© The Author 2012. Published by Oxford University Press on behalf of the Infectious
Diseases Society of America. All rights reserved. For Permissions, please e-mail:
Control of HIV-1 With Peg–Interferon Alfa-2a • JID 2013:207 (15 January) • 213
intensive investigation, no strategy so far has resulted in
sustained control of HIV in the absence of antiretroviral
Interferon α belongs to a family of type 1 interferons
produced by dendritic and others cells as part of the host’s
Toll-like receptor–mediated antiviral response . While in-
terferon-mediated gene expression is increased in advanced
HIV disease , the role of this innate response (ie, viral
control mechanism vs chronic activation mediator contribut-
ing to disease progression) remains under debate [8–10]. The
results of several clinical trials support a predominantly antivi-
ral activity (ie, approximately 0.5 log decrease in plasma viral
load) when interferon alfa is administered to HIV-1–infected
persons in the absence of ART [11–14]. However, in this
setting interferon alfa is not completely suppressive, possibly
because of the deterioration of immune system effectors due
to the ongoing viral replication. The degree to which interfer-
on alfa monotherapy may contribute to virus control (eg, sup-
pression of residual replication) after ART-mediated immune
reconstitution has not been tested.
Given the growing interest in identifying novel strategies
aimed at controlling HIV in the absence of ART, we sought to
establish a proof of concept that interferon alfa can suppress
HIV replication in subjects in whom the detrimental effects of
uncontrolled HIV replication on immune function have been
partially reversed by ART.
Written informed consent was obtained from patients accord-
ing to the directives of the institutional review boards at the
Wistar Institute, University of Pennsylvania, Philadelphia
FIGHT, and Drexel University. All human experimentation
was conducted in accordance with the guidelines of the US
Department of Health and Human Services and those of the
authors’ institutions. The study was registered at Clinical-
Trials.gov (http://www.clinicaltrials.gov/; NCT00594880).
A schematic representation of the study is provided in
Figure 1A, together with a CONSORT (Consolidated Stan-
dards of Reporting Trials) study subject disposition chart
(Figure 1B). A total of 23 eligible subjects (all were in stable
health, were receiving ART, had a plasma HIV RNA load of
<50 copies/mL, had a current CD4+T-cell count of >450 cells/
μL, and had a nadir CD4+T-cell count of >200 cells/μL) were
randomized. Exclusion criteria included hepatitis C virus coin-
fection, active hepatitis B virus infection, a history of major
depression or autoimmune diseases, a Framingham score of
>15% (for men) or >10% (for women) for a 10-year risk for
myocardial infarction, and retinal clouds on fundoscopic
After an 8-week observation period, subjects were randomly
assigned to receive either 180 μg/week (for arm A) or 90 μg/
week (for arm B) of pegylated (Peg) interferon alfa-2a
(Pegasys; Roche) in an open-label manner. After the first 5
weeks of treatment (study week 13), subjects had their ART
regiment interrupted and continued to receive Peg–interferon
The prespecified primary end point was the proportion of
subjects who maintained a viral load of <400 copies/mL after
12 weeks of Peg–interferon alfa-2a monotherapy (study week
25). Primary end point failure was defined as 2 consecutive
viral load measurements of ≥400 copies/mL, as a decrease in
CD4+T-cell count to <300 cells/μL, or as the presence of sig-
nificant adverse events, including Patient Health Question-
naire 2/9 depression scores of >10. Subjects with a sustained
viral load of <400 copies/mL at the primary end point were
allowed to continue Peg–interferon alfa-2a monotherapy for
an additional 12 weeks (study week 37). All subjects restarted
their prestudy ART regimen at the time of virologic rebound
or treatment failure or by study week 37 and were then fol-
lowed for an additional 12-week period.
Clinical Laboratory Testing and Immunoassays
All clinical laboratory tests were performed by Quest Diagnos-
tics (Madison, NJ). Interferon alfa levels were assessed in cryo-
preserved plasma, using a cytometric bead array (Invitrogen,
Carlsbad, CA) and were analyzed on a Luminex (Austin, TX)
platform at the Center for AIDS Research Immunology Core
at the University of Pennsylvania.
Ultrasensitive Amplification of HIV RNA in Plasma
Longitudinal plasma HIV RNA levels were measured using
the Food and Drug Administration–approved isothermal
transcription-mediated amplification assay (Aptima, Gen-Probe,
San Diego, CA). This assay has a 50% detection limit of 3.6
RNA copies/mL when performed singly . Data were col-
lected in 5 replicates for each time point (1.5–2.5 mL plasma
total), improving the overall limit of detection to <3.5 RNA
copies/mL; means are reported. The output for each replicate
is a signal/cutoff (S/Co) ratio (range, 0–30), with S/Co of <1.0
considered to be HIV RNA negative. S/Co levels were derived
by averaging the S/Co values across replicates [16, 17].
Measures of Integrated HIV DNA
The integration assay was performed on peripheral blood mono-
nuclear cells (PBMCs) as described in detail previously .
