http://orcid.org/0000-0002-8449-222X
Figures
Abstract
Background
Youth have residual thymic tissue and potentially greater capacity for immune reconstitution than adults after initiation of combination antiretroviral therapy (cART). However, youth face behavioral and psychosocial challenges that may make them more likely than adults to delay ART initiation and less likely to attain similar CD4 outcomes after initiating cART. This study compared CD4 outcomes over time following cART initiation between ART-naïve non-perinatally HIV-infected (nPHIV) youth (13–24 years-old) and adults (≥25–44 years-old).
Methods
Retrospective analysis of ART-naïve nPHIV individuals 13–44 years-old, who initiated their first cART between 2008 and 2011 at clinical sites in the HIV Research Network. A linear mixed model was used to assess the association between CD4 levels after cART initiation and age (13–24, 25–34, 35–44 years), accounting for random variation within participants and between sites, and adjusting for key variables including gender, race/ethnicity, viral load, gaps in care (defined as > 365 days between CD4 tests), and CD4 levels prior to cART initiation (baseline CD4).
Results
Among 2,595 individuals (435 youth; 2,160 adults), the median follow-up after cART initiation was 179 weeks (IQR 92–249). Baseline CD4 was higher for youth (320 cells/mm3) than for ages 25–34 (293) or 35–44 (258). At 239 weeks after cART initiation, median unadjusted CD4 was higher for youth than adults (576 vs. 539 and 476 cells/mm3, respectively), but this difference was not significant when baseline CD4 was controlled. Compared to those with baseline CD4 ≤200 cells/mm3, individuals with baseline CD4 of 201–500 and >500 cells/mm3 had greater predicted CD4 levels: 390, 607, and 831, respectively. Additionally, having no gaps in care and higher viral load were associated with better CD4 outcomes.
Citation: Agwu AL, Fleishman JA, Mahiane G, Nonyane BAS, Althoff KN, Yehia BR, et al. (2017) Comparing longitudinal CD4 responses to cART among non-perinatally HIV-infected youth versus adults: Results from the HIVRN Cohort. PLoS ONE 12(2): e0171125. https://doi.org/10.1371/journal.pone.0171125
Editor: Alan Landay, Rush University, UNITED STATES
Received: September 20, 2016; Accepted: January 15, 2017; Published: February 9, 2017
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: The authors confirm that all data underlying the study findings are fully available without restriction from the HIV Research Network (HIVRN) cohort https://cds.johnshopkins.edu/hivrn/. The network invites collaboration and shares data. However, due to ethical restrictions from the study’s informed consent documents and the HIVRN human subject protection plan, the data are available upon formal request and subsequent review and approval by the HIVRN data subcommittee. A formal request must be made to the study data manager, Rebecca Diaz Reyes (rdiazre1@jhu.edu) or https://cds.johnshopkins.edu/hivrn/.
Funding: Supported by the Agency for Healthcare Research and Quality (contract HHSA2902011000007C), and the National Institute on Aging (R01 AG026250) (K.A.G.), and the Johns Hopkins Center for AIDS Research (1P30AI094189). Dr. Agwu was supported by the National Institutes of Allergy and Infectious Diseases (1K23 AI084549). Dr. Berry was supported by the National Institutes of Health (K23 AI084854)). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: All other authors have declared that no competing interests exist.
Introduction
Thirty years into the HIV epidemic, effective combination antiretroviral therapy (cART) has transformed HIV into a chronic disease. Life expectancy for persons living with HIV (PLHIV) has improved dramatically, mirroring that of uninfected adults. [1,2] However, non-perinatally HIV-infected (nPHIV) youth between the ages of 13 and 24 comprise a unique and challenging group, in which the incidence of HIV infection is increasing. [3] Youth tend to have higher CD4 counts than adults when presenting for care, though 40% have CD4 counts below 350 cells/mm3 at presentation. [4,5] [6–10]
Increasingly, studies support earlier initiation of cART [11,12] and guidelines now recommend initiation of cART for all PLHIV, regardless of CD4 count. [13,14] However, studies have shown that youth are less likely than adults to initiate cART when meeting treatment criteria. [5,15] Age, their cognitive developmental stage (concrete thinking and invincibility), and their high rates of social marginalization and stigma, make diagnosing, engaging, and treating youth difficult. [16–18] For example, youth have higher rates of nonadherence to cART and attrition from care than adults. [6–10] Youth have residual thymic tissue and may potentially have greater capacity for immune reconstitution than adults. [19–23] However, it is not clear if the potential for enhanced CD4 recovery for youth, given their residual thymic tissue, compensates for their delays in initiating cART.
