Abstract

Structured treatment interruption (STI) may help to alleviate the problems associated with long-term antiretroviral therapy (ART) in human immunodeficiency virus (HIV)–infected patients. We analyzed the role that baseline levels of cytokines in plasma play as markers of a favorable outcome of STI. Two groups of patients were defined: STI responders and STI nonresponders. STI responders showed a higher baseline concentration of interleukin (IL)–15 in plasma than did STI nonresponders and showed lower levels of tumor necrosis factor (TNF)–α during STI. No differences were observed in levels of IL-2, IL-7, or interferon-α in plasma. Our data show that (1) levels of TNF-α in plasma correlate with HIV viremia and (2) monitoring baseline levels of IL-15 in plasma allows for the identification of a favorable outcome of STI

In a substantial proportion of human immunodeficiency virus (HIV)–infected individuals, antiretroviral therapy (ART) modifies the natural history of HIV infection, thereby producing remarkable benefits. However, most HIV-infected populations residing in developing countries have no access to continuous ART, and long-term therapy is limited by the need for strict daily adherence, occurrence of drug toxicity, elevated treatment costs, and sustained risks of emergence of drug resistance [1]. Structured treatment interruption (STI) may help to reduce patients’ reliance on ART. However, this strategy could be associated with the potential risk of HIV-mediated loss of CD4 T cells, severe constitutional symptoms (such as primary infection), and the emergence of viral mutations conferring drug resistance [2]. Thus, data to justify STI as a safe and effective approach remain controversial [3, 4]

In most individuals, STI produces rapid rebound of HIV viremia, which is restored to an undetectable level by highly active ART [5]. Nevertheless, some patients experience sustained ability to control viral replication, probably as a consequence of virological and/or immunological factors [3]. We recently have observed that most patients undergoing STI who experience a rapid rebound of HIV viremia have a sustained number of HIV-specific CD8 cells producing interferon (IFN)–γ, but the majority of these cells present a phenotype that is nonfunctionally cytotoxic [6]. This impaired function of cytotoxic T lymphocytes and the accumulation of preterminally differentiated CD8 T cells may be a consequence of an increased turnover and/or may be driven by an altered CD4 T cell–helper activity. We also have observed that Vγ9Vδ2 T cells are rapidly lost in patients undergoing STI who are unable to control viral replication [7]. Since this γδ T cell subset recognizes a broad range of microbial, nonpeptidic antigens, the impairment of Vγ9Vδ2 T cells may increase the risk of opportunistic infections [8]. These data underline the importance of a careful application of STI and suggest that the identification of immunological changes during STI is crucial for assessing its clinical value. In a recent study, Garcia et al. have shown that the level of memory CD4 T cells and the proliferative response to recall antigens may predict a good response to STI [9]. Similarly, in the present study, we investigated the role of levels of cytokines in plasma as markers of a favorable STI outcome

Patients, materials, and methodsWe examined the levels of cytokines in plasma from 25 HIV-1–infected patients who had undergone a single cycle of STI. This cohort of patients has been described elsewhere [6]. In brief, asymptomatic patients with chronic HIV-1 infection were recruited from the National Institute for Infectious Diseases “Lazzaro Spallanzani” of Rome. The criteria for inclusion in the study were the following: >2 years of successful ART, stable CD4 T cell counts (>500 cells/μL for 12 months before entry into the study), undetectable levels of HIV-1 RNA in plasma (<50 copies/mL, by branched DNA assay [bDNA version 3.0; Bayer Diagnostics], for >12 months before entry into the study), and the patient’s consent that ART could be discontinued for presence of side effects or by spontaneous request. Informed consent was obtained from patients, and all clinical research experiments were performed in accordance with a protocol approved by the ethical committee of the “Lazzaro Spallanzani” Institute. The patients underwent a single cycle of STI, for ⩾1 month of discontinuation. Reintroduction of ART was decided on the basis of the pro tempore guidelines for ART (CD4+ T cell counts <350 cells/μL and/or levels of HIV-RNA in plasma >30,000 copies/mL) [6]. Response was defined as a rebound of HIV-1 RNA in plasma of >30,000 copies/mL (time for nonresponders, <1 month from STI; time for responders, >4 months from STI) (table 1) [6]. Clinical and immunological follow-up and plasma collection were performed at the time of suspension of ART (T0) and 1 month after suspension of ART (T1). Patients with a rapid rebound of HIV viremia (STI nonresponders) were further evaluated, at the time of resumption of ART (T2) (median, 27.8±1.8 d from T1) and 30 d after resumption of ART (T3). Patients with either a delayed or an absent rebound of HIV viremia (STI responders) (median, 365 d of STI after T1) were further evaluated, each month at similar times (1 month after T1 [T1a] and 2 months after T1 [T1b]). Plasma samples were stored at −80°C until being analyzed

