Primary HIV-1 infection (PHI) involves a highly dynamic relation between the virus and host, with extremely high initial levels of plasma viremia, coincident with antibody seroconversion. In this stage, a typical HIV reactivity pattern can be observed, evolving toward complete maturation of HIV-specific antibodies, as detected by standard assays, such as enzyme-linked immunosorbent assay (ELISA) and Western blot, which are routinely used to diagnose HIV infection.1
Similar to what happens in most infections, antibodies produced in the early phases show a low avidity for antigens, which increases progressively in time after exposure to an immunogen, such that determination of antibody avidity may help in identifying recent infections.2-5
Antiretroviral combination therapy during PHI may limit the extent of viral replication, modifying the pattern of cellular and humoral immune responses.6-10
Some reports documented a reduced humoral response in patients treated during PHI, with declining levels of HIV-1 p24-specific antibodies attributable to the loss of p24-specific CD4 T lymphocytes.6,11-13 More recently, it was observed that an immediate therapeutic intervention during PHI, associated with durable virologic suppression, may block the formation of HIV-1-specific antibodies and, in a few individuals, even determine complete or partial HIV-1 seroreversion.14-17 This phenomenon, mainly reported in children vertically infected with HIV-1 treated in the first month of life,18,19 seems to be uncommon in adults. Obviously, the absence of antibodies in these cases does not indicate absence of infection or viral eradication but demonstrates an incompletely evolved HIV-1 antibody response.
Little is known about the evolution of avidity of the anti-HIV-1 antibodies during the progression of the disease or about possible changes in avidity occurring after the introduction of antiretroviral treatment. One study, which evaluated the antibody avidity in patients chronically infected with HIV-1, using the Avidity Index (AI) as a marker of antibody maturation, found a significant, low, and slowly progressive reduction of anti-HIV-1 IgG avidity after the initiation of highly active antiretroviral therapy (HAART).20 A previous report, focused on early HIV infection, described low avidity and slower antibody maturation in early treated patients.2 Conversely, other authors have reported that AI is not influenced by antiretroviral treatment, including protease inhibitors, irrespective of chronic or acute HIV infection.4 No details on AI trends over time are available, however. A more recent study did not find a significant impact of HAART on avidity evolution of antibodies against specific HIV-1 antigens, suggesting that HAART profoundly affects the amount of antibodies more than their quality.21
The objective of this study was to determine the effects of early treatment of PHI with HAART, and of the subsequent therapy interruption (TI), on the maturation of HIV-1 antibody response as measured by the AI, in parallel with the evolution of Western blot reactivity.
Subjects and Samples
Thirteen individuals (11 male and 2 female) with symptomatic PHI were retrospectively identified from the institutional register of PHIs. PHI was defined by the presence of HIV-1 RNA in the plasma and simultaneous negative or indeterminate HIV-1 antibody test results (HIV-1/2 ELISA and Western blot) in persons with a compatible clinical syndrome. Subjects were diagnosed between 1999 and 2004. Five patients have never been treated; 8 individuals were treated with HAART at the time of diagnosis with 2 nucleoside reverse transcriptase inhibitors combined with a protease or nonnucleoside reverse transcriptase inhibitor. In 4 subjects, HAART was discontinued after a variable time lapse from diagnosis of PHI (≥1 year); 1 of these 4 subjects (T6) underwent multiple cycles of TI.
Clinical and demographic data, including age, gender, HIV risk behavior, baseline CD4 cell count and virus load, seroconversion status, and time of initiation of antiretroviral treatment, were collected at the time of diagnosis or obtained from patients' charts. Adherence and any change in regimen were monitored and assessed by chart review.
Longitudinal serum and/or plasma samples, collected as residual laboratory samples and stored at −80°C, were evaluated. The samples were collected at the time of diagnosis (before starting HAART for treated subjects) and, subsequently, at 3, 6, and 12 months (±20 days) after the diagnosis. Time of diagnosis was defined as the date of the first positive laboratory test result allowing PHI diagnosis.
For those 4 subjects who discontinued HAART, additional specimens were analyzed, on the basis of their availability and volume, to follow in more detail the evolution of the AI in parallel with that of HIV-1-specific Western blot bands.