Unfortunately, studies examining CD4 responses in PLHIV have had limited capacity to look specifically at youth, as they commonly use broad age ranges (e.g., 18–39) and do not distinguish youth. [21,24,25] Further, other studies of youth have been conducted in clinical trials; generalization to non-trial settings may be limited. [8,26,27] In this study, we compared CD4 responses following cART initiation between nPHIV youth (13–24 years-old) and adults (≥25–44 years-old) followed in a large observational clinical cohort.
Methods
This retrospective study compared trends in CD4 levels over time after initiation of cART between nPHIV youth aged 13–24 and adults aged 25 through 44 years at study entry. Studies that have examined differences in immunologic responses between young and older adults have used varying age comparisons,[28–30] e.g., 18–30 vs. 30–40, 40–50, 50-<60 and ≥60 [31], ≥16-<32.7 vs. 32.7–37.4, 27.2–44.4, and ≥44.5 [32], <50 vs. ≥50 years of age. [24,33] Differences between younger and older adults tended to emerge when comparing younger individuals with those ≥50 [24], with a few studies finding differences when comparing younger individuals to those > = 44.5 and 40–50 years of age.[31,32] Notably, most of the studies lumped all younger adults (e.g., 18–49) and had limited numbers of the youngest patients (i.e., <24), making it challenging to decipher differences in immune responses to cART among the youngest age groups. We therefore intentionally chose 45 as the upper age cutoff for our study to specifically compare CD4 response to cART among individuals where it would least be expected, by the cumulative evidence, that there should be differences and where any differences seen would not be attributed to the expected changes with older age.
Eligible patients were recruited from those treated in HIV clinics participating in the HIV Research Network (HIVRN), a consortium of 19 U.S. clinic sites that provide primary and subspecialty care to HIV-infected patients.[34,35] As described previously, HIVRN sites abstract specified data elements from patients’ medical records, including demographic data, service utilization, medications, and laboratory tests; abstracted data are assembled into a single database after quality assurance review. [34,35] The study was approved by the Johns Hopkins University School of Medicine Institutional review board (IRB) and the IRBs of each participating institution.
Eligibility criteria
Patients were eligible if they were at least 12 years old but less than 45 years old, enrolled in the HIVRN clinic between January 1, 2008 and December 31, 2011, acquired HIV through risk behaviors (e.g., sexual activity, injection drug use), were ART-naïve at clinic enrollment, actively receiving HIV care (defined as having at least one CD4 count and one outpatient HIV provider visit within a calendar year), and initiated cART at some point between enrollment and December 31, 2011. In the analytic sample, the minimum age was 13. To focus on a therapy-naïve group, participants who had prior history of ART usage and whose first recorded HIV-1 RNA test was ≤400 copies/mL were excluded.
Data collection and measures
Demographic and clinical data were collected from the HIVRN clinical database. CART was defined as concomitant use of ≥ 3 antiretroviral drugs either from ≥ 2 classes (nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs), non-nucleoside reverse transcriptase inhibitors, protease inhibitors (PIs), integrase inhibitors, and entry inhibitors) or 3 NRTIs that qualified as cART (i.e., abacavir, lamivudine, and zidovudine). The date of initiation of cART was also recorded. Initial cART therapy was categorized as NNRTI, PI, other, or unknown.
Each HIVRN site recorded the dates and outcomes of all CD4 tests performed during the observation period of Jan 1, 2008 –December 31, 2014. The baseline CD4 test was the one performed closest in time prior to the cART initiation date and ≤16 weeks prior to cART initiation. If a patient had no recorded CD4 prior to cART initiation, a CD4 test performed up to 7 days after cART initiation was used. Patients with no CD4 test in the appropriate time window to establish a baseline were excluded. Baseline CD4 level was categorized as ≤200, 201–500, and >500 cells/mm3.