Table 1

Characteristics of patients with human immunodeficiency virus (HIV) undergoing structured treatment interruption (STI)

Table 1

Characteristics of patients with human immunodeficiency virus (HIV) undergoing structured treatment interruption (STI)

Commercially available ELISA kits for tumor necrosis factor (TNF)–α, interleukin (IL)–2, and IL-15 (R&D Systems) and for IFN-α and IL-7 (ImmunoKontact) were used according to manufacturer’s instructions. Data are expressed as mean ± SD. Comparisons between groups were made by either Student’s t test or paired Student’s t test, as appropriate

ResultsClinical, immunological, and virological baseline parameters for patients who underwent STI are reported in table 1. Eighteen of 25 patients showed a rapid rebound of HIV viremia after undergoing STI (STI nonresponders), whereas 7 patients presented a delayed or absent rebound of HIV viremia (STI responders; 4 of whom are still undergoing STI after >1 year). In terms of sex, age, pre-ART HIV-1 load set-point, or CD4 and CD8 cell counts at the time of STI, no differences between STI responders and STI nonresponders were observed (P>.05, for all comparisons) (table 1) [6]. We evaluated, in plasma, the levels of cytokines that (1) may have a direct application for treatment of humans (IFN-α and IL-2), (2) are known to influence T cell maturation and differentiation (IL-7 and IL-15), or (3) play a role in the acute phase of response (TNF-α). Levels of these cytokines in plasma were analyzed, at different times, for the 2 groups of patients with HIV: STI nonresponders (figure 1, left panels) and STI responders (figure 1, right panels)

Figure 1.

Levels of cytokines in plasma from structured treatment interruption (STI) nonresponders (rapid rebound of human immunodeficiency virus [HIV] viremia) and from STI responders (delayed rebound of HIV viremia) undergoing STI. Levels of interleukin (IL)–15 (A and B), tumor necrosis factor (TNF)–α (C and D), IL-2 (E and F), IL-7 (G and H), and interferon (IFN)–α (I and J) in plasma were analyzed in STI nonresponders (left panels) and in STI responders (right panels). Levels of cytokines in plasma (expressed as pg/mL) represent a single patient. Copies of HIV-1 RNA in plasma (expressed as media for each time) are indicated by a gray area along the right panels’ Y-axes. Evaluated times (suspension of antiretroviral therapy [ART] [T0], 1 month after suspension of ART [T1], 1 month after T1 [T1a], 2 months after T1 [T1b], resumption of ART [T2], and 30 d after resumption of ART [T3]) are indicated beneath the X-axes

Figure 1.

Levels of cytokines in plasma from structured treatment interruption (STI) nonresponders (rapid rebound of human immunodeficiency virus [HIV] viremia) and from STI responders (delayed rebound of HIV viremia) undergoing STI. Levels of interleukin (IL)–15 (A and B), tumor necrosis factor (TNF)–α (C and D), IL-2 (E and F), IL-7 (G and H), and interferon (IFN)–α (I and J) in plasma were analyzed in STI nonresponders (left panels) and in STI responders (right panels). Levels of cytokines in plasma (expressed as pg/mL) represent a single patient. Copies of HIV-1 RNA in plasma (expressed as media for each time) are indicated by a gray area along the right panels’ Y-axes. Evaluated times (suspension of antiretroviral therapy [ART] [T0], 1 month after suspension of ART [T1], 1 month after T1 [T1a], 2 months after T1 [T1b], resumption of ART [T2], and 30 d after resumption of ART [T3]) are indicated beneath the X-axes