Briefly, Alu-gag–based polymerase chain reaction with repeti-
tive sampling techniques was used to increase the sensitivity
and accuracy of the assay in order to measure low levels of
integration in patient samples. Our integration standard
accounts for the effect of distance between integration sites
and host-Alu sequences. Results are expressed as number of
214 • JID 2013:207 (15 January) • Azzoni et al
integrated HIV DNA copies per number of CD4+T cells,
as follows: [number of copies of integrated HIV DNA
per PBMC]× [(monocyte count +lymphocyte count)/(CD4+
A total of 2× 106PBMCs were frozen in DNAzol (Molecular
Research Center, Cincinnati, OH), and DNA was extracted as
per the manufacturer’s protocol. HLA class I loci were geno-
typed by the sequence-based typing method, as recommended
by the 13th International Histocompatibility Workshop (http://
(IL-28B) gene alleles were genotyped for the C or T allele at
rs12979860, using a custom-designed TaqMan allelic discrimi-
nation assay (Applied Biosystems), as described elsewhere .
Subjects were randomly allocated 1:1 to receive either 90 or
180 μg/week of Peg–interferon alfa-2a with random block
over 49 weeks. Treatment is represented by dark grey boxes (pegylated [Peg] interferon alfa-2a at 180 or 90 μg/week), light grey boxes (antiretroviral
therapy [ART], as previously treated), or white boxes (ART interruption, Peg–interferon alfa-2a monotherapy). B, CONSORT (Consolidated Standards of
Reporting Trials) flow diagram of the study subject disposition. PHQ2/9, Patient Health Questionnaire 2/9.
Trial schema and subject disposition. A, Schema of the trial visit and relative activities (top row) during the 14 visits scheduled to occur
Control of HIV-1 With Peg–Interferon Alfa-2a • JID 2013:207 (15 January) • 215
sizes. A Web-based system (http://www.CEDTweb.org) was
used to sequentially assign randomization outcomes.
We evaluated whether immunotherapy with 180 or 90 μg/week
of Peg–interferon alfa-2a maintained control of HIV replica-
tion (ie, for 12 weeks) in the absence of antiretroviral therapy
to greater levels than observed in the absence of immunomod-
ulatory therapy. On the basis of our previous studies, we ex-
pected that, at most, 9% of subjects would maintain a viral
load of <400 copies/mL for 12 weeks in the absence of any
immunomodulatory therapy [20–22]. A relatively high thresh-
old of ≥400 copies/mL was used to define virologic failure in
an effort to avoid classifying patients with transient viral
blips as having virologic failure; such viral blips have been de-
scribed for subjects receiving ART . The primary analysis
tested the null hypothesis that the proportion of individuals
with a viral load of<400 copies/mL is equal to 0.09 against
the 1-sided alternative that this proportion is >0.09, using a
1-sided exact binomial test with an α level of .05. The analysis
was initially stratified by arm and then performed with com-
bined arms. Subjects who withdrew consent or were lost to
follow-up (n = 3) were excluded from the primary analysis; a
sensitivity analysis including all subjects was also performed.
We also evaluated whether immunotherapy with 180 or
90 μg/week of Peg–interferon alfa-2a maintained control of
HIV replication at a viral load of <48 copies/mL (ie, for 12
weeks) in the absence of antiretroviral therapy, using the same
approach as with the primary end point analysis described
above and with an expected suppression rate of 3% observed
in historical control subjects from a structured-treatment-
interruption cohort in Philadelphia [20–22].
To determine control over time and on the basis of our pre-
vious studies, the proportion of subjects maintaining viral
suppression at a viral load of <400 copies/mL over time was il-
lustrated using a Kaplan-Meier plot and differences between
study arms and the historical cohort  (7% female, 28%
African American, and 7% Hispanic; mean [±SD] CD4+T-cell
count before interruption, 677±224 cells/μL) were tested using
a log-rank test. To address potential differences in CD4+T-cell
count distributions between cohorts, we fit a Cox proportional
hazard model that controlled for CD4+T-cell count at the time
of interruption. Differences in medians between visits for the
reported variables were assessed using the Wilcoxon signed
rank test or the Fisher exact test. Differences in the levels of
HIV integration between visits were tested using Wilcoxon
signed rank tests, and differences in end point/baseline ratios
between arms were tested using a Wilcoxon rank sum test.
A total of 23 study participants were randomly assigned to
one of the study arms in the 49-week study (Figure 1): 12
were assigned to arm A (180 μg/week), and 11 were assigned
to arm B (90 μg/week). The study population included 19
men (83%) and 14 African Americans (61%) and had a
median age of 45 years (interquartile range [IQR], 40–49
years). Across both arms, the median baseline CD4+T-cell
count was 840 cells/μL (IQR, 631–1112 cells/μL). Subject base-
line characteristics are described in Table 1. No significant dif-
ference between arms was observed for baseline variables.
Polymorphisms in genes encoding for IL-28B and HLA
alleles have been associated with the ability to respond to in-
terferon alfa-based therapy during hepatitis C virus infection
, to control HIV replication in the absence of ART , or
to control HIV disease progression [25–29]. We did not detect
any obvious enrichment in alleles analyzed in the subjects
with sustained viral control (Table S1).