Routine HIV antibody testing was performed using a commercial HIV-1 assay (Genscreen HIV1/2; Biorad, Marnes-la-Coquette, France) and an HIV-1 Western blot (New Lav Blot-1; Biorad). Results were interpreted according to manufacturer's instructions.
To calculate the AI, the AxSYM HIV 1/2gO (Abbot Diagnostics Division, Delkenheim, Germany) test was used. This assay uses recombinant-derived antigens corresponding to 4 viral proteins (HIV-1 group M envelope, HIV-1 group O envelope, HIV-1 core, and HIV-2 envelope) and 2 synthetic peptides corresponding to HIV-1 and HIV-2 envelope.
According to the procedure indicated by Suligoi et al,4 2 aliquots for each serum/plasma sample were prepared by a 1:10 dilution with phosphate-buffered saline (PBS) and with 1 M of guanidine (G) as a denaturing agent. Both aliquots were shaken, incubated at room temperature for 10 to 15 minutes, and subsequently analyzed, and the AI of HIV antibodies was calculated using the following formula:
where S/CO is the sample/cutoff.
The AI is expected to be lower than 1 in the initial period after infection and to tend to 1 as the antibodies mature. We used an AI >0.80 to define full HIV-1 antibody maturation, according to previous studies, showing that AI values lower than 0.80 suggest recent HIV infections (ie, within 6 months of seroconversion), whereas AI values >0.90 suggest long-standing infections (acquired >1 year earlier).4
For each plasma sample, CD4 T-lymphocyte counts (assessed by standard flow cytometry methods) and HIV viral load (VL) data were available from laboratory records. The latter parameter was determined by a second-generation assay based on nucleic acid sequence-based amplification (NASBA) for samples collected until 2001 and by the branched-chain DNA assay (Versant HIV RNA 3.0; Bayer Diagnostics, Erany, France) for the subsequent samples. All values were reported as in the original record, irrespective of the VL assay, without correction.
Simple descriptive statistical measures for the AI were calculated (median and range). The Fisher exact test and Wilcoxon rank sum test were used to test for differences in baseline VL (patients were categorized on the basis of a level of viremia higher or lower than 10 × 105 copies/mL) and CD4 cell count, respectively, in the 2 groups of patients (untreated and treated). The slope of the regression line fitted on each patient's data was used to summarize the change of AI over the follow-up period, and differences between untreated and treated patients 6 and 12 months after the PHI diagnosis were tested for significance by the Wilcoxon rank sum test.
The demographic, virologic, and immunologic characteristics at diagnosis of the 13 observed patients are described in Table 1. All subjects were Italian, and 8 of the 11 male patients were men who have sex with men (MSM). The median age was 34 years (range: 20 to 54 years). Of 13 patients, 8 received HAART. The median time from diagnosis was 14.5 days (range: 8 to 52 days).
All patients developed a complete Western blot pattern within 6 months from diagnosis of PHI. For 1 patient, who had an indeterminate Western blot result during the first 3 months after diagnosis of PHI, the assay became positive on the first subsequent occasion (ie, 10 months after diagnosis of PHI).
Overall, at diagnosis, the VL ranged from 289 to 38,000,000 copies/mL and the CD4 count ranged from 338 to 882 cells/mm3. On average, the HIV VL at diagnosis tended to be lower in untreated as compared with treated patients (2 of 5 untreated patients vs. 6 of 8 treated patients had an HIV RNA level higher than 10 × 105 copies/mL; P = 0.29). Consistently, the CD4 cell count tended to be higher in untreated patients (the median values were 700 and 466 cells/mm3 in untreated and treated subjects, respectively; P = 0.08).
In untreated patients, after a peak, the VL reached the set point in approximately 4 to 6 months.
In treated patients, all of whom were adherent subjects, the treatment was effective, leading to suppression of the VL lasting for the whole period of observation of this study while patients were receiving HAART (12 months). No difference between subjects receiving protease inhibitor-based regimens and those receiving nonnucleoside reverse transcriptase inhibitor-based regimens was observed.
