Kaufmann, Gilbert R.a,e; Bloch, Markb; Finlayson, Robertc; Zaunders, Johnd; Smith, Dona; Cooper, David A.a,d
Potent antiretroviral therapy (ART) efficiently inhibits viral replication, resulting typically in a biphasic decline in plasma HIV-1 RNA . The degree of viral load reduction is inversely associated with the short and medium-term increase in CD4 cell count [2,3], allowing for a slow but gradual quantitative and qualitative recovery of the immune system [4,5].
The CD4 T lymphocyte response to ART is highly variable, which has resulted in early reports of discordant virological and immunological responses . The reasons for the high variability of T cell responses have been insufficiently studied and still remain unclear. It is conceivable that immunological as well as virological factors may be involved. In treatment-naive individuals, a syncytium-inducing T cell tropic viral strain leads to a more rapid decline of CD4 T lymphocytes than a macrophage-tropic strain . As viral replication is not completely suppressed by ART [7,8], a highly cytopathic viral strain may similarly impede the recovery of CD4 T lymphocytes in individuals receiving ART . A second potential factor determining the immunological recovery may be the thymus. Young children with more functional thymus tissue show a more rapid reconstitution of the immune system than older children or adults . Third, the degree of HIV-1-related immunological damage may affect immune restoration. Individuals with advanced HIV-1 infection have progressively lost naive CD4 T lymphocytes . It has been suggested that the number of naive cells predicts the subsequent level of CD4 T lymphocytes that can be reached by ART .
In numerous studies [12–16], only a partial normalization of HIV-1-associated immunological alterations has been reported. However, the short follow-up of these investigations does not allow a definitive conclusion about the potentially achievable level of immunological recovery. In this study, we analysed the long-term dynamics of CD4 and CD8 T lymphocytes over 4 years in an observational cohort of HIV-1-infected subjects who commenced ART. Inclusion criteria were based on the aim to evaluate the maximum level of immune reconstitution that can be reached by ART. Therefore, only those individuals were considered who showed a long-term undetectable viral load on ART. In this selected cohort, the characteristics of good and poor immunological responders, as well as baseline parameters that may serve as predictors of the CD4 T cell response to ART, were studied.
Subjects were recruited from two primary care practices (Holdsworth House General Practice and Taylor Square Private Clinic, Sydney, Australia). Individuals were eligible if they: (i) commenced ART between January 1996 and December 1997; (ii) reduced plasma HIV-1 RNA to undetectable levels (< 400 copies/ml); and (iii) maintained undetectable plasma HIV-1-RNA values during the observation period. Changes of ART were allowed in the case of adverse events or as requested by patients, as long as the plasma HIV-1 viral load remained below the threshold of detection.
A total of 210 individuals presented with good initial virological responses to ART, suppressing HIV-1 RNA to undetectable levels. However, 115 of these subjects were excluded from analysis, because they either showed a subsequent viral load rebound as a result of the discontinuation of ART or virological failure of the drug regimen, had been pre-treated with a protease inhibitor (PI), or had missing baseline plasma HIV-1-RNA data.
The final analysis was based on data from 95 subjects, who were followed for a median observation period of 45 months. Study participants had a mean age of 40 ± 8 years and 100% were men because of the large predominance of the male homosexual HIV risk group in Sydney (see Table 1). Half of the subjects were treatment naive and half had been pre-treated with nucleoside analogues. Twenty-eight subjects (30%) received indinavir, 27 (29%) saquinavir (hard-gel), and 21 (22%) a combination of ritonavir and saquinavir. A small number of individuals were treated with ritonavir (n = 2), a combination of nelfinavir and saquinavir (n = 1), nevirapine (n = 8) or a combination of a non-nucleoside reverse transcriptase inhibitor (NNRTI) and a PI (n = 4). In addition, four individuals were treated with double nucleoside analogue combinations. The reverse transcriptase inhibitor regimen mainly consisted of stavudine and lamivudine (55%), zidovudine and lamivudine (29%), or stavudine and didanosine (6%). During the observation period, 45 subjects (47%) changed the PI or the NNRTI. Twenty-three individuals (24%) discontinued the PI and received an NNRTI. The reverse transcriptase inhibitor regimen was changed in 61 individuals (64%) because of adverse events.
