JAIDS Journal of Acquired Immune Deficiency Syndromes:
Pretreatment Factors Associated With 3-Year (144-Week) Virologic and Immunologic Responses to Potent Antiretroviral Therapy
Bosch, Ronald J PhD*; Bennett, Kara MS*; Collier, Ann C MD†; Zackin, Robert ScD*§; Benson, Constance A MD‡; for the AIDS Clinical Trials Group Protocol A5001 Team
From the *Center for Biostatistics in AIDS Research, Harvard University School of Public Health, Boston, MA; †Department of Medicine, University of Washington School of Medicine, Seattle, WA; and ‡Division of Infectious Diseases, University of California, San Diego, San Diego, CA.
Received for publication June 9, 2006; accepted October 20, 2006.
Supported by the National Institutes of Health (AI38855, AI25915, AI25924, AI25868, AI50410, AI27670, AI25897, AI34832, AI27661, AI27673, AI32775, AI25903, AI46383, AI39156, AI25859, AI27659, AI27665, AI27658, AI25879, AI27675, AI27660, AI32770, AI46339, AI46386, AI32782, AI34853, AI46370, AI46376, AI27668, AI27664, AI38858, and AI46381); General Clinical Research Centers, National Institutes of Health (RR00046, RR00096, RR00044, RR00051, RR00047, and RR00052); and HIV Clinical Research Programme, Instituto Superiore di Sanità.
Presented in part at the 10th Conference on Retroviruses and Opportunistic Infections, Boston, MA, February 10-14, 2003 (abstract 572).
Informed consent was obtained from all participating individuals, and guidelines of the US Department of Health and Human Services and the authors' institutions were followed in the conduct of the research.
Reprints: Ronald J. Bosch, PhD, Center for Biostatistics in AIDS Research, Harvard University School of Public Health, Boston MA, 02115 (e-mail: firstname.lastname@example.org).
Objective: To examine pretreatment factors associated with longer term (144 weeks) responses to antiretroviral therapy (ART).
Methods: Of 1498 ART-naive subjects randomized to ART regimens, including ≥3 agents, 1083 patients who had plasma HIV RNA (vRNA) levels and CD4 cell counts at baseline and week 144 were analyzed. Primary baseline factors evaluated were CD4 cell count, vRNA level, gender, race, and age, using multivariable Cox, log-binomial, and linear regression models.
Results: Shorter time to achieving a vRNA level <50 copies/mL was associated with lower baseline vRNA level (P < 0.001), older age (P = 0.007), and lower baseline CD4 cell count (P = 0.055). After adjusting for race, gender, and baseline CD4 cell count, older age was associated with a vRNA level <50 copies/mL at week 144 (P = 0.018). Greater CD4 count increases from baseline to week 144 (mean = 284 cells/μL) were seen in younger men, blacks, and subjects with higher pretreatment vRNA levels; the effect of pretreatment vRNA level was most apparent in women.
Conclusions: Older age was the most important baseline predictor of a vRNA level <50 copies/mL at week 144; lower pretreatment vRNA level and older age were the most important predictors of time to a vRNA level <50 copies/mL. The influence of pretreatment factors on increases in CD4 cell counts differed between men and women.
The availability and widespread use of potent combination antiretroviral therapy (ART) have changed the course of HIV disease in developed countries from one of relentlessly progressive morbidity and mortality to one of a chronic and more manageable infection. Individuals using ART have reduced disease-related symptoms and, in those with advanced disease, longer survival than untreated patients. In clinical trials and prospective observational studies evaluating contemporary ART regimens, the proportion of treatment-naive persons able to achieve initial suppression of their plasma HIV RNA (vRNA) level ranges from 75% to 90%, although a substantial fraction experience loss of virologic control over time.1-8
Factors shown to predict virologic and immunologic responses to ART (and failure of those responses) include pretreatment CD4 cell count, vRNA level, age, gender, regimen tolerability and adherence, and ART regimen potency.1-18 These factors and others have prognostic significance in determining the risk for clinical disease progression and death in those treated with ART.2,4,11,19-21
Most studies, however, are based on relatively limited follow-up or relatively small numbers of patients followed for longer periods. There remains a need for longer term studies that provide a more robust assessment of the durability of responses to ART. Within the context of such longer term observations, there is also a need to define better the population and individual characteristics that influence and/or predict not only the initial response to ART but the durability of responses and the occurrence of and risks for developing adverse effects associated with ART over time.
AIDS Clinical Trials Group (ACTG) protocol A5001 (ACTG Longitudinal Linked Randomized Trial [ALLRT]) is a prospective, longitudinal, randomized trial cohort composed of subjects enrolled in specific ACTG clinical trials in which subjects receive randomized ART regimens, immune-based therapies, or intervention strategies. The A5001 protocol is unique compared with most other observational cohort studies1-5,9-12,15,19,20 in that it only includes persons from studies with specific randomized interventions, follows participants long term regardless of their continued participation in the original trial, and has prospective standardized data collection at specified intervals. These features offer the potential to avoid the treatment selection bias that may be introduced into analyses when clinicians or patients select initial treatments, as is usually the case in observational or nonrandomized clinical cohort studies. The standardized approaches to follow-up and management of toxicities and treatment failure in the context of rigorously conducted clinical trials may also avoid the bias introduced when clinicians independently manage these events in the context of clinical practice.
In the context of the A5001 protocol, we analyzed a variety of pretreatment factors among ART-naive subjects beginning randomized potent combination ART to determine predictors of durable longer term (144 weeks) virologic and immunologic responses.