A dramatic difference was present between the concentration of IL-15 in plasma from STI responders and that in plasma from STI nonresponders, at baseline and at all times (figure 1A and 1B). Specifically, STI responders had a significantly higher baseline level of IL-15 in plasma (T0, 30.1±11.1 pg/mL) than did nonresponders (T0, 6.7+7.1 pg/mL) (P<.001). Independent of STI, levels of IL-15 in plasma from STI responders remained constantly higher (T1, 25.0±10.5 pg/mL; T1a, 26.8±8.6 pg/mL; T1b, 29.2±9.2 pg/mL) than those in plasma from STI nonresponders (T1, 7.2±8.0 pg/mL; T2, 5.5±5.9 pg/mL; T3, 7.2±7.3 pg/mL) (P<.001, for all comparisons)

After STI, STI nonresponders presented a significant increase in levels of TNF-α in plasma (T0, 1.5±0.6 pg/mL; T1, 4.2±5.6 pg/mL; P<.01, by paired Student’s t test) (figure 1C). Moreover, after resumption of ART, levels of TNF-α in plasma decreased to pre-STI levels (T3, 2.1±0.7 pg/mL; T0, 1.5±0.6 pg/mL; P>.05), thus correlating with HIV-1 virus load (figure 1C). In contrast, during STI, STI responders had no significant variations in levels of TNF-α in plasma (T0, 2.1±2.0 pg/mL; T1, 1.4±1.3 pg/mL; T1a, 2.0±1.5 pg/mL; T1b, 2.1±1.3 pg/mL; P>.05, all comparisons vs. T0, by paired Student’s t test) (figure 1D)

At T0, no significant differences were observed in the levels of IL-2 in plasma from either STI nonresponders (19.2±16.6 pg/mL) or STI responders (19.5±8.8 pg/mL) (P>.05) (figure 1E and 1F). Moreover, no significant changes were observed in levels of IL-2 in plasma from STI responders (T1, 16.9±9.6 pg/mL; T1a, 22.5±9.1 pg/mL; T1b, 22.1±10.2 pg/mL; P>.05, vs. T0). Similarly, during the without-treatment period, STI nonresponders did not show a significant change in the levels of IL-2 in plasma (T1, 22.8±21.2 pg/mL; T2, 22.6±22.1 pg/mL; P>.05, vs. T0). Nevertheless, a significant increase in the levels of IL-2 in plasma, 1 month after resumption of ART, was observed (T3, 38.9±20.7 pg/mL; P<.01, vs. T0, by paired Student’s t test). Finally, no variations in the levels of either IL-7 in plasma (figure 1G and 1H) or IFN-α in plasma (figure 1I and 1J) were observed in either STI nonresponders or STI responders

DiscussionIn HIV-infected patients, STI of ART may help to reduce the side effects of long-term treatment. Nevertheless, not all patients with HIV undergoing STI are able to control viral replication, a finding suggesting that host factors may play a role in the spontaneous control of HIV [6]. Specifically, in STI patients, we have previously observed an impaired HIV-specific CD8-maturation and -effector capability associated with uncontrolled rebound of HIV viremia. Moreover, Garcia et al. [9] have shown recently that the best predictors of the outcome of STI may be baseline levels of memory CD4 T cells and the proliferative response to recall antigens, a finding indicating that CD4 T cells play a central role in the spontaneous control of viral replication. In our cohort of STI patients, we have also observed that a good response to STI is associated with an increased number of memory CD4 T cells (data not shown). Since, in patients with HIV, a preserved ability to produce cytokines is associated with prolonged survival [10], we investigated the cytokine profile in plasma from both STI responders and STI nonresponders