After whole-blood lysis (FACSlysing Solution, Becton-Dickinson, San Jose, CA, USA), T lymphocyte counts were determined by three-colour flow cytometry using CD45-peridin chlorophyll protein versus side-scatter gating, CD3-FITC, CD4-phycoerythrin and CD8-phycoerythrin monoclonal antibodies (Becton-Dickinson, San Jose, CA, USA). The analysis was performed on an EPICS XL flow cytometer (Coulter Electronics) in accordance with Centers for Disease Control and Prevention (CDC) guidelines.
Plasma HIV-1 RNA was measured using a quantitative reverse transcriptase–polymerase chain reaction using the HIV Monitor test version 1.5 with a lower limit of detection of 400 copies/ml (Roche Molecular Systems, Branchburg, NJ, USA).
All results are presented as medians and interquartile ranges, except for age, which is shown as mean ± standard deviation. T cell dynamics were analysed by calculating 3–6 monthly changes of CD4 and CD8 cell counts in the first year of ART and yearly changes thereafter. To allow for irregular visits, missing laboratory values were interpolated using the mean of the two adjacent measurements.
Three major immunological endpoints were analysed, including (i) the absolute CD4 cell count at 48 months (last value after 3 years of ART carried forward); (ii) the increase in CD4 cell count during the observation period and (iii) the proportion of individuals with a CD4 count above 200 cells/μl (threshold for opportunistic infections), above 350 cells/μl (threshold for the initiation of ART ), and above 500 cells/μl (lower limit of the normal range). In addition, the proportion of patients who reached normal CD8 cell counts (< 1000 cells/μl) was evaluated. The percentages of T lymphocytes were not analysed, because clinical decisions are usually based on absolute T cell counts.
The relationship between baseline parameters and immunological endpoints was evaluated using either an analysis of variance of ranked data or a logistic regression analysis. A Spearman's rank correlation test was used to evaluate the relationship between T cell changes at different time intervals. Paired baseline and follow-up samples were analysed using a Wilcoxon signed rank test. A two-sided P value less than 0.05 was considered statistically significant. Statistical tests were performed using SPSS version 8.0 (SPSS Inc., Chicago, IL, USA).
In 60% of subjects, the plasma HIV-1-RNA level declined within 3 months from a median of 4.7 log10 copies/ml to a level below the threshold of detection (< 400 copies/ml), 24% reached an undetectable viral load after 3 and 6 months, and 16% reached an undetectable viral load after 6 months. The median CD4 cell count increased from 325 cells/μl at baseline to 624 cells/μl at 48 months (Fig. 1a). A maximum CD4 cell count of 780 cells/μl was attained at 37 months (interquartile range 26–42). The greatest increase in CD4 cell count was observed in the first 3 months (22.6 cells/μl per month). Thereafter, it became gradually smaller, reaching 8.6 cells/μl per month between months 3 and 12, 6.8 cells/μl per month in the second year, 3.3 cells/μl per month in the third, and 1.7 cells/μl per month in the fourth year (Fig. 2a). See Table 1 for baseline characteristics of all subjects.
The proportion of subjects who reached normal CD4 cell counts (> 500 cells/μl) increased from 17% at baseline to 71–75% at 36–48 months (Fig. 3), whereas the percentage of individuals who reached CD4 cell counts above 350 cells/μl and above 200 cells/μl increased from 43 to 88–91% and from 76 to 94–98%, respectively.