A total of 1083 ART-naive individuals with vRNA levels and CD4 counts at baseline and 144 weeks after initiating ART were analyzed. Pretreatment predictors of ART responses included in the primary analysis were vRNA level, CD4 cell count, age, gender, and race/ethnicity. We also assessed associations with self-reported adherence at week 144. In supplemental analyses, we examined pretreatment naive CD4 cell percentage (n = 560), activated CD8 cell percentage (n = 568), body mass index (BMI; n = 1039), hemoglobin level (n = 1081) and CD8 cell count (n = 1082). The initial ART regimens were randomly assigned in the 3 contributing ACTG parent protocols: ACTG 384,6,22,23 ACTG 388,14 and ACTG A5014.24 All initial regimens contained 3 or 4 antiretroviral drugs and included at least 1 protease or nonnucleoside reverse transcriptase inhibitor. Regarding subsequent regimens, 51 subjects with a vRNA level <200 copies/mL at least 18 months after initiating treatment were enrolled into an ACTG treatment simplification study (ACTG A5116), and 749 subjects from ACTG 384 were assigned to switch, based on their initial randomization, to specific 3-, 4-, or 5-drug regimens at virologic failure or intolerance.6,22,23 The subjects included in this analysis represent 72% (1083 of 1498 subjects) of the ART-naive participants in these parent studies. Most (311) of the 415 subjects excluded from analysis were not followed in the parent study past week 96 and were not enrolled in ACTG A5001 (including 54 participants at Italian sites who were not eligible for ACTG A5001). Clinical and laboratory evaluations were performed according to the parent protocol and at 16-week intervals specifically for ACTG A5001 after subjects completed the parent protocol.
Plasma vRNA levels were measured by the Roche ultrasensitive assay (Amplicor HIV-1 Monitor; Roche Diagnostic Systems, Branchburg, NJ). CD4 cell counts were obtained on the basis of consensus methodology.25 Laboratories met performance standards of the flow cytometry quality assurance program of the National Institute of Allergy and Infectious Diseases, Division of AIDS. Pretreatment naive (CD45RA+/CD62L+) CD4 cell percentage and activated (CD38+/human leukocyte antigen-D-related-positive [HLA-DR+]) CD8 cell percentage were enumerated by lymphocyte immunophenotyping.25
Baseline measures within 90 days before initiating ART were averaged using the log10 transformation for vRNA level. Wilcoxon rank sum, χ2, and Pearson correlation analyses compared baseline factors. During follow-up, windows were created every 16 weeks, corresponding to the ALLRT protocol visit schedule, and the center-most vRNA level and CD4 cell count measures were selected for each 16-week window.
Success for a binary week 144 viral suppression endpoint was defined as a vRNA level <50 copies/mL at the latest 2 available time points among weeks 112, 128, and 144, although a single vRNA level was used for 69 subjects. Week 48 and 96 time points for a vRNA level <50 copies/mL were also analyzed; if unavailable, a vRNA level obtained 16 weeks earlier was used. Binary regression with a log-binomial model was used to describe effects in terms of relative risks;26 logistic regression results (not shown) were similar. Cox proportional hazards models evaluated factors associated with the time to a vRNA level <50 copies/mL; for this Cox model, the vRNA level was analyzed based on windows created every 4 weeks so as to include the frequent early vRNA level evaluations in the parent protocols.
The week 144 CD4 cell count endpoint was the average of values from weeks 112, 128, and 144. We used linear regression to evaluate the influence of baseline factors on the change in CD4 cell count from baseline to week 144. A secondary analysis examined the change from week 16 to week 144. In addition, we analyzed the subgroup of subjects with sustained viral suppression, defined as a vRNA level <400 copies/mL at every 16-week evaluation starting from week 16; those with a sustained vRNA level <50 copies/mL and those with ≥2 vRNA levels between 50 and 400 copies/mL gave similar results (not shown).
Models were adjusted for the 11 randomized initial ART regimens (effects not shown); results were similar without this adjustment. Effects were compared between male and female subjects, and, when significant, we modeled gender-specific effects. Self-reported adherence using a standardized questionnaire was also examined, evaluating 100% versus <100% adherence to ART. BMI (kg/m2) was categorized into underweight (<18.5 kg/m2), normal (18.5-25 kg/m2), overweight (25-30 kg/m2), and obese (>30 kg/m2).27 Hemoglobin levels were analyzed in terms of their toxicity grade.28 No adjustments have been made for the multiple factors and outcomes analyzed.
Among the 1083 subjects, 83% were men and the median age was 38 years (Table 1); 46% were white/non-Hispanic, 33% were black/non-Hispanic, and 22% were Hispanic/other; they were from 73 sites in the United States, Puerto Rico, and Italy. The median CD4 count and vRNA level were 221 cells/μL and 143,120 copies/mL, respectively. Women had significantly lower vRNA levels than men (median of 83,480 vs. 161,920 copies/mL; P < 0.001) and higher CD4 counts (241 versus 216 cells/μL; P = 0.047).
Baseline characteristics were generally similar for the 415 subjects excluded from these analyses who began ART in the same time interval but did not have week 144 data. They had a lower median age of 35 (vs. 38) years, and fewer were Hispanic/other (16% vs. 22%). Early loss to follow-up was the main reason (297 [72%] of 415 subjects) why subjects were excluded from analyses. Twenty-four excluded subjects died before week 144, 10 of HIV-associated disease.
Plasma HIV RNA Responses and Predictors of Plasma HIV RNA Suppression at Week 144
Nearly all subjects had initial vRNA responses; 1075 (99%) attained at least a 1.5 log10 decline in vRNA level from baseline or a vRNA level <50 copies/mL by week 24. (Initial vRNA responses were similar among those without week 144 data, with 94% [330 of 350 subjects] among those with vRNA levels available during weeks 8 through 24 having this level of response.) By weeks 24 and 144, respectively, 83% and 97% of participants had a vRNA <50 copies/mL at least once. In multivariable Cox models, a shorter time to achieving <50 copies/mL was associated with lower baseline vRNA level (P < 0.001), older age (P = 0.007), and lower baseline CD4 cell count (P = 0.055); gender and race were not additionally predictive.