TNF-α is a key proinflammatory cytokine that is released in the acute-phase response and therefore is strictly related to active replication of HIV [11]. In our study, STI nonresponders presented a significant increase in levels of TNF-α in plasma after STI of ART. This clinical context is characterized by a rapid rebound of HIV viremia and presents aspects similar to those present during acute HIV-1 infection or during AIDS [12], a finding suggesting that levels of TNF-α in plasma represent the proinflammatory host response to an increased virus load. Thus, TNF-α is a serological marker that parallels viral replication by confirming a lack of control of viral replication (indicated by the increased HIV viremia), but it is not predictive of STI outcome

IL-2 produced by CD4 T cells plays a central role in the induction of an effective immune response. HIV infection affects several parts of the immune system, leading to a progressive loss of CD4 T cells and immunosuppression. Because it is possible that IL-2 plays a role in the induction of an effective immune recovery in patients with HIV, its therapeutic use in large clinical trials should be considered [13]. Our data on STI response do not show any significant change between the levels of IL-2 in plasma from STI nonresponders and those in plasma from STI responders. In both groups of patients, a significant increase in the level of IL-2 in plasma was observed 1 month after resumption of ART, a finding confirming that (1) CD4 T cell function is noncompromised by STI and (2) the massive exposure to viral antigens may boost CD4 T cell functions

IL-7 and IFN-α are known to be involved in T cell maturation and antiviral activities, which are both affected during HIV infection [14, 15]. However, in our study, during STI, no significant modifications of these cytokines were observed in either STI nonresponders or STI responders. IL-15 is a cytokine that plays an important role in the early steps of the immune response to infections and is needed for the maturation and differentiation of T cells [16]. It has been shown recently that IL-15 strongly enhances the survival and the effector function of HIV-specific CD8 T cells, which may play a pivotal role in the control of HIV infection [17]. Moreover, coadministration of HIV vaccine vectors with vaccinia viruses expressing IL-15 induces a strong CD8 T cell–mediated immunity [18], a finding indicating that IL-15 represents a cytokine adjuvant for the generation of HIV immunity. In agreement with these data, levels of IL-15 in plasma from STI nonresponders were significantly different than those in plasma from STI responders. Specifically, STI responders had a significantly higher baseline level of IL-15 than did nonresponders, and this difference persisted at all the times during our study. These results indicate that higher levels of IL-15 in plasma correlate with a delayed or an absent rebound of HIV viremia, a delay or absence that results from the ability to control viral replication. To determine a better outcome of STI, baseline levels of IL-15 in plasma could therefore be used as markers. Moreover, an inability to spontaneously control viral replication—an inability that is due to a selective impairment of HIV-specific immune response—could be caused by defective IL-15 production. Accordingly, these findings may support the hypothesis that, in most patients with HIV who, in the absence of ART, are unable to control viral replication, treatment with IL-15 may represent a useful strategy to enhance HIV-specific CD8 T cell response [17, 18]. Altogether, our data indicate that, in patients with chronic HIV infection, levels of IL-15 in plasma may predict a favorable outcome of STI, a finding suggesting that, in patients unable to spontaneously control viral replication, treatment with IL-15 could be useful in attempts to improve the antiviral immune response