In contrast, CD8 cell counts did not show significant changes over time, only slightly declining from 1105 cells/μl at baseline to 1079 cells/μl at 48 months (P = 0.41;Fig. 1a). Median changes in CD8 T lymphocytes in the first, second and third year were −3.7 cells/μl per month, −1.9 cells/μl per month and 3.8 cells/μl per month, respectively (Fig. 2b). In the fourth year, an increase in CD8 cell count of 6.9 cells/μl per month was observed. At 48 months, the CD8 cell count reached the normal range (< 1000 cells/μl) in 36 individuals (44%), but only 23 subjects (28%) showed a normalization of both CD4 and CD8 cell counts. The CD4 to CD8 cell ratio increased from 0.26 at baseline to 0.59 at 48 months, following a similar time course as CD4 T lymphocytes (Fig. 1b).
Twenty-two individuals (26%) met the definition of poor immunological responders, and did not reach a CD4 cell count greater than 500 cell/μl at 48 months. Interestingly, 50% of these subjects intermittently exhibited a CD4 cell count above 500 cells/μl, but could not maintain CD4 cell counts in this range.
Thirty-six subjects (43%) showed a substantial CD4 cell decline of more than 20% from the maximum attained CD4 cell count. In 19 individuals (23%) CD4 cell counts declined by more than 30%, despite suppressing viral load below 400 copies/ml.
Relationship between baseline parameters and the CD4 T cell response
Younger individuals showed greater increases in CD4 cell count than older subjects (P = 0.002;Table 2). In addition, a higher nadir CD4 cell count was significantly associated with a greater increase in CD4 T lymphocytes (P = 0.004).
Higher CD4 cell counts at 48 months were associated with higher baseline CD4 cell counts (P < 0.001), higher nadir CD4 cell counts (P < 0.001), a less advanced CDC stage (P = 0.001) and younger age (P = 0.002). In a multivariate model, a higher nadir CD4 cell count (P < 0.001) and younger age (P = 0.006) independently predicted higher CD4 cell counts at 48 months (Table 3).
Six out of 18 subjects (33%) with fewer than 200 CD4 cells/μl at baseline attained a CD4 cell count above 500 CD4 cells/μl at 48 months, whereas 19 out of 27 individuals (70%) with CD4 cell counts between 200 and 350 cells/μl, and all subjects (100%) with more than 350 cells/μl achieved a normal CD4 cell count (> 500 cells/μl;P < 0.001). On the basis of these results, the longitudinal CD4 and CD8 cell count as well as the CD4 to CD8 cell ratio were analysed after the stratification of individuals into three categories of baseline CD4 cell count. This analysis revealed that CD4 T lymphocytes and the CD4 to CD8 cell ratio in the three subgroups followed parallel time courses. In contrast, the CD8 cell count did not show such a pattern (Fig. 4). Consequently, the monthly CD4 T cell changes were similar among the three baseline CD4 cell strata (Fig. 2).
Risk factors for a poor immunological response
Nadir and baseline CD4 cell counts were significantly lower in poor immunological responders than in individuals who reached the normal range for CD4 T lymphocytes at 48 months (99 versus 160 cells/μl for nadir CD4 cell count; 160 versus 373 cells/μl for baseline CD4 cell count;P < 0.0001 and P = 0.0001, respectively). Moreover, a trend was observed that individuals with poor immunological responses were older (42 versus 37 years;P = 0.08) and were in a more advanced CDC category (P = 0.08). In contrast, baseline CD8 cell counts, plasma HIV-1-RNA levels and the duration of HIV-1 infection were similar in poor and good responders. In a multivariate logistic regression analysis, the relative risk of a poor immunologial response increased 3.2 times per 100 cells decline in the nadir CD4 cell count, whereas the baseline CD4 cell count did not represent an independent predictor as a result of its significant relationship with the nadir CD4 cell count (r = 0.67;P < 0.001;Table 4).