At weeks 48, 96, and 144, vRNA levels <50 copies/mL were seen in 75% (811 of 1083), 77% (828 of 1080), and 68% (739 of 1083) of subjects, respectively. In a model adjusting for gender, race, pretreatment CD4 cell count, and randomized ART, older age was associated (P = 0.018) and a lower pretreatment vRNA level was marginally associated (P = 0.074) with achieving a week 144 vRNA level <50 copies/mL (Table 2). Similar results were seen when analyzing male and female subjects separately (not shown). Assessment of adherence by self-report at week 144 revealed that older individuals were significantly more likely to report better adherence to their antiretroviral regimen (details not shown). Hemoglobin, BMI, CD8 cell count, CD8 cell activation, and naive CD4 cell percentage were not additionally predictive of week 144 vRNA suppression. In contrast to the week 144 results, lower pretreatment vRNA level was significantly associated with viral suppression at earlier time points (P < 0.001 and P = 0.005) when the same factors were evaluated for their association with week 48 and week 96 vRNA levels <50 copies/mL (models not shown).
CD4 Cell Responses and Predictors of CD4 Cell Count Changes From Baseline to Week 144
The mean increase in the CD4 cell count from baseline to week 144 was 284 cells/μL (median, 25th-75th percentile: 266, 149-393 cells/μL). Evaluating the same pretreatment factors as described previously, we found that a higher baseline vRNA level (P < 0.001) and younger age (P = 0.011) were independently associated with greater CD4 cell increases from baseline to week 144 (models not shown). There was a significant interaction between gender and age (P = 0.006); thus, we then evaluated male- and female-specific age effects (Table 3). A favorable effect of younger age on CD4 cell increases was apparent among male subjects but not among female subjects. No other interactions with gender were apparent (P > 0.35). Subjects with a sustained vRNA level <400 copies/mL had a mean CD4 cell increase of 323 cells/μL; analyses of these subjects further emphasized the favorable effect of younger age for men (P < 0.001). The impact of race/ethnicity on CD4 cell increases was also accentuated, with the largest increases observed among black subjects.
Because initial rises in CD4 cell counts early after initiation of ART have been attributed to redistribution from lymphoid tissue29 rather than to production of naive CD4 cells, the same model was applied to CD4 cell count changes from week 16 to week 144 for subjects with sustained viral suppression. In this group, the mean increase from week 16 to week 144 was 210 cells/μL (median, 25th-75th percentile: 197, 109-303 cells/μL). The effect of pretreatment vRNA level on the change in CD4 cell count from week 16 to week 144 was significantly different between male subjects and female subjects (P = 0.002); thus, we modeled male- and female-specific vRNA level effects. The association of higher pretreatment vRNA levels with greater CD4 cell count increases was apparent for female subjects (P = 0.001) but not for male subjects (P = 0.62). The favorable effect of higher pretreatment vRNA levels on subsequent CD4 cell counts in female subjects remained after additionally adjusting for pretreatment CD8 cell activation in a subset of participants (model not shown). In contrast to these analyses from week 16 to week 144, in the models of change from baseline among all participants (and among those with sustained virologic suppression), a differential effect of pretreatment vRNA level between men and women was not statistically significant, although gender-specific vRNA effects were consistently larger for female subjects (models not shown). As was seen in the model for change from baseline to week 144 in virally suppressed subjects, the adjusted change in CD4 cell count from week 16 to week 144 was larger among blacks and younger male subjects. Lower pretreatment CD4 cell counts were associated with larger CD4 cell count increases from week 16 to week 144 (P = 0.006).
Median CD4 cell counts over time are shown in Figure 1 for all subjects and also for subjects with sustained viral suppression. Most subjects with pretreatment CD4 cell counts less than 200 cells/μL achieved confirmed increases to greater than 200 cells/μL by week 144, including 74% of those who began treatment with CD4 cell counts less than 50 cells/μL (Table 4). In general, CD4 cell count increases and median CD4 cell counts over time were only minimally higher in the subjects with sustained vRNA suppression (see Table 4; see Fig. 1). The substantial CD4 cell count increases in subjects not meeting our definition of sustained vRNA suppression prompted analyses revealing that most (81%) of these subjects had at least 5 vRNA measurements between weeks 16 and 144 that were at least 1.5 log10 less than baseline or less than 400 copies/mL, showing that they had a substantial virologic response despite not maintaining a vRNA level <400 copies/mL.
We evaluated the impact of several other pretreatment factors on change in CD4 cell count from baseline to week 144 by adding each factor to the model (see Table 3). Hemoglobin, CD8 cell count, and activation of CD8 cells were not additionally predictive of CD4 cell increases (P > 0.05 for each). Higher BMI was predictive of larger increases from baseline in CD4 cell count in men (P < 0.001, n = 861; for women: P = 0.72, n = 178). After adjustment for baseline vRNA level, CD4 cell count, age, and race, and relative to men with a pretreatment BMI 18.5 to 25 kg/m2, underweight men had week 144 CD4 cell increases that were 75 cells/μL lower, and men with a pretreatment BMI 25 to 30 kg/m2 and >30 kg/m2 had increases that were 28 and 85 cells/μL higher, respectively. A similar effect of BMI on CD4 cell increases was seen in men with sustained vRNA suppression (P < 0.001). Consistent with the significant effect of younger age on CD4 cell count increases among men, we also found that there was a significant positive association between baseline naive CD4 cell percentage and CD4 cell count change from baseline (P < 0.001; n = 467 men, 24 additional CD4 cells/μL per each 10% higher baseline naive CD4 cell percentage). Baseline naive CD4 cell percentage was more significantly associated with CD4 cell count increases than baseline naive CD4 cell count, and this marker was positively correlated with baseline CD4 cell count (r = 0.34, 95% confidence interval [CI]: 0.25 to 0.41). As expected, the strength of the association between younger age and CD4 cell count increases among men was diminished when naive CD4 cell percentage was added to the model (from P < 0.001 to P = 0.029; n = 467), consistent with the negative correlation between the 2 factors (r = −0.31, 95% CI: −0.39 to −0.23). In women, higher baseline naive CD4 cell percentage was similarly associated with larger CD4 cell count increases (P = 0.036, model not shown) and with baseline CD4 cell count.