Receive One of Two Dosages of Pegylated Interferon Alfa-2a
Baseline Characteristics of Subjects With Human Immunodeficiency virus Infection Who Were Randomly Assigned to
Study ArmPrimary Outcome
A (180 μg/week)
B (90 μg/week)
CD4+T-cell count (cells/μL)
Data are median (interquartile range) or no. (%) of subjects.
216 • JID 2013:207 (15 January) • Azzoni et al
Peg–interferon alfa-2a treatment can result in a number of
known adverse events, among which the most prominent are
depression, lymphocytopenia, neutropenia and liver toxicities.
Three subjects had a Patient Health Questionnaire 2/9 score of
>10, indicating moderate depression, and Peg–interferon alfa-
2a therapy was discontinued. Although no subject had to dis-
continue the treatment because of a decrease in CD4+T-cell
count (Figure 2), an initial and expected drop during the first
5 weeks of Peg–interferon alfa-2a treatment was noted
(median CD4+T-cell count at week eight, 840 cells/μL [IQR,
630.5–1112]; median CD4+T-cell count at week thirteen, 711
cells/μL [IQR, 558–1020]); P= .0004). These levels remained
stable over the subsequent monotherapy period. There was no
relationship between initial CD4+T-cell count decline and
dose of Peg–interferon alfa-2a (P= .12). One subject experi-
enced grade 3 neutropenia while still receiving ART, resulting
in study discontinuation.
Of the 23 subjects enrolled, 2 withdrew from the study before
the end point (one was incarcerated, and one withdrew out of
concern with protracted grade 2 liver function test results), and
1 was lost to follow-up after week 4, resulting in a sample size
of 20 for the primary analysis. During the postrandomization
period of Peg–interferon alfa-2a administration, 7 individuals
exhibited virologic failure, 3 had moderate depression scores
as revealed by the Patient Health Questionnaire 2/9, and 1 had
a grade 3 neutropenia; all of these subjects were considered to
have achieved end-point failure. Of the 20 subjects with end
points by week 12 of Peg–interferon alfa-2a monotherapy
(study week 25), 9 subjects (45%) successfully maintained a
viral load of <400 copies/mL. The proportion of individuals
with a viral load of <400 copies/mL was greater than the antic-
ipated proportion of 9% in either arm or combined (P=.0088
for arm A, P=.0010 for arm B, and P<.0001 for the combined
arms, using the exact binomial test). As shown in Table 2, the
proportion of subjects who achieved successful outcomes was
not significantly different between the 2 arms (P=1, by the
Fisher exact test). We performed a sensitivity analysis that in-
cluded all 23 randomized subjects, assigning a failure outcome
to those who withdrew/were lost to follow-up: the exact bino-
mial test P values were .0180 for arm A, .0017 for arm B, and
<.0001 for the combined arms. Secondary analysis at the
primary end point, using the <48 copies/mL threshold
achieved by 4 of 20 subjects (20%), showed significant differ-
ences in both arms, compared with an anticipated rate of 3%
based on reports in the literature (P=.03 for arm A, P=.0028
for arm B, and P<.0027 for the combined arms).
arm B (open squares). Treatment is represented by black (antiretroviral therapy [ART]) or white horizontal boxes (Peg–interferon alfa-2a). The left panel
represents visits up to failure or study end point. The right panel represents visits after ART resumption.
CD4+T-cell count response to Peg–interferon alfa-2a. CD4+T cell count was assessed at each study visit for arm A (closed circles) and
Immunodeficiency virus Infection Who Were Randomly Assigned
to Receive One of Two Dosages of Pegylated Interferon Alfa-2a
Primary End Point Results for Subjects With Human
Study Arm, No. (%)
Overall, No. (%)Outcome A (180 μg/week)B (90 μg/week)
P= 1, by the Fisher exact test, for differences in failure and success rates
between arms A and B.
Control of HIV-1 With Peg–Interferon Alfa-2a • JID 2013:207 (15 January) • 217
Plasma interferon alfa-2a concentrations were higher than
baseline values in most subjects, with levels at week 5 slightly
higher in arm A (P=.0782; data not shown) and no signifi-
cant difference between subjects with treatment success or
failure outcomes, suggesting that either dose tested can
achieve drug levels associated with suppression.
Assessment of Plasma HIV Levels During 24-Week Peg–
Interferon Alfa-2a Monotherapy
The 9 subjects with HIV levels controlled to <400 copies/mL at
12 weeks on Peg–interferon alfa-2a monotherapy (4 of 9 had a
sustained viral load of <48 copies/mL) were allowed to continue
to receive monotherapy for another 12 weeks. One subject with-
drew consent after week 12, resulting in 8 subjects followed for
24 weeks. As illustrated in Figure 3, 6 of these 8 subjects main-
tained a viral load of <400 copies/mL for the entire period (3 of
6 with a viral load of <48 copies/mL). A log-rank test per-
formed using data over the 24-week period revealed that the
rate in each arm of maintaining viral suppression was signifi-
cantly different than that of historical controls (P=.0046 for
arm A and P=.0011 for arm B, compared with an ART-inter-
ruption arm of historical control subjects ); accounting for
baseline CD4+T-cell count did not alter these findings.