Factors associated with the decline in CD4 cell count
Subjects who experienced substantial CD4 cell declines of 20–30% from the maximum attained CD4 cell count showed no distinct characteristics compared with individuals who maintained stable CD4 cell counts (data not shown). In particular, no relationship was observed between changes in ART and a decline in CD4 T lymphocytes. Moreover, subjects with a significant decline (> 20%) in CD4 cell count were found with a similar frequency in the group on NNRTI and in the group on PI (40 versus 49%;P = 0.51).
Relationship between changes in CD4 T lymphocytes in different time intervals
Early CD4 cell count increases during the first 3 and 6 months of ART were significantly associated with the total increase in CD4 T lymphocytes over 48 months (r = 0.52, P < 0.001 and r = 0.55, P < 0.001, respectively). However, early changes in CD4 T lymphocytes were not associated with increases in CD4 cell counts in the second, third and fourth year. Interestingly, changes in CD4 cell count in the second year were inversely correlated with CD4 cell count changes in the third year of ART (r = −0.41, P < 0.001). Similarly, CD4 cell count changes in the third year were inversely correlated with CD4 cell count changes in the fourth year (r = −0.65;P < 0.001).
Influence of antiretroviral therapy on T cell responses
The subgroup of subjects treated continuously with PI (n = 42) and individuals who received NNRTI or were switched from a PI to an NNRTI (n = 53) showed similar baseline CD4 cell counts (323 versus 344 cells/μl;P = 0.56) and subsequent increases in CD4 T lymphocytes during the observation period (313 versus 303 cells/μl;P = 0.38). Moreover, the frequency of poor immunological responders was not significantly different in both groups (20 versus 32%;P = 0.31) and T cell dynamics were similar (data not shown).
The level of immune competence that can be reached in HIV-1-infected individuals receiving ART is of major clinical significance. It is important to detect early those subjects who are likely to belong to the group of poor immunological responders, and have a lengthy exposure to the risk of opportunistic infections. We hypothesized that certain baseline parameters may be associated with the degree of immunological recovery to ART and allow the early detection of these individuals. Such parameters may include: baseline or nadir CD4 cell counts, which reflect the degree of HIV-1-related immunodeficiency; age, which is inversely correlated with thymic function; and baseline viral load, which may serve as a marker of viral fitness.
At 4 years, more than 70% of treated individuals reached the normal range of CD4 T lymphocytes (> 500 cells/μl). Moreover, CD4 cell counts increased in almost all of these individuals to levels above 200 cells/μl, providing protection against major opportunistic infections . HIV-1-related immunodeficiency, as reflected in the baseline or nadir CD4 cell count, was strongly associated with the recovery of CD4 T lymphocytes. A minority of individuals with fewer than 200 CD4 T lymphocytes/μl at baseline reached the normal range of CD4 T lymphocytes after 4 years of ART, whereas a larger proportion of individuals with baseline CD4 cell counts of between 200 and 350/μl, and all subjects with more than 350 CD4 T lymphocytes/μl achieved this goal. This finding supports the concept that the immunological damage caused by HIV-1 over time appears to determine the degree of recovery of CD4 T lymphocytes.
CD4 T lymphocyte dynamics followed three major phases. Consistent with previous reports, a rapid early increase in the CD4 cell count was observed in the first 3 months [4,5,19]. This early phase was followed by a second slower phase, which was characterized by an almost linear increase in the CD4 cell count in the first 2 years. In the third year, the CD4 cell count approached a plateau level, with only minor subsequent changes. Even individuals who commenced ART with low CD4 cell counts showed this pattern of CD4 T lymphocyte dynamics. The largest effect of ART on the recovery of the immune system therefore appears to occur in the first 2 years.
Interestingly, only 28% had a combination of normal CD4 and CD8 cell counts, because most individuals maintained high CD8 cell counts. Persistently elevated CD8 cell counts probably indicate ongoing viral activity and antigen-driven expansion of CD8 T lymphocytes . Residual viral activity may partly explain the large individual variability in the recovery of CD4 T lymphocytes.