To inform clinical decision making further relative to the timing of initiation of potent combination ART, we evaluated the influence of a defined set of pretreatment factors on the vRNA level and CD4 cell count responses 144 weeks after initiating ART, because improvements in these parameters have been associated with a reduced risk of clinical events and disease progression.2,20,21 This analysis of longer term responses to potent ART describes one of the larger ART-naive cohorts to date and is unique in that all subjects initially received ART regimens that were randomly assigned in the clinical trials that served as the platform for the recruitment of the cohort. We targeted clinically relevant features of the durable response to potent ART by analyzing CD4 cell counts and vRNA levels in the 3 years after initiating treatment rather than shorter term endpoints,1,3,4,9,10,12,15 assessing CD4 cell count responses among those with sustained virologic suppression, and defining vRNA suppression as 2 consecutive values <50 copies/mL. In our models, we adjusted for the treatment arms in the contributing randomized trials, but treatment regimen differences were not the focus of this investigation.
Nearly all subjects had an initial vRNA response to ART, with 97% achieving a vRNA level <500 copies/mL by week 32 after initiation of treatment, contrasting with 85% of subjects in a pooled clinic-based cohort.4 This and other potential differences between the characteristics of our clinical trial population compared with observational clinical cohorts should be considered in interpreting our findings. Differences include random assignment of specified initial ART regimens in the context of a clinical study rather than with regimens selected by clinicians and/or patients and prospective follow-up with a uniform set of measurements at prescribed intervals rather than according to clinical practice. These features serve to minimize initial treatment selection bias and bias introduced by differential follow-up that might influence responses described in observational cohort studies. Our analyses confirmed a key finding from other studies,4,11 showing that HIV-infected individuals with a higher pretreatment vRNA level take a longer time to achieve vRNA suppression below assay limits.
Older age was the most significant pretreatment factor associated with longer term (week 144) vRNA suppression. Additionally, after controlling for pretreatment vRNA levels, older individuals showed a shorter time to achieve vRNA suppression. Others9,11,30-33 have also reported better viral suppression in older individuals in adjusted analyses and have indicated that older age is related to better adherence (as we also demonstrated12,34,35), fewer treatment interruptions,31 and possibly to other factors that influence viral replication.33 Although improved vRNA suppression in older individuals corresponded to similar age-related effects on adherence in our study, potential confounders that were not collected in ALLRT include age-related differences in socioeconomic factors,33 access to medical services, and the prevalence of sexually transmitted diseases,33 which might influence HIV replication. A lower pretreatment vRNA level was marginally associated with a week 144 vRNA level <50 copies/mL but was significantly related to vRNA suppression at earlier time points. This finding suggests that influences other than pretreatment viral burden, such as regimen tolerability, adherence, and drug resistance,36 play an increasing role over time in maintaining viral suppression, diluting the effect of pretreatment vRNA level over the longer term.
CD4 cell count increases were seen across the range of pretreatment CD4 cell counts, even among those who initiated ART with CD4 counts <50 cells/μL. Median CD4 cell counts continued to increase over time for those with baseline CD4 counts less than 350 cells/μL, consistent with other reports.37,38 One hundred forty-four weeks after initiating ART, most of those who started ART with a CD4 count <350 cells/μL still had CD4 cell counts below normal levels, however.25 Because we have analyzed the same group of subjects over time, these findings of longitudinal increases are likely to be more clinically applicable than analyses in which the number of evaluated subjects substantially diminishes over time.37
As in other studies, younger age was a significant independent predictor of greater CD4 cell count increases for the analyses of change from baseline to week 144 and of change from week 16 to week 144.16,18,37,39-41 This age effect was only apparent among men, however. The smaller number of women included may have limited our statistical power to detect an age effect in women, although the age distribution was similar for men and women (median, 25th-75th percentile: 38, 33-45 for men and 38, 32-45 for women). This age effect likely is related to thymic function, which declines with age.29,40 Pretreatment naive CD4 cell percentage was significantly and positively associated with greater CD4 cell count increases in men and women. Including this biomarker of immunologic potential42-45 in our models diminished the impact of younger age, as would be expected for 2 covariates that reflect immune reconstitution potential. A lower pretreatment CD4 cell count was also a significant predictor of greater CD4 cell count increases from week 16 to week 144 in those with sustained vRNA suppression. Greater CD4 cell count increases in subjects with lower levels at baseline have also been observed in other cohorts,11,37,46 and others have postulated that this effect represents the recovery of CD4 cell counts toward a normal range, such that those starting therapy with lower values have more opportunity to increase in response to ART.46 Considered together, the greater CD4 increases seen with lower baseline CD4 cell counts but higher baseline naive CD4 cell percentages suggest that these 2 pretreatment factors represent different host functional characteristics (ie, a higher naive CD4 cell percentage may represent better thymic reserve function, whereas a lower pretreatment CD4 cell count may reflect longer duration of infection or greater viral burden or pathogenicity). Combining these values and analyzing the relation between pretreatment naive CD4 cell counts and subsequent CD4 cell count increases might negate these opposing effects.40,44
Among those with sustained vRNA suppression, black participants had larger CD4 cell count increases than white subjects after adjustment for other pretreatment factors. Smith et al46 also reported a trend toward larger long-term CD4 cell count increases in nonwhite subjects. We also identified a relation between higher pretreatment BMI and greater CD4 cell increases in men, which has not been previously reported. This BMI effect remained when we restricted the analysis to those with sustained vRNA suppression and when we further adjusted for pretreatment naive CD4% (details not shown). It is possible that BMI is another indirect marker of immunologic reserve (ie, availability of micronutrients needed for cellular replication may be related to this effect). The relation between BMI and CD4 cell responses to ART warrants further study. Interestingly, other studies have shown that a higher BMI is also associated with slower clinical HIV disease progression after adjustment for other factors.47-49 It is unclear if this effect could be mediated by more favorable immune responses.