Evolution of Avidity Index
Avidity Index in Untreated and Treated Patients
At diagnosis, the median AI was not different in untreated and treated patients (0.42 [range: 0.33 to 0.43] and 0.44 [range: 0.40 to 0.72], respectively; P = 0.242). In untreated patients, an increasing trend of AI, reaching a value of 0.80 at 3 months and tending to unity 6 months after diagnosis, was observed. By contrast, in all treated patients, the overall trend of AI was stable during the whole observation time under HAART (12 months), with values always remaining less than 0.80 (Fig. 1). A significant difference in the median values of AI between untreated and treated patients was observed at 6 months, with AI values in untreated patients significantly higher than those observed in treated patients (0.89 vs. 0.45, respectively; P = 0.005). Similarly, the difference was significant when the median slope of the regression lines was computed for the 2 groups of patients (0.082 vs. −0.007, respectively; P = 0.005).
Avidity Index and Western Blot Reactivity in Subjects Who Stopped HAART
Of 8 treated patients, 4 stopped HAART for different reasons at variable time points after diagnosis of PHI: patient T2 interrupted treatment after 15 months because of intolerance, patient T4 interrupted treatment after 12 months because of toxicity, and patient T6 interrupted treatment after 41 months and patient T8 interrupted treatment after 23 months because of their own decision to do so. At TI, all patients had a CD4 count greater than 500 cells/mm3 and undetectable viremia. Patient T6, who underwent 2 cycles of TI, resumed HAART after 19 months, then stopped again after 7 months, and resumed HAART after a further 4 months. The therapeutic scheme for these 4 patients is illustrated in Figure 2.
For these patients, the AI was calculated in every residual serum/plasma sample available from the time of PHI diagnosis until the completion of the study. Western blot reactivity was evaluated at the time of diagnosis, at intermediate time points while receiving HAART, and after discontinuation of therapy. For patient T6, a further sample collected after the second resumption of HAART was evaluated.
Figure 2 shows the time course of the AI, in parallel with VL and evolution of the Western blot pattern, for each of the 4 patients discontinuing HAART.
In all these patients, the AI was still low (<0.80 in all) at the time of TI (ie, as long as 41 months after the diagnosis of PHI in patient T6) and promptly increased thereafter, reaching a value of 0.80 in ≤6 months in all patients. The increasing trend of AI was parallel to the rise of HIV VL attributable to rebound of viral replication in the absence of therapy. It is to be noted that in patient T6, who resumed HAART after 2 cycles of TI, the AI remained at values exceeding 0.80 from the first interruption onward, even after HAART resumption.
All these patients developed a fully reactive Western blot test result within 6 months of diagnosis while receiving HAART. Thereafter, a weakening of the antibody response was observed in all and was particularly evident in patient T2 (showing seroreversion of 2 previously positive bands in the sample collected at 9 months after diagnosis) and in patient T8 (showing reversion of 3 bands, with a positive result only for p24 and p55 in the sample collected at 6 months after diagnosis).
After TI-driven VL rebound, Western blot analyses were reactive for all bands in every subject. The full reactivity lasted in patient T6 after virologic suppression after resumption of therapy.
DISCUSSION AND CONCLUSIONS
The study demonstrates that in patients with PHI, immediate establishment of HAART prevents the gradual evolution of the AI of HIV-1-specific antibodies. In fact, among treated patients with PHI, the AI values remained constantly lower than 0.80 during the entire period of treatment (ie, for periods as long as 41 months after diagnosis of PHI, as seen in patient T6). In contrast, in all untreated patients, a progressive increase of the AI, reaching values greater than 0.90 at 6 months after the diagnosis of PHI, was observed.
These data suggest that the early and rapid suppression of HIV replication, driven by HAART administered within 60 days after PHI diagnosis, may lead to an altered pattern of antibody maturation. In particular, it seems that maturation of antibody avidity evolves only in the presence of ongoing viral replication, presumably as a result of continuous HIV antigen stimulation.
The relation between the viral suppression established by HAART and the low antibody avidity observed during the whole treatment period is supported by the AI trend in patients who have interrupted HAART. In these subjects, the restart of viral replication after discontinuation of HAART prompted the resumption of the evolution pattern of antibody maturation, as indicated by the subsequent increase of AI values.