Twenty-five per cent of individuals belonged to the category of poor or moderate immunological responders. CD4 cell counts increased very slowly in these individuals, and did not reach the normal range (500 cells/μl) at 48 months. Whether poor responders will eventually reach a CD4 cell count of 500 cell/μl after a longer observation period cannot conclusively be answered in this study, but it may take at least 6–8 years in some individuals to reach this level of CD4 cell count. The risk of a poor immunological response increased with a lower baseline CD4 cell count and a lower nadir CD4 cell count. Moreover, there was a trend that older patients were more likely to experience poorer immunological responses. It is conceivable that poor immunological responders are individuals who have experienced more severe HIV-1-related damage to the immune system. The relationship between age and the immunological response supports the concept of an age-related decline in thymic function and probably other regenerative mechanisms, such as the peripheral expansion of CD4 T lymphocytes [10,20, 21].
The CD4 cell count began to decline substantially in a large proportion of subjects after the maximum CD4 cell count had been reached. The reason for this finding remains unclear. None of the evaluated baseline parameters was associated with the decline in CD4 cell count. In particular, the current drug regimen or changes in ART did not seem to be associated with a decline in CD4 cell count. As the subgroups receiving distinct ART regimens were small and CD4 cell count changes were observed very late at 2–3 years, this finding requires further investigation in a larger study with a longer follow-up.
Interestingly, early increases in CD4 T lymphocytes in the first 6 months were highly associated with the total change in CD4 T lymphocyte number over 48 months, but not with CD4 T cell changes in the second, third or fourth year. The relationship between the early and total increase in CD4 cell count indicates that early changes in CD4 cell counts contribute substantially to the overall recovery of CD4 T lymphocytes. The lack of a relationship between early and late CD4 cell increases suggests that distinct mechanisms may be responsible for the recovery of CD4 T lymphocytes in different phases of immune reconstitution. There is growing evidence that the mechanism responsible for the initial increase in CD4 cell count is mainly the entrapment of CD4 T lymphocytes in the lymphoid tissue, which are released upon the initiation of ART . In contrast, the late phase increase may be the result of the peripheral expansion of circulating memory cells .
The greatest impact of current ART regimens on the recovery of CD4 T lymphocytes occurs within the first 2 years. The degree of immunodeficiency before the initiation of ART and age are significant predictors of the subsequent recovery of CD4 T lymphocytes. This finding suggests that ART should be commenced before severe immunological damage to ensure immune restoration to normal levels.
The authors would like to thank the practitioners at the Holdsworth House General Practice and Taylor Square Private Clinic for their contribution to this study.
1. Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995, 373: 123–126.
2. Staszewski S, Miller V, Sabin C. et al
. Determinants of sustainable CD4 lymphocyte count increases in response to antiretroviral therapy. AIDS 1999, 13: 951–956.
3. Kaufmann D, Pantaleo G, Sudre P, Telenti A. CD4-cell count in HIV-1-infected individuals remaining viraemic with highly active antiretroviral therapy (HAART). Swiss HIV Cohort Study.
Lancet 1998, 351: 723–724.
4. Autran B, Carcelain G, Li T. et al
. Positive effects of combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease. Science 1997, 277: 112–116.
5. Kelleher AD, Carr A, Zaunders J, Cooper DA. Alterations in the immune response of human immunodeficiency virus (HIV)-infected subjects treated with an HIV-specific protease inhibitor, ritonavir. J Infect Dis 1996, 173: 321–329.
6. Koot M, Keet IP, Vos AH. et al
. Prognostic value of HIV-1 syncytium-inducing phenotype for rate of CD4+ cell depletion and progression to AIDS. Ann Intern Med 1993, 118: 681–688.