The effect of pretreatment vRNA level on the change in CD4 cell count from week 16 to week 144 was different in men and women; higher pretreatment vRNA level was associated with a greater increase in CD4 cell count from week 16 to week 144 only for women. This effect of pretreatment vRNA level in women but not men was also seen in supplemental analyses of CD4 cell count changes from week 48 to 144 (details not shown). Compared with the men, women in our study had lower pretreatment vRNA levels, which is a finding that is consistent with many other reports but is as yet unexplained.50-52 The greater influence in women of pretreatment vRNA level on CD4 cell count increases may be related to the mechanisms contributing to the lower vRNA levels seen in women or their disease progression at lower vRNA levels50 or, similarly, their higher progression rates adjusted for CD4 cell counts and vRNA levels.21,50 A relation between gender-related steroids and immune modulation related to chemokine receptor expression and regulation of immune inflammatory mediators that may influence viral replication and disease progression has been postulated.50-52 Similar relations, should they be confirmed in HIV-infected women, may also play a role in the association between the pretreatment vRNA level and increases in CD4 cell counts from week 16 to week 144 in women demonstrated in our study.
Ultimately, the decision to initiate ART is influenced by a number of pretreatment factors other than vRNA level and CD4 cell count that may lessen the impact of our findings relative to recommendations for the optimal timing for initiating ART.53,54 Participants in our analyses were followed long term after being randomly assigned to potent combination ART regimens within prospective clinical trials, however, minimizing the relation between pretreatment patient characteristics and the choice of specific ART regimens that were initiated. Although the potency and tolerability of different regimens used in the studies that served as the platform for recruitment of our subjects may have influenced the initial and subsequent virologic responses, the overall proportion of those with substantial and durable responses was high throughout follow-up period, suggesting that regimen potency may only be one of several factors that influence the long-term response to ART. Our findings that the impact of pretreatment factors differed between men and women deserve further study. Although no gender differences were noted in recent large-sample analyses of responses to ART,11 those analyses did not assess whether there were differential influences of pretreatment factors between men and women. Careful characterization of such differences could have implications for the criteria used to determine the timing of ART initiation among women versus men, which is an issue that has been discussed in the context of vRNA differences between women and men.51 Pretreatment naive CD4 cell percentages were lower in those with lower CD4 cell counts in men and women, suggesting that in addition to their decline with age, naive CD4 cell percentages decline with advanced immunosuppression in untreated HIV infection. The role of the naive CD4 cell subset as yet another factor to consider in determining when to start ART remains to be more specifically defined.
Our study has several limitations. Of the 1498 participants originally randomized in the clinical trials that determined their involvement in the cohort, 415 were not followed through week 144. Most of these completed their parent study but were unwilling to be followed long term in the ACTG A5001 protocol, possibly because of poorer health or other confounding patient characteristics. In addition, patients followed in clinical trials, particularly those who volunteer to be followed long term, may not be representative of all patients with HIV infection.2,9 Further, we restricted analyses to those with week 144 data, although alternate approaches could have been used to control for those subjects missing week 144 data, such as carrying forward the last observed value. Nevertheless, there are statistical biases also inherent in that approach, and we elected to use a more definitive endpoint for this study. Finally, we considered pretreatment CD4 cell count as the primary indicator of HIV disease severity and degree of immunosuppression. Diagnoses of prior opportunistic infections are correlated with low CD4 cell counts; thus, we chose not to incorporate prior diagnoses into our pretreatment models. These or other potential unmeasured confounders (eg, host genetic factors) could have a further impact on our findings, however.
In summary, we have shown that a number of pretreatment factors in addition to pretreatment CD4 cell count and vRNA level are important predictors of longer term virologic and immunologic responses to potent ART. This study represents an initial analysis of the ACTG A5001/ALLRT cohort, which continues in long-term follow-up; other analyses are ongoing, including the influence of initial ART regimens on long-term responses, an assessment of opportunistic infections occurring after starting ART, and host genetic factors that influence treatment responses. Although determining the optimal strategies for initiation of ART, as defined by patient characteristics and clinically available biomarkers, might be best addressed by large, long-term, randomized studies, changes in treatment paradigms and improvements in the efficacy and toxicities of treatments over time pose significant challenges to acquiring such data. In the absence of such studies, analyses of data generated in subjects followed prospectively and long term in cohorts that incorporate strategies to minimize some of the selection bias of conventional observational clinical cohorts may provide useful insights for making individual treatment decisions (eg, emphasizing adherence counseling, particularly for younger patients). In addition, understanding the subject characteristics predictive of longer term virologic and immunologic responses may provide insights into the design of strategies to improve these responses.