References

1
Markowitz
M
Resistance, fitness, adherence, and potency: mapping the paths to virologic failure
JAMA
 , 
2000
, vol. 
283
 (pg. 
250
-
1
)
2
Youle
M
Is interruption of HIV therapy always harmful
J Antimicrob Chemother
 , 
2000
, vol. 
45
 (pg. 
137
-
8
)
3
Lori
F
Lisziewicz
J
Structured treatment interruptions for the management of HIV infection
JAMA
 , 
2001
, vol. 
286
 (pg. 
2981
-
7
)
4
Montaner
LJ
Structured treatment interruptions to control HIV-1 and limit drug exposure
Trends Immunol
 , 
2001
, vol. 
22
 (pg. 
92
-
6
)
5
Daar
ES
Mohri
H
Ho
DD
Virologic markers in the assessment of antiretroviral therapy
AIDS
 , 
1991
, vol. 
5(Suppl 2)
 (pg. 
S175
-
9
)
6
D’Offizi
G
Montesano
C
Agrati
C
, et al.  . 
Expansion of pre-terminally differentiated CD8 T cells in chronic HIV-positive patients presenting a rapid viral rebound during structured treatment interruption
AIDS
 , 
2002
, vol. 
16
 (pg. 
2431
-
8
)
7
Martini
F
Poccia
F
Goletti
D
, et al.  . 
Acute human immunodeficiency virus replication causes a rapid and persistent impairment of Vγ9Vδ2 T cells in chronically infected patients undergoing structured treatment interruption
J Infect Dis
 , 
2002
, vol. 
186
 (pg. 
847
-
50
)
8
Poccia
F
Gougeon
ML
Agrati
C
, et al.  . 
Innate T-cell immunity in HIV infection: the role of Vgamma9Vdelta2 T lymphocytes
Curr Mol Med
 , 
2002
, vol. 
2
 (pg. 
769
-
81
)
9
Garcia
F
Plana
M
Mestre
G
, et al.  . 
Immunological and virological factors at baseline may predict response to structured therapy interruption in early stage chronic HIV-1 infection
AIDS
 , 
2002
, vol. 
16
 (pg. 
1761
-
5
)
10
Ostrowski
SR
Gerstoft
J
Pedersen
BK
Ullum
H
Impaired production of cytokines is an independent predictor of mortality in HIV-1–infected patients
AIDS
 , 
2003
, vol. 
17
 (pg. 
521
-
30
)
11
Tracey
KJ
Cerami
A
Cachectin/tumor necrosis factor and other cytokines in infectious disease
Curr Opin Immunol
 , 
1989
, vol. 
1
 (pg. 
454
-
61
)
12
Bedini
JL
Garcia
F
Aznar
E
, et al.  . 
Serum levels of beta2-microglobulin, neopterin, TNF-alpha and soluble receptors of TNF-alpha and interleukin-2 in intravenous drug abusers according to HIV-1 status and stage of the HIV-1 infection
Clin Microbiol Infect
 , 
1998
, vol. 
4
 (pg. 
4
-
10
)
13
Katlama
C
Carcelain
G
Duvivier
C
, et al.  . 
Interleukin-2 accelerates CD4 cell reconstitution in HIV-infected patients with severe immunosuppression despite highly active antiretroviral therapy: the ILSTIM study—ANRS 082
AIDS
 , 
2002
, vol. 
16
 (pg. 
2027
-
34
)
14
Fry
TJ
Connick
E
Falloon
J
, et al.  . 
A potential role for interleukin-7 in T-cell homeostasis
Blood
 , 
2001
, vol. 
97
 (pg. 
2983
-
90
)
15
Hatzakis
A
Gargalianos
P
Kiosses
V
, et al.  . 
Low-dose IFN-alpha monotherapy in treatment-naive individuals with HIV-1 infection: evidence of potent suppression of viral replication
J Interferon Cytokine Res
 , 
2001
, vol. 
21
 (pg. 
861
-
9
)
16
Sprent
J
Surh
CD
Generation and maintenance of memory T cells
Curr Opin Immunol
 , 
2001
, vol. 
13
 (pg. 
248
-
54
)
17
Mueller
YM
Bojczuk
PM
Halstead
ES
, et al.  . 
IL-15 enhances survival and function of HIV-specific CD8+ T cells
Blood
 , 
2003
, vol. 
101
 (pg. 
1024
-
9
)
18
Oh
S
Berzofsky
JA
Burke
DS
Waldmann
TA
Perera
LP
Coadministration of HIV vaccine vectors with vaccinia viruses expressing IL-15 but not IL-2 induces long-lasting cellular immunity
Proc Natl Acad Sci USA
 , 
2003
, vol. 
100
 (pg. 
3392
-
7
)
Financial support: Research Project of the Italian Ministry of Health; Istituto Superiore di Sanità (grants 30C23 and 45D/1.21)