Detection of Residual HIV-1 RNA
We measured plasma samples from subjects with a viral load
of <400 copies/mL, using an ultrasensitive polymerase chain
reaction technique able to detect 1–100 HIV RNA copies/mL
[16, 30] (Table 3). First, we determined the effects of 5 weeks
of Peg–interferon alfa-2a added to a suppressive ART regimen
(viral load, <400 copies/mL at both time points) from all avail-
able subjects (n= 20).
Residual HIV plasma levels during ART (median S/Co
value at week eight, 3.3 [IQR, 0.15–7.9]) were not significantly
jects With Human Immunodeficiency virus Infection, by Primary Outcome
Transcription-Mediated Amplification (TMA)–Based Assessment of Viral Replication During the Study Period Among Sub-
48 1325 37
PCR viral load (RNA copies/mL)
4813 Failure visit
PCR viral load (RNA copies/mL)
Data are median (interquartile range) or no. of subjects.
Abbreviations: PCR, polymerase chain reaction; S/Co, signal/cutoff ratio.
aTMA was not done for individuals with a viral load of >400 copies/mL.
dOne individual had a viral load of 486 copies/mL, but this was not confirmed because the individual withdrew from the study immediately after week 4.
was assessed at each visit and is represented as the Kaplan–Meier plot
of the viral response to Peg–interferon alfa-2a treatment. The vertical
axis represents the fraction of subjects in arm A (solid line) and arm B
(dotted line) with a VL of <400 RNA copies/mL. The dashed line repre-
sent the comparator historical structured treatment interruption (STI)
group . Abbreviation: ART, antiretroviral therapy.
Viral load (VL) response to Peg–interferon alfa-2a. The VL
218 • JID 2013:207 (15 January) • Azzoni et al
different after 5 weeks of Peg–interferon alfa-2a (median S/Co
value at week thirteen, 2.4 [IQR, 0.1–4.95]; P= .5619).
Second, we sought to determine whether a change in resid-
ual viral load occurred in responders (<400 copies/mL) from
the time of ART interruption to 12 or 24 weeks of Peg–
interferon alfa-2a monotherapy. The 9 responder subjects
showed an increase in residual HIV RNA levels, from a
median S/Co value of 2.6 (IQR, 1.4–8) before ART interrup-
tion (week 13) to a median S/Co value of 20.9 (IQR, 8.9–30.2)
at the primary end point (week 25; P= .0078, by the Wilcoxon
signed rank test). No change was detected between 12 weeks
(S/Co value at week twenty-five, 9.2 [IQR, 4.9–26.5]) and 24
weeks (S/Co value at week thirty-seven, 22.4 [IQR, 6.1–31.5];
P=.625) after ART interruption among the 6 subjects whose
viral load remained at <400 copies/mL for 24 weeks despite
not receiving ART.
Evaluation of Integrated HIV DNA Copies Per CD4+T Cell in
Association With Treatment Success
HIV persists during ART as integrated DNA in memory T
cells and perhaps other cells. We compared the amount of in-
tegrated viral DNA present in PBMCs from 6 subjects who
exhibited virologic failure and 7 who maintained viral sup-
pression at week 12 (representing those subjects with primary
end point success and with a sufficient PBMC yield to com-
plete the analysis; Figure 4). Samples were studied at weeks 8,
13, and 25. The number of integrated HIV DNA copies per
CD4+T cell (Figure 4A) did not change significantly between
baseline and the time of virologic failure in the 6 subjects who
had virologic failure. In contrast, the number of integrated
HIV DNA copies per CD4+T cell between baseline and the
primary end point declined significantly in the 7 subjects who
exhibited durable viral control (mean baseline value [±SD],
7.83 × 10−4± 4.08 × 10−4copies/CD4+T cell; mean end point
value [±SD], 2.92 × 10−4± 1.19 ×10−4copies/CD4+T cell;
P= .0313). The ratio of integrated HIV DNA copies/CD4+
T cell between end point and baseline was also significan-
tly lower in subjects sustaining viral control (Figure 4B;
We report that administration of Peg–interferon alfa-2a (90 or
180 μg/week) to individuals with an ART-suppressed HIV
RNA load results in a sustained control of viral replication in
45% of subjects when ART is interrupted. The observed re-
sponse rate was significantly greater than that reported in trials
evaluating ART interruption alone and remained significant
when including all randomized subjects (39% suppression).
Prior approaches based on interferon alfa administration
without ART (ie, in treatment-naive subjects or subjects with
ART interruption) have failed to achieve similar suppression
rates [12, 14]. Differences in study design might have contrib-
uted to this disparity. We started Peg–interferon alfa-2a treat-
ment in subjects receiving suppressive ART (viral load, <50
reaction, as described in the Methods section. The number of integrated HIV DNA copies/CD4+T cell are represented. The left panel shows 6 subjects
experiencing protocol failure (viral load, ≥400 HIV RNA copies/mL) before the 12-week end point. Assessed are protocol week 8 (antiretroviral therapy
[ART]), week 13 (ART+Peg–interferon alfa-2a), and after failure (after ART resumption). The right panel shows 7 subjects experiencing protocol-defined
success (a sustained viral load of <400 copies/mL until the 12-week end point). Assessed are protocol week 8 (ART), week 13 (ART +Peg–interferon
alfa-2a; subject 1=not done), and week 25 (Peg–interferon alfa-2a monotherapy), as indicated. P values calculated by the Wilcoxon signed rank test
are indicated. B, Boxes represent the median, quartiles, and extremes of the distribution of end point:baseline ratios of integrated DNA copies/CD4+T
cell in subjects with virological failure or success. Individual values are superimposed as open circles. A P value calculated by the Wilcoxon rank sum
test is indicated.