The authors thank the ACTG sites and study participants for their time and effort, Frontier Science Foundation for data management, and Mizue Krygowski and Marlene Smurzynski for their valuable contributions. They gratefully acknowledge the contributing protocol chairs: G. Robbins and R. Shafer for ACTG 384, M. Fischl for ACTG 388, and A. Landay and M. Lederman for ACTG A5014. The authors dedicate this work to Robert Zackin.
1. Deeks SG, Hecht FM, Swanson M, et al. HIV RNA and CD4 cell count response to protease inhibitor therapy in an urban AIDS clinic: response to both initial and salvage therapy. AIDS
2. Ledergerber B, Egger M, Opravil M, et al. Clinical progression and virological failure on highly active antiretroviral therapy in HIV-1 patients: a prospective cohort study. Swiss HIV Cohort Study. Lancet
3. Wit FW, van Leeuwen R, Weverling GJ, et al. Outcome and predictors of failure of highly active antiretroviral therapy: one-year follow-up of a cohort of human immunodeficiency virus type 1-infected persons. J Infect Dis
4. Phillips AN, Staszewski S, Weber R, et al. HIV viral load response to antiretroviral therapy according to the baseline CD4 cell count and viral load. JAMA
5. Phillips AN, Miller V, Sabin C, et al. Durability of HIV-1 viral suppression over 3.3 years with multi-drug antiretroviral therapy in previously drug-naïve individuals. AIDS
6. Robbins GK, DeGruttola V, Shafer RW, et al. Comparison of sequential three-drug regimens as initial therapy for HIV-1 infection. N Engl J Med
7. Gulick RM, Ribaudo HJ, Shikuma CM, et al. Triple-nucleoside regimens versus efavirenz-containing regimens for the initial treatment of HIV-1 infection. N Engl J Med
8. Hicks C, King MS, Gulick RM, et al. Long-term safety and durable antiretroviral activity of lopinavir/ritonavir in treatment-naïve patients: 4 year follow-up study. AIDS
9. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions. Ann Intern Med
10. Staszewski S, Miller V, Sabin C, et al. Determinants of sustainable CD4 lymphocyte count increases in response to antiretroviral therapy. AIDS
11. Moore AL, Kirk O, Johnson AM, et al, for the EuroSIDA group. Virologic, immunologic, and clinical response to highly active antiretroviral therapy: the gender issue revisited. J Acquir Immune Defic Syndr
12. Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med
13. Staszewski S, Morales-Ramirez J, Tashima KT, et al. Efavirenz plus zidovudine and lamivudine, efavirenz plus indinavir, and indinavir plus zidovudine and lamivudine in the treatment of HIV-1 infection in adults. Study 006 Team. N Engl J Med
14. Fischl MA, Ribaudo HJ, Collier AC, et al. A randomized trial of 2 different 4-drug antiretroviral regimens versus a 3-drug regimen, in advanced human immunodeficiency virus disease. J Infect Dis
15. Matthews GV, Sabin CA, Mandalia S, et al. Virological suppression at 6 months is related to choice of initial regimen in antiretroviral-naïve patients: a cohort study. AIDS
16. Yamashita TE, Phair JP, Muñoz A, et al. Immunologic and virologic response to highly active antiretroviral therapy in the Multicenter AIDS Cohort Study. AIDS
17. Jacobson LP, Phair JP, Yamashita TE. Update on the virologic and immunologic response to highly active antiretroviral therapy. Curr Infect Dis Rep
18. Viard JP, Mocroft A, Chiesi A, et al, for the EuroSIDA Study Group. Influence of age on CD4 cell recovery in human immunodeficiency virus-infected patients receiving highly active antiretroviral therapy: evidence from the EuroSIDA study. J Infect Dis
19. Egger M, May M, Chene G, et al, and ART Cohort Collaboration. Prognosis of HIV-1-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet
20. Antiretroviral Therapy (ART) Cohort Collaboration. Prognostic importance of initial response in HIV-1 infected patients starting potent antiretroviral therapy: analysis of prospective studies. Lancet
21. Gebo KA, Gallant JE, Keruly JC, et al. Absolute CD4 vs. CD4 percentage for predicting the risk of opportunistic illness in HIV infection. J Acquir Immune Defic Syndr
22. Smeaton LM, DeGruttola V, Robbins GK, et al. ACTG (AIDS Clinical Trials Group) 384: a strategy trial comparing consecutive treatments for HIV-1. Control Clin Trials
23. Shafer RW, Smeaton LM, Robbins GK, et al. Comparison of four-drug regimens and pairs of sequential three-drug regimens as initial therapy for HIV-1 infection. N Engl J Med
24. Landay AL, Spritzler J, Kessler H, et al. Immune reconstitution is comparable in antiretroviral-naive subjects after 1 year of successful therapy with a nucleoside reverse-transcriptase inhibitor- or protease inhibitor-containing antiretroviral regimen. J Infect Dis
25. Valdez H, Connick E, Smith KY, et al. Limited immune restoration after 3 years' suppression of HIV-1 replication in patients with moderately advanced disease. AIDS
26. Skov T, Deddens J, Petersen MR, et al. Prevalence proportion ratios: estimation and hypothesis testing. Int J Epidemiol
27. National Heart Blood and Lung Institute. BMI calculator with categories found. Available at: http://nhlbisupport.com/bmi/
. 2005. Accessed November 16, 2006.
29. Lederman MM. Immune restoration and CD4+ T-cell function with antiretroviral therapies. AIDS
30. Paredes R, Mocroft A, Kirk O, et al. Predictors of virological success and ensuing failure in HIV-positive patients starting highly active antiretroviral therapy in Europe: results from the EuroSIDA study. Arch Intern Med
31. Wellons MF, Sanders L, Edwards LJ, et al. HIV infection: treatment outcomes in older and younger adults. J Am Geriatr Soc
32. Grabar S, Kousignian I, Sobel A, et al. Immunologic and clinical responses to highly active antiretroviral therapy over 50 years of age. Results from the French Hospital Database on HIV. AIDS
33. Goodkin K, Shapshak P, Asthana D, et al. Older age and plasma viral load in HIV-1 infection. AIDS
. 2004;18(Suppl 1):S87-S98.