It is already known that reduction of antibody level, or even complete or partial HIV-1 seroreversion, documented by the disappearance of previously present Western blot bands, may occur in adult patients receiving HAART during acute or early HIV infection15-17 or in neonates who are treated in the first months of life.18,19 This phenomenon is not common in adults. Kassutto et al15 found an incomplete HIV-1 antibody response in 3 of 150 patients treated, on average, 8 days after presentation, and Hare et al16 observed seroreversion on at least 1 enzyme immunoassay (EIA) antibody assay in 6 of 87 patients receiving HAART within 28 days from enrollment. More recently, among 84 patients receiving HAART for longer than 5 years, Amor et al17 reported a reduced reactivity in all Western blot bands and disappearance of reactivity to gp41 using synthetic peptides in a patient who had started antiretroviral therapy early after diagnosis of PHI.
Our results confirm that HAART affects the humoral immune response and suggest that maturation of antibody response may show different patterns of evolution over time, depending on whether the development of the repertoire of Western blot bands or the increase of the AI is considered. In fact, we observed that in all 8 patients who suppressed HIV viremia early after PHI diagnosis, the increase of the AI was blocked and resumed only after the rebound of virus replication that followed interruption of HAART. Differently from the AI, Western blot pattern maturation progressed to reach a complete repertoire after 6 months from the diagnosis of PHI in 7 of these patients and after 10 months for 1 patient (see section on materials and methods), even in the presence of continuous suppression of viral replication, although a transient and partial reduction of antigen-specific bands may occur during HAART (see patients T2 and T8 in Fig. 2). These findings are in agreement with those regarding other infections, such as rubella22 or tuberculosis,23 and suggest that antibody affinity is controlled by mechanisms independent of those regulating the extent and repertoire of antibody production.
The block of AI evolution in patients undergoing HAART since the diagnosis of PHI, here reported for the first time, apparently contradicts results previously reported by Suligoi et al4 and by Adalid-Peralta et al.21 Suligoi et al4 found a marked increase in AI during the first year after seroconversion, reaching 0.90 around 15 months after diagnosis of PHI, with stabilization at the value of 1 during the subsequent 1-year period. They did not find any difference in the AI trend in 4 of 16 patients with acute HIV disease who received HAART immediately after clinical diagnosis, however. This disagreement may be related to different patients' characteristics, such as risk behavior, immune pattern before HAART, or, most importantly, definition criteria and earliness of diagnosis of PHI. Adalid-Peralta et al,21 using Western blot analysis, found that during PHI, the differences between the maturation of IgG avidity against specific HIV antigens in naive and early treated patients were not significant, with the exception of anti-p55 only, concluding that antibody maturation was not prevented by HAART. The apparent discrepancy with respect to our findings can be partially explained by the different method used to determine the AI. Because the method used in the present study did not allow us to measure the avidity of HIV-1 antigen-specific IgG antibodies, however, no direct comparison of data is possible. Additionally, other factors, such as patient selection criteria and elevated individual variability, can play a role. In particular, our patient groups are quite homogeneous and the individual variability of the AI at each time point is low, such that the observed differences, even if referring to the overall AI of all plasma anti-HIV antibodies, are highly significant.
A limit of our study is its retrospective nature. In fact, at diagnosis, the 2 groups of patients (untreated and treated) differed in their VL and CD4 values, which are parameters taken into account to judge the clinical necessity to establish early HAART.
Our results are interesting for a number of reasons. First, because the AI is currently used in seroepidemiologic studies for determining the approximate time of acquisition of HIV infection, the altered kinetics of the AI attributable to early control of viremia during HAART could lead to inconsistent estimates in a substantial proportion of patients. Another point of interest is that our data add new information about the complexity of the initial host-virus interaction during PHI. In fact, the altered pattern of maturation of antibody response in patients achieving rapid control of HIV replication suggests the interplay of factors in immune response to HIV not yet fully investigated.
Whether the AI may have a role as a marker to evaluate the use of HAART in acute or early HIV infection or to predict treatment outcome remains unknown, however.
Additional studies could help us to understand the correlates of effective immune control of HIV better and could eventually contribute to the development of an effective vaccine strategy or other approaches for immune-mediated viral containment.
The authors thank Carla Nisii for her editorial assistance. All the authors were involved in the analysis of data, interpretation of data, and development of discussions included in this article.