Assessment of integrated HIV DNA levels per circulating CD4+T cell. A, Integrated HIV DNA was measured by Alu–polymerase chain
Control of HIV-1 With Peg–Interferon Alfa-2a • JID 2013:207 (15 January) • 219
copies/mL) and with high degree of immune reconstitution
(CD4+T-cell count >450 cells/μL), which likely resulted in re-
stored immune subsets (eg, natural killer [NK] cells and CD8+
T cells) and activation of antiviral host factors (eg, apolipopro-
tein-B messenger RNA editing enzyme [APOBEC]), mediating
some of the antiviral effects of type 1 interferons. Moreover,
unlike prior studies [12, 13], Peg–interferon alfa-2a was coad-
ministered with ART for 5 weeks, potentially allowing for
steady-state drug levels and immunomodulation to be
achieved before ART was interrupted.
The mechanisms by which Peg–interferon alfa-2a restricted
viral replication in our subjects remain to be elucidated. Type
1 interferons control viremia by reducing viral replication in
infected cells and preventing infection of new targets. The
mechanisms of type 1 interferon–dependent viral control fall
into 3 areas: direct interferon α anti-HIV activity, mediated by
host proteins (eg, tetherin, APOBEC 1, and protein kinase R
activation [31, 32]); enhancement of adaptive effector function
(eg, HIV-specific CD8+ T cells and ADCC); and enhancement
of innate immune effector function (eg, NK cells).
A role for interferon-mediated immune clearance of HIV-
infected cells is supported by clinical studies in which the ad-
ministration of interferon alfa to patients with HIV infection
or melanoma was associated with increased perforin and gran-
zyme expression in NK and CD8+T cells, respectively [33,
34]. While several studies suggest that the majority of integrat-
ed HIV DNA is defective [1, 35, 36], latently infected cells
may still express HIV protein and be subject to immune
We report a significant reduction of CD4+ T cell–integrated
HIV DNA in subjects with viral control. Our therapeutic ap-
proach (ie, ART +Peg–interferon alfa-2a therapy, followed by
Peg–interferon alfa-2a alone) may have contributed to this de-
crease by allowing residual HIV replication to trigger immune
responses that are restored by long-term ART and that are
capable of clearing infected cells expressing HIV proteins. The
implications of this observation remains unclear, but one hy-
pothesis is that Peg–interferon alfa-2a resulted in an immuno-
mediated reduction in the size of the cellular latent reservoir
(as defined by the level of integrated HIV DNA). As indicated
by the transcription-mediated amplification-based assessment,
residual viremia or enhanced viral release occurred when ART
was stopped. This may have contributed to the amplification
of cytotoxic T-lymphocyte responses, which are reduced in
frequency in ART-treated subjects as compared to long-term
nonprogressors . Regardless of the mechanism, our data
suggest that detection of residual HIV in plasma can be disas-
sociated from changes in HIV integration levels within periph-
eral cells. Further studies in larger populations will be required
to establish whether ART interruption and residual viremia
are necessary to produce the observed reduction in viral
It will be important for future studies to determine whether
Peg–interferon alfa-2a treatment can stabilize the viral set
point to these low levels, sustaining viral control over time. Of
the 8 subjects who elected to continue monotherapy past the
primary end point to week 24 (9 subjects were eligible), 6 had
a viral load of <400 copies/mL at week 48 (3 of 8 had a viral
load of <48 copies/mL), supporting the view that in some in-
dividuals interferon alfa-mediated immune control may be ex-
tended beyond 24 weeks. Interestingly, 2 of the 6 subjects with
a viral load of <400 copies had a stable set point between 50
and 400 copies/mL throughout the 24 weeks of monotherapy.
We did not observe any rise in integrated HIV DNA levels
in patients in whom the viral load was not controlled to <400
copies/mL. This was unexpected and may indicate delayed ki-
netics of reservoir change after ART interruption and viremic
rebound; importantly, in our study the viremic episodes were
of short duration, since our patients restarted ART as soon as
their viral load was confirmed to be <400 copies/mL. Our data
suggest that future studies aimed at determining the kinetics
of CD4+T cell–integrated HIV DNA rebound may require
longer periods of ART interruption with sustained viremia.
Overall, our observations are consistent with interferon alfa
contributing to long-term control despite residual viral repli-
cation, as observed in “elite” suppressors (ie, individuals sus-
taining low-level viremia without ART [16, 17]). Notably, the
latter individuals also preserve functional circulating plasma-
cytoid dendritic cells better than chronic progressors do [38,
39], and they have low levels of viral integration , support-
ing the hypothesis that type 1 interferon–mediated mecha-
nisms contribute to HIV control in vivo.