34. Glass TR, De Geest S, Weber R, et al. Correlates of self-reported nonadherence to antiretroviral therapy in HIV-infected patients: the Swiss HIV Cohort Study. J Acquir Immune Defic Syndr
35. Carrieri MP, Leport C, Protopopescu C, et al. Factors associated with nonadherence to highly active antiretroviral therapy: a 5-year follow-up analysis with correction for the bias induced by missing data in the treatment maintenance phase. J Acquir Immune Defic Syndr
36. Hirsch MS, Brun-Vezinet F, Clotet B, et al. Antiretroviral drug resistance testing in adults infected with human immunodeficiency virus type 1: 2003 recommendations of an International AIDS Society-USA Panel. Clin Infect Dis
37. Hunt PW, Deeks SG, Rodriguez B, et al. Continued CD4 cell count increases in HIV-infected adults experiencing 4 years of viral suppression on antiretroviral therapy. AIDS
38. Smith K, Aga E, Bosch RJ, et al. Long-term changes in circulating CD4 T lymphocytes in virologically suppressed patients after 6 years of highly active antiretroviral therapy. AIDS
39. Kaufmann GR, Bloch M, Finlayson R, et al. The extent of HIV-1-related immunodeficiency and age predict the long-term CD4 T lymphocyte response to potent antiretroviral therapy. AIDS
40. Cohen Stuart J, Hamann D, Borleffs J, et al. Reconstitution of naive T cells during antiretroviral treatment of HIV-infected adults is dependent on age. AIDS
41. Kaufmann GR, Perrin L, Pantaleo G, et al. CD4 T-lymphocyte recovery in individuals with advanced HIV-1 infection receiving potent antiretroviral therapy for 4 years. Arch Intern Med
42. 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
43. Hengel RL, Jones BM, Kennedy MS, et al. Lymphocyte kinetics and precursor frequency-dependent recovery of CD4(+)CD45RA(+)CD62L(+) naive T cells following triple-drug therapy for HIV type 1 infection. AIDS Res Hum Retroviruses
44. Michael CG, Kirk O, Mathiesen L, et al. The naive CD4+ count in HIV-1-infected patients at time of initiation of highly active antiretroviral therapy is strongly associated with the level of immunological recovery. Scand J Infect Dis
45. Mildvan D, Bosch RJ, Kim RS, et al. Immunophenotypic markers and antiretroviral therapy (IMART): T cell activation and maturation help predict treatment response. J Infect Dis
46. Smith CJ, Sabin CA, Youle MS, et al. Factors influencing increases in CD4 cell counts of HIV-positive persons receiving long-term highly active antiretroviral therapy. J Infect Dis
47. Shor-Posner G, Campa A, Zhang G, et al. When obesity is desirable: a longitudinal study of the Miami HIV-1-Infected Drug Abusers (MIDAS) cohort. J Acquir Immune Defic Syndr
48. Shuter J, Chang CJ, Klein RS. Prevalence and predictive value of overweight in an urban HIV care clinic. J Acquir Immune Defic Syndr
49. Jones CY, Hogan JW, Snyder B, et al, for the HIV Epidemiology Research Study Group. Overweight and human immunodeficiency virus (HIV) progression in women: associations HIV disease progression and changes in body mass index in women in the HIV epidemiology research study cohort. Clin Infect Dis
. 2003;37(Suppl 2):S69-S80.
50. Farzadegan H, Hoover DR, Astemborski J, et al. Sex differences in HIV-1 viral load and progression to AIDS. Lancet
51. Gandhi M, Bacchetti P, Miotti P, et al. Does patient sex affect human immunodeficiency virus levels? Clin Infect Dis
52. Touloumi G, Pantazis N, Babiker AG, et al, for the CASCADE Collaboration. Differences in HIV RNA levels before the initiation of antiretroviral therapy among 1864 individuals with known HIV-1 seroconversion dates. AIDS
53. Hammer SM, Saag MS, Schechter M, et al. Treatment for adult HIV infection: 2006 recommendations of the International AIDS Society-USA Panel. JAMA
54. Wood E, Hogg RS, Yip B, et al. Using baseline CD4 cell count and plasma HIV RNA to guide the initiation of highly active antiretroviral therapy. Rev Invest Clin
The following institutions and investigators participated in the contributing ACTG studies: R. Murphy and B. Berzins (Northwestern University, Evanston, IL); B. Sha and J. Fritsche (Rush University, Chicago, IL); O. Adeyemi and J. Despotes (Cook County Hospital, Chicago, IL); S. Koletar and D. Gochnour (Ohio State University, OH); C. Marcus and J. Eron (University of North Carolina, Chapel Hill, NC); T. Lane and L. Dasnoit (Moses Cone Hospital, Greensboro, NC); L. Meixner and T. Cotlon-Pineda (University of California, San Diego, San Diego, CA); J. Feinberg and D. Daria (University of Cincinnati, Cincinnati, OH), J. Santana and O. Méndez (University of Puerto Rico, PR); H. Balfour and C. Fietzer (University of Minnesota, Minneapolis, MN); J. Katseres and J. Stapleton (University of Iowa, Iowa City, IA); S. Swindells and F. Van Meter (University of Nebraska, Omaha, NE); J. Leedom and V. Clemente (University of Southern California, Los