1. Fiebig EW, Wright DJ, Rawal BD, et al. Dynamics of HIV viremia and antibody seroconversion in plasma donors: implications for diagnosis and staging of primary HIV infection. AIDS
2. Le Guillou H, Le Meur A, Bourdon S, et al. Avidites des anticorps: application au diagnostic d'infection récente à VIH. Ann Biol Clin (Paris)
3. Suligoi B, Galli C, Massi M, et al. Precision and accuracy of a procedure for detecting recent human immunodeficiency virus infections by calculating the antibody avidity
index by an automated immunoassay-based method. J Clin Microbiol
4. Suligoi B, Massi M, Galli C, et al. Identifying recent HIV infections using the Avidity Index and an automated enzyme immunoassay. J Acquir Immune Defic Syndr
5. Janssen RS, Satten GA, Stramer SL, et al. New testing strategy to detect early HIV-1 infection for use in incident estimate for clinical and prevention purposes. JAMA
6. Markowitz M, Vesanen M, Tenner-Racz K, et al. The effects of commencing combination antiretroviral therapy soon after human immunodeficiency virus type 1 infection on viral replication
and antiviral immune responses. J Infect Dis
7. Binley JM, Trkola A, Ketas T, et al. The effect of highly active antiretroviral therapy
on binding and neutralizing antibody responses to human immunodeficiency virus type I infection. J Infect Dis
8. Rosemberg ES, Altfeld M, Poon SH, et al. Immune control of HIV-1 after early treatment of acute infection. Nature
9. Kaufmann GR, Bloch M, Zaunders JJ, et al. Long-term immunological response in HIV-1-infected subjects receiving potent antiretroviral therapy. AIDS
10. Lacabaratz-Porret C, Urrutia A, Doisne JM, et al. Impact of antiretroviral therapy and changes in viral load on human immunodeficiency virus (HIV)-specific T cell responses in primary HIV infection. J Infect Dis
11. Zaunders JJ, Cunningham PH, Kelleher AD, et al. Potent antiretroviral therapy of primary human immunodeficiency virus type (HIV-1) infection: partial normalization of T lymphocyte subsets and limited reduction of HIV DNA despite clearance of plasma viremia. J Infect Dis
12. Lafeuillade A, Poggi C, Tamalet C, et al. Effects of a combination of zidovudine, didanosine, and lamivudine on primary human immunodeficiency virus type 1 infection. J Infect Dis
13. Morris BL, Binley JM, Clas BA, et al. HIV-1 antigen-specific and nonspecific B cell responses are sensitive to combination antiretroviral therapy. J Exp Med
14. Jurriaans S, Sankatsing SUC, Prins JM, et al. HIV-1 seroreversion in an HIV-1 seropositive patient treated during acute infection with highly active antiretroviral therapy
and mycophenolate mofetil. AIDS
15. Kassutto S, Johnston MN, Rosemberg ES. Incomplete HIV type 1 antibody evolution and seroreversion in acutely infected individuals treated with early antiretroviral therapy. Clin Infect Dis
16. Hare CB, Pappalardo BL, Busch MP, et al. Seroreversion in subjects receiving antiretroviral therapy during acute/early HIV infection. Clin Infect Dis
17. Amor A, Toro C, Jimenez V, et al. Seroreversion of HIV antibodies in patients with prolonged suppression of viremia under HAART. AIDS
18. Hainaut M, Peltier CA, Gerard M, et al. Effectiveness of antiretroviral therapy initiated before the age of 2 months in infants vertically infected with human immunodeficiency virus type 1. Eur J Pediatr
19. Luzuriaga K, Bryson Y, Krogstad P, et al. Combination treatment with zidovudine, didanosine, and nevirapine in infants with human immunodeficiency virus type 1 infection. N Engl J Med
20. Le Guillou-Guillemette H, Renier G, Vielle B, et al. Immune restoration under HAART in patients chronically infected with HIV-1: diversity of T, B, and NK immune responses. Viral Immunol
21. Adalid-Peralta L, Grangeot-Keros L, Rudent A, et al. Impact of highly active antiretroviral therapy
on the maturation of anti-HIV-1 antibodies during primary HIV-1 infection
. HIV Med
22. Pullen GR, Fitzgerald MG, Hoskiing CS. Antibody avidity
determination by ELISA using thiocyanate elution. J Immunol Methods
23. Pereira Arias-Bouda LM, Kuijper S, Van der Werf A, et al. Changes in avidity and level of immunoglobulin G antibodies to Mycobacterium tuberculosis in sera of patients undergoing treatment for pulmonary tuberculosis. Clin Diagn Lab Immunol