7. Lafeuillade A, Chollet L, Hittinger G, Profizi N, Costes O, Poggi C. Residual human immunodeficiency virus type 1 RNA in lymphoid tissue of patients with sustained plasma RNA of < 200 copies/mL. J Infect Dis 1998, 177: 235–238.
8. Zhang L, Ramratnam B, Tenner-Racz K. et al
. Quantifying residual HIV-1 replication in patients receiving combination antiretroviral therapy. N Engl J Med 1999, 340: 1605–1613.
9. Renaud M, Katlama C, Mallet A. et al
. Determinants of paradoxical CD4 cell reconstitution after protease inhibitor-containing antiretroviral regimen. AIDS 1999, 13: 669–676.
10. Cohen Stuart JW, Slieker WA, Rijkers GT. et al
. Early recovery of CD4+ T lymphocytes in children on highly active antiretroviral therapy. Dutch study group for children with HIV infections.
AIDS 1998, 12: 2155–2159.
11. Roederer M, Dubs JG, Anderson MT, Raju PA, Herzenberg LA. CD8 naive T cell counts decrease progressively in HIV-infected adults. J Clin Invest 1995, 95: 2061–2066.
12. Notermans DW, Pakker NG, Hamann D. et al
. Immune reconstitution after 2 years of successful potent antiretroviral therapy in previously untreated human immunodeficiency virus type 1-infected adults. J Infect Dis 1999, 180: 1050–1056.
13. Arno A, Ruiz L, Juan M. et al
. Impact on the immune system of undetectable plasma HIV-1 RNA for more than 2 years. AIDS 1998, 12: 697–704.
14. Connors M, Kovacs JA, Krevat S. et al
. HIV infection induces changes in CD4+ T-cell phenotype and depletions within the CD4+ T-cell repertoire that are not immediately restored by antiviral or immune-based therapies. Nat Med 1997, 3: 533–540.
15. Gray CM, Schapiro JM, Winters MA, Merigan TC. Changes in CD4+ and CD8+ T cell subsets in response to highly active antiretroviral therapy in HIV type 1-infected patients with prior protease inhibitor experience. AIDS Res Hum Retroviruses 1998, 14: 561–569.
16. Plana M, Garcia F, Gallart T, Miro JM, Gatell JM. Lack of T-cell proliferative response to HIV-1 antigens after 1 year of highly active antiretroviral treatment in early HIV-1 disease. Immunology Study Group of Spanish EARTH-1 Study.
Lancet 1998, 352: 1194–1195.
18. Furrer H, Egger M, Opravil M. et al
. Discontinuation of primary prophylaxis againstPneumocystis cariniipneumonia in HIV-1-infected adults treated with combination antiretroviral therapy. Swiss HIV Cohort Study.
N Engl J Med 1999, 340: 1301–1306.
19. Pakker NG, Roos MT, van Leeuwen R. et al
. Patterns of T-cell repopulation, virus load reduction, and restoration of T-cell function in HIV-infected persons during therapy with different antiretroviral agents. J Acquir Immune Defic Syndr 1997, 16: 318–326.
20. Zhang L, Lewin SR, Markowitz M. et al
. Measuring recent thymic emigrants in blood of normal and HIV-1-infected individuals before and after effective therapy. J Exp Med 1999, 190: 725–732.
21. Douek DC, McFarland RD, Keiser PH. et al
. Changes in thymic function with age and during the treatment of HIV infection. Nature 1998, 396: 690–695.
22. Bucy RP, Hockett RD, Derdeyn CA. et al
. Initial increase in blood CD4(+) lymphocytes after HIV antiretroviral therapy reflects redistribution from lymphoid tissues. J Clin Invest 1999, 103: 1391–1398.
23. Walker RE, Carter CS, Muul L. et al
. Peripheral expansion of pre-existing mature T cells is an important means of CD4+ T-cell regeneration in HIV-infected adults. Nat Med 1998, 4: 852–856.
© 2002 Lippincott Williams & Wilkins, Inc.