Angeles, CA); M. Saag and J. Lennox (University of Alabama at Birmingham, Birmingham, AL, and Emory University, Atlanta, GA); D. Clifford and L. Kessels (Washington University, St. Louis, MO); D. McMahon and B. Rutecki (University of Pittsburgh, Pittsburgh, PA); P. Kumar and K. Hawkins (Georgetown University, Washington, DC); J. Bartlett and M. Silberman (Duke University, Durham, NC); M. Goldman and H. Rominger (Indiana University, Indianapolis, IN); J. Black and B. Zwickl (Methodist Hospital of Indiana, Indianapolis, IN); M. Dube and G. Clement (Wishard Hospital, Indianapolis, IN); A. Sbrolla and K. Habeeb (Massachusetts General Hospital, Boston, MA); H. Fitch and N. Kim (Beth Israel Deaconess Medical Center, Boston, MA); P. Skolnik and B. Adams (Boston Medical Center, Boston, MA); P. Sax and L. Dumas (Brigham and Women's Hospital, Boston, MA); C. Gonzalez and S. Holland (New York University, Bellevue, NY); J. Jacobson and M. Dolan (Mount Sinai Medical Center, New York, NY); J. Reid and R. Reichman (University of Rochester Medical Center, Rochester, NY); T. O'Hara and R. Cruz (State University of New York, Buffalo, NY); N. El-Daher and M. Shoemaker (McCree McCuller Wellness Center, Rochester, NY); C. Hurley and R. Corales (AIDS Community Health Center, Rochester, NY); M. Chance, K. Medvik, K. Shina, and J. Baum (Case Western Reserve University, Cleveland, OH); K. Whitely and M. Wild (MetroHealth Medical Center, Cleveland, OH); J. Castro and M. Fischl (University of Miami, Miami, FL); J. Currier, S. Chafey (University of California, Los Angeles School of Medicine, Los Angeles, CA); M. Witt and D. Duran (Harbor General Hospital, University of California, Los Angeles, Los Angeles, CA); B. Putnam and C. Basler (University of Colorado Health Sciences Center, Denver, CO); J. Nicotera and D. Haas (Vanderbilt University, Nashville, TN); I. Frank and I. Matozzo (University of Pennsylvania, Philadelphia, PA); J. Noel-Connor and M. Torres (Columbia University, New York, NY); V. Hughes and R. Gulick (Cornell University, New York, NY); M. Glesby and T. Stroberg (Chelsea Clinic, New York, NY); A. Zolopa and S. Valle (Stanford University, Stanford, CA); S. Stoudt (San Mateo County AIDS Program, San Mateo, CA); D. Slamowitz and P. Cain (Santa Clara Valley Medical Center, San Jose, CA); J. Norris (Willow Clinic, Menlo Park, CA); W. O'Brien and G. Casey (University of Texas, Galveston, TX); N. Hanks and J. Frederick (University of Hawaii, Honolulu, HI); D. Mildvan and G. Costantini (Beth Israel Medical Center, New York, NY); C. Lohner and D. Margolis (University of Texas, Southwestern Medical Center, Dallas, TX); C. Flexner and I. Wiggins (Johns Hopkins University, Baltimore, MD); J. Schouten and N. J. Conley (University of Washington, Seattle, WA); M. Payne and C. B. Hare (University of California, San Francisco, San Francisco General Hospital, San Francisco, CA); J. Volinski and C. Lindquist (Marin County Specialty Clinic, Greenbrae, CA); R. Redfield, C. Davis, and O. Erondu (University of Maryland, Baltimore, MD); K. Tashima and P. Poethke (Miriam Hospital, Providence, RI); R. Pollard and N. Fitch (University of California, Davis, Davis, CA); S. Vella, A. Chiesi, R. Arcieri, M. Pirillo, C. Galluzzo, E. Germinario, R. Amici, M. Marzi, A. Nobile, R. Di Nallo, and C. Polizzi (Istituto Superiore di Sanita, Rome, Italy); O. Coronado and G. Fasulo (Ospedale Maggiore, Bologna, Italy); G. Carosi and F. Castelli (Spedali Civili, Brescia, Italy); M. Di Pietro and F. Vichi (Ospedale S.M. Annunziata, Florence, Italy); G. Sterrantino and S. Ambu (Ospedale Careggi, Florence, Italy); A. Cargnel, P. Meraviglia, F. Niero, and A. Capetti (Ospedale Luigi Sacco, Milan, Italy); M. Soranzo and A. Macor (Ospedale Amadeo Di Savoia, Turin, Italy); G. d'Ettorre and G. Forcina (Universita di Roma, Rome, Italy); D. Bassetti and A. Di Biagio (Universita di Genova, Genoa, Italy); F. Ghinelli and L. Sighinolfi (Archispedale S. Anna, Ferrara, Italy); A. Riva and G. Scalise (Azienda Ospedaliera Umberto I, Ancona, Italy); D. Santoro and E. Rinaldi (Ospedale Sant'Anna, Como, Italy); G. Guaraldi and R. Esposito (Universita delgi Studi di Modena, Modena, Italy); C. Ferrari and G. Pasetti (Azienda Ospedaliera di Parma, Parma, Italy); N. Abrescia, A. Busto, A. Chirianni, M. Gargiulo, C. Izzo, and C. Sbreglia (Azienda Ospedaliera D. Cotugno, Naples, Italy); F. Alberici and D. Sacchini (Azienda U.S.L. of Piacenza-Ospedale Civile, Piacenza, Italy); and G. Magnani and G. Zoboli (Arcispedale S. Maria Nuova, Reggio Emilia, Italy).
antiretroviral therapy; CD4+ cell count; combination therapy; HIV RNA level
© 2007 Lippincott Williams & Wilkins, Inc.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Highlight selected keywords in the article text.
Data is temporarily unavailable. Please try again soon.