Our study has some limitations. First, the original design
(including a period of ART interruption without Peg–interferon
alfa-2a for each subject, providing for the determination of in-
dividual viral set points) was modified at the request of the
Food and Drug Administration in the course of the Investiga-
tional New Drug submission, on the basis of the results of the
SMART study , which showed that CD4+ T cell–guided
sexually transmitted infections are associated with disease pro-
gression, and on the basis of extensive literature reporting
viral rebound following ART interruption in >95% of subjects.
To address this limitation, we compared the proportion of fail-
ures in each arm with an expected rate obtained from prior
studies [20–22, 40–43], and we elected to use an estimated
proportion of subjects with sustained viral suppression (9%)
that was higher than the proportions reported in similar pop-
ulations. In a secondary analysis, we also compared (using a
log-rank test) each study arm to historical cohort of control
subjects for whom ART was interrupted, confirming that the
2 arms had significantly higher rates of suppression than the
ART-interruption group, even after adjustment for initial
CD4+T-cell levels. A second limitation is the unavailability of
many pre-ART viral loads (most subjects received long-time
220 • JID 2013:207 (15 January) • Azzoni et al
ART from multiple providers), which prevented us from di-
rectly assessing the relationship between viral set points before
ART and those during Peg–interferon alfa-2a monotherapy.
However, we did establish that our study group did not
include an overrepresentation of individuals with HLA and
KIR alleles associated with HIV control. Finally, because our
study cohort represents a proof-of-concept study with a
limited sample size, it will be important to confirm our results
in future, larger longitudinal studies.
In conclusion, we report that treatment with 90 or 180 μg/
week of Peg–interferon α2a can support viral control and re-
duction of peripheral HIV integration levels in subjects for
whom ART has been interrupted. Our study provides a proof
of concept that immunotherapy can be pursued in HIV-
infected, ART-dependent subjects to reach a status of viral
control beyond ART and could complement current research
approaches to current “functional cure” and eradication.
Supplementary materials are available at The Journal of Infectious Diseases
online (http://jid.oxfordjournals.org/). Supplementary materials consist of
data provided by the author that are published to benefit the reader. The
posted materials are not copyedited. The contents of all supplementary
data are the sole responsibility of the authors. Questions or messages
regarding errors should be addressed to the author.
of study participants and to their providers, who contributed to overall
study oversight. We acknowledge clinical recruitment support provided by
Jane Shull and the Philadelphia FIGHT staff (A. Kapalko, K. Richards,
S. Smith), Drexel University (S. Lewis, D. Downie, C. M. Randazzo,
A. Johnson, S. Skinner), and the University of Pennsylvania (C. Carty,
K. Maffei, J. Quinn, J. Gilmore, J. Hines, Z. Dorey-Stein). We are grateful
to Dr J. Chehimi, for initial study support, and to Drs K. Brady, J. Hines,
E. Hollen, S. Hansen-Flaschen, and R. Maniglia, for patient referrals.
Pharmacy support was provided by the University of Pennsylvania
(K. Rockwell, D. Kim) and Drexel University (K. George). We thank the
data safety monitoring board members K. Squires, MD; J. Merz, PhD;
G. Bisson, MD; A. Troxel, PhD; S. Bellamy, PhD; R. Goldfein, Esq; and
G. Brake. We acknowledge statistical analysis support (X. Yin); Web devel-
opment support for data collection (M. Picone); ultrasensitive amplifica-
tion of HIV RNA (V. Winkelman at Creative Testing Solutions, Temple
AZ); National Institutes of Health program support
M. Dehlinger, E. Pouilot); genotyping support (F.-M. Dui, M. P. Martin),
administrative support for institutional review board, data safety monitor-
ing board, National Institutes of Health, and Food and Drug Administra-
tion reporting (B. O’Brien, M. O’Neill, J. Dubin); sample/subject transport
support (J. Starnes, B. Fisher); and laboratory support (G. Reynolds,
B. Ross, A. Mackiewicz, N. Opsitnick, C. Calloway, M. Cummins,
L. Hubbard, A. Quartlebaum).
Author contributions are as follows: Livio Azzoni, study design, data
management, analysis, and manuscript preparation; Andrea S. Foulkes,
study design, analysis, and manuscript preparation; Emmanouil Papasav-
vas, study management, analysis, and manuscript preparation; Angela
M. Mexas, integrated DNA assays, data analysis, and manuscript prepara-
tion; Kenneth M. Lynn, clinical care, study coordination, analysis, and
manuscript preparation; Michael Busch, transcription-mediated amplifica-
tion assays,analysis,and manuscript preparation;
We are grateful for the altruism and determination
transcription-mediated amplification assays, analysis, and manuscript
preparation; Mary Carrington, HLA haplotypes, analysis, and manuscript
preparation; Karam Mounzer, clinical care, study coordination, analysis,
and manuscript preparation; Pablo Tebas, clinical care, study coordination,
analysis, and manuscript preparation; Ian Frank, clinical care, study coor-
dination, analysis, and manuscript preparation; Una O’Doherty, integrated
M. Jacobson, clinical care, study coordination, analysis, and manuscript
preparation; Jay Kostman, clinical principal investigator, study coordina-
tion, analysis, and manuscript preparation; and Luis J. Montaner, principal
investigator, study design and management, data collection, analysis, and
The content of this publication does not necessarily
reflect the views or policies of the Department of Health and Human Ser-
vices, nor does mention of trade names, commercial products, or organi-
zations imply endorsement by the US government.
This work was supported by NIH/NIAID grant
U01AI065279 to LJM, K08AI073102 to AM, K02 AI078766 and R21
AI087461 to UO. Additional support was provided by Genentech/Roche
(drug support under protocol PEG224), The Philadelphia Foundation
(Robert I. Jacobs Fund), Henry S. Miller, Jr. and J. Kenneth Nimblett,
AIDS funds from the Commonwealth of Pennsylvania and from the Com-
monwealth Universal Research Enhancement Program, Pennsylvania De-
partment of Health, the Penn Center for AIDS Research (P30 AI 045008),
and Cancer Center Grant (P30 CA10815), the American Foundation for
AIDS Research, the National Cancer Institute, National Institutes of
Health (Contract No. HHSN261200800001E), and the Intramural Re-
search Program of the NIH, National Cancer Institute, Center for Cancer
Potential conflicts of interest.
All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential
Conflicts of Interest. Conflicts that the editors consider relevant to the
content of the manuscript have been disclosed.
1. Chun TW, Carruth L, Finzi D, et al. Quantification of latent tissue
reservoirs and total body viral load in HIV-1 infection. Nature 1997;
2. Wilen CB, Wang J, Tilton JC, et al. Engineering HIV-resistant human
CD4+ T cells with CXCR4-specific zinc-finger nucleases. PLoS Pathog
3. Angel JB, Routy JP, Tremblay C, et al. A randomized controlled trial
of HIV therapeutic vaccination using ALVAC with or without
Remune. AIDS 2011; 25:731–9.
4. Henry K, Katzenstein D, Cherng DW, et al. A pilot study evaluating
time to CD4 T-cell count <350 cells/mm(3) after treatment interrup-
tion following antiretroviral therapy +/- interleukin 2: results of
ACTG A5102. J Acquir Immune Defic Syndr 2006; 42:140–8.
5. Matalon S, Rasmussen TA, Dinarello CA. Histone deacetylase inhibi-
tors for purging HIV-1 from the latent reservoir. Mol Med 2011;
6. Chang JJ, Altfeld M. Innate immune activation in primary HIV-1
infection. J Infect Dis 2011; 202(Suppl. 2):S297–301.
7. Rotger M, Dang KK, Fellay J, et al. Genome-wide mRNA expression
correlates of viral control in CD4+ T-cells from HIV-1-infected indi-
viduals. PLoS Pathog 2010; 6:e1000781.
8. Chehimi J, Papasavvas E, Tomescu C, et al. Inability of plasmacytoid
dendritic cells to directly lyse HIV-infected autologous CD4+ T cells
despite induction of tumor necrosis factor-related apoptosis-inducing
ligand. J Virol 2010; 84:2762–73.
9. Tomescu C, Chehimi J, Maino VC, Montaner LJ. NK cell lysis of
HIV-1-infected autologous CD4 primary T cells: requirement for IFN-
mediated NK activation by plasmacytoid dendritic cells. J Immunol
10. Boasso A, Shearer GM. Chronic innate immune activation as a cause
of HIV-1 immunopathogenesis. Clin Immunol 2008; 126:235–42.
Control of HIV-1 With Peg–Interferon Alfa-2a • JID 2013:207 (15 January) • 221
11. Lane HC, Kovacs JA, Feinberg J, et al. Anti-retroviral effects of inter- Download full-text
feron-alpha in AIDS-associated Kaposi’s sarcoma. Lancet 1988; 2:
12. Boue F, Reynes J, Rouzioux C, et al. Alpha interferon administration
during structured interruptions of combination antiretroviral therapy
in patients with chronic HIV-1 infection: INTERVAC ANRS 105 trial.
AIDS 2011; 25:115–8.
13. Dianzani F, Rozera G, Abbate I, et al. Interferon may prevent HIV
viral rebound after HAART interruption in HIV patients. J Interferon
Cytokine Res 2008; 28:1–3.
14. Asmuth DM, Murphy RL, Rosenkranz SL, et al. Safety, tolerability,
and mechanisms of antiretroviral activity of pegylated interferon alfa-
2a in HIV-1-monoinfected participants: a phase II clinical trial.
J Infect Dis 2010; 201:1686–96.
15. Linnen JM, Gilker JM, Menez A, et al. Sensitive detection of genetic
variants of HIV-1 and HCV with an HIV-1/HCV assay based on tran-
scription-mediated amplification. J Virol Methods 2002; 102:139–55.
16. Hatano H, Delwart EL, Norris PJ, et al. Evidence for persistent low-
level viremia in individuals who control human immunodeficiency
virus in the absence of antiretroviral therapy. J Virol 2009; 83:329–35.
17. Hatano H, Delwart EL, Norris PJ, et al. Evidence of persistent low-
level viremia in long-term HAART-suppressed, HIV-infected individ-
uals. AIDS 2010; 24:2535–9.
18. Graf EH, Mexas AM, Yu JJ, et al. Elite suppressors harbor low levels
of integrated HIV DNA and high levels of 2-LTR circular HIV DNA
compared to HIV+ patients on and off HAART. PLoS Pathog 2011; 7:
19. Ge D, Fellay J, Thompson AJ, et al. Genetic variation in IL28B pre-
dicts hepatitis C treatment-induced viral clearance. Nature 2009;
20. El-Sadr WM, Lundgren JD, Neaton JD, et al. CD4+ count-guided inter-
ruption of antiretroviral treatment. N Engl J Med 2006; 355:2283–96.
21. Papasavvas E, Kostman JR, Mounzer K, et al. Randomized, controlled
trial of therapy interruption in chronic HIV-1 infection. PLoS Med
22. Davey RT Jr, Bhat N, Yoder C, et al. HIV-1 and T cell dynamics after
interruption of highly active antiretroviral therapy (HAART) in pa-
tients with a history of sustained viral suppression. Proc Natl Acad Sci
U S A 1999; 96:15109–14.
23. Nettles RE, Kieffer TL, Kwon P, et al. Intermittent HIV-1 viremia
(blips) and drug resistance in patients receiving HAART. JAMA 2005;
24. Bashirova AA, Thomas R, Carrington M. HLA/KIR restraint of HIV:
surviving the fittest. Annu Rev Immunol 2011; 29:295–317.
25. McMichael AJ, Jones EY. First-class control of HIV-1. Science 2010;
26. Thomas R, Apps R, Qi Y, et al. HLA-C cell surface expression and
control of HIV/AIDS correlate with a variant upstream of HLA-C.
Nat Genet 2009; 41:1290–4.
27. Poropatich K, Sullivan DJ Jr. Human immunodeficiency virus type 1
long-term non-progressors: the viral, genetic and immunological basis
for disease non-progression. The Journal of General Virology 2011;
28. Kaur G, Mehra N. Genetic determinants of HIV-1 infection and pro-
gression to AIDS: immune response genes. Tissue Antigens 2009;
29. Leslie A, Matthews PC, Listgarten J, et al. Additive contribution of
HLA class I alleles in the immune control of HIV-1 infection. J Virol
30. Roland ME, Elbeik TA, Kahn JO, et al. HIV RNA testing in the
context of nonoccupational postexposure prophylaxis. J Infect Dis
31. Pillai SK, Abdel-Mohsen M, Guatelli J, et al. Role of retroviral restric-
tion factors in the interferon-alpha-mediated suppression of HIV-1 in
vivo. Proc Natl Acad Sci U S A 2012; 109:3035–40.
32. Goujon C, Malim MH. Characterization of the alpha interferon-
induced postentry block to HIV-1 infection in primary human macro-
phages and T cells. J Virol 2010; 84:9254–66.
33. Guillot B, Portales P, Thanh AD, et al. The expression of cytotoxic
mediators is altered in mononuclear cells of patients with melanoma and
increased by interferon-alpha treatment. Br J Dermatol 2005; 152:690–6.
34. Portales P, Reynes J, Pinet V, et al. Interferon-alpha restores HIV-
induced alteration of natural killer cell perforin expression in vivo.
Aids 2003; 17:495–504.
35. Chun TW, Stuyver L, Mizell SB, et al. Presence of an inducible HIV-1
latent reservoir during highly active antiretroviral therapy. Proc Natl
Acad Sci U S A 1997; 94:13193–7.
36. Chun TW, Engel D, Berrey MM, Shea T, Corey L, Fauci AS. Early
establishment of a pool of latently infected, resting CD4(+) T cells
during primary HIV-1 infection. Proc Natl Acad Sci U S A 1998;
37. Migueles SA, Osborne CM, Royce C, et al. Lytic granule loading of
CD8+ T cells is required for HIV-infected cell elimination associated
with immune control. Immunity 2008; 29:1009–21.
38. Soumelis V, Scott I, Gheyas F, et al. Depletion of circulating natural
type 1 interferon-producing cells in HIV-infected AIDS patients.
Blood 2001; 98:906–12.
39. Machmach K, Leal M, Gras C, et al. Plasmacytoid dendritic
cells reduce HIV production in Elite controllers. J Virol 2012; 86:
40. Autran B, Murphy RL, Costagliola D, et al. Greater viral rebound and
reduced time to resume antiretroviral therapy after therapeutic immu-
nization with the ALVAC-HIV vaccine (vCP1452). AIDS 2008;
41. Kilby JM, Bucy RP, Mildvan D, et al. A randomized, partially blinded
phase 2 trial of antiretroviral therapy, HIV-specific immunizations,
and interleukin-2 cycles to promote efficient control of viral replica-
tion (ACTG A5024). J Infect Dis 2006; 194:1672–6.
42. Neumann AU, Tubiana R, Calvez V, et al. HIV-1 rebound during in-
terruption of highly active antiretroviral therapy has no deleterious
effect on reinitiated treatment. Comet Study Group. AIDS 1999;
43. Steingrover R, Pogany K, Fernandez Garcia E, et al. HIV-1 viral
rebound dynamics after a single treatment interruption depends on
time of initiation of highly active antiretroviral therapy. AIDS 2008;
222 • JID 2013:207 (15 January) • Azzoni et al