The age when each patient initiated cART was calculated from month and year of birth and the date of cART initiation. Age was categorized as 13–24, 25–34, and 35–44 years. Self-identified racial/ethnic group was classified as non-Hispanic White, non-Hispanic Black, Hispanic, or “other” (American Indian or Alaskan Native, Asian or Pacific Islander). HIV transmission group was classified as men who have sex with men (MSM), injection drug use (IDU), heterosexual activity (HET), or other.
Dates and results of all HIV-1 RNA (viral load (VL)) tests during the observation period were recorded. For analysis, VL values were categorized as ≤400, 401–1000, and >1000 copies/mL. VL tests were matched with CD4 tests by date. Most (87%) CD4 tests had a VL test on the same date. If no exact match by date existed, the VL test closest in time to the CD4 test (preferentially prior) was used. For the 3,270 CD4 records without an exact match to VL date, the median number of days between CD4 and VL tests was 1 day prior (IQR: 8 days prior to 32 days after).
As an indicator of gaps in care following cART initiation, we derived a dichotomous variable, “year_gap,” that equaled 1 if more than 365 days had elapsed between a CD4 test and the CD4 test immediately prior. Follow-up data on relevant clinical and demographic variables were available through December 31, 2014. All patient data were included until CD4 test data ended, due to death, elective withdrawal from HIVRN cohort, or the end of the observation period.
Statistical analysis
The primary aim of the analysis was to examine the extent of increase in CD4 levels over time, subsequent to initiation of cART, and to assess whether age groups differed in their rates of increase, especially comparing those aged 13–24 to older adults. We distinguish the “baseline” CD4, which is the CD4 test immediately prior to cART initiation, from the outcome variables, which are CD4 tests subsequent to cART initiation. Analyses included all CD4 tests post-cART initiation and within the observation period, regardless of changes in ART regimen or nonadherence. To provide a context for interpreting the CD4 results, and given the critical importance of adherence as measured by virologic suppression, we also examined change over time in viral load.
Longitudinal analyses of CD4 levels used a linear mixed model. Because each patient contributed multiple CD4 measures over time, CD4 outcome measures for a single patient were not independent. Further, because patients were clustered in clinics, observations on patients in the same clinic were not independent. The three-level mixed model (CD4 tests, patients, and clinics) included random effects for clinic and for patient, thereby adjusting for correlated observations at these levels. In addition to age group, the model adjusted for initial treatment regimen, gender, race/ethnicity, HIV acquisition risk, year when cART was initiated, and baseline CD4 level. Viral load and “year_gap” were included as time-varying covariates. Of note, as the focus was CD4 outcome, VL temporally close to each CD4 measurement was used in the analysis rather than baseline VL, as this was thought, a priori, to be the more relevant VL measure associated with CD4, as it is a surrogate for adherence.
To assess CD4 trends, the time origin was the cART initiation date, and the time metric was time since cART initiation; this allows for comparisons of participants at the same time since cART initiation. The statistical model thus included a variable reflecting the number of weeks between the date of cART initiation and the dates of subsequent CD4 tests. To facilitate comparison with prior studies, which examined CD4 outcomes as 24 and 48 weeks [8,26,36], the time variable was scaled in 24-week units; thus, a time value of “2” represents 48 weeks from baseline. To incorporate potential non-linearity in the CD4 time trend, the model included both linear and squared time variables. To incorporate patient-level variation in CD4 time trends, both time variables had random effects, which were correlated. To incorporate age differences in time trends, the model included interactions between age group and each time variable.
We estimated two models. The first excluded baseline CD4, and the second included it. Comparing the models provides a sense of the degree to which effects of time-constant characteristics (e.g., gender, race/ethnicity) are mediated by baseline CD4. The second model is the main model in the analysis. To reflect differential time trends as a function of baseline CD4, the second model also included an interaction between the linear time variable and baseline CD4 category. The second model also included an interaction between age group and baseline CD4 category in order to evaluate whether the CD4 response of youth vs. adults is related to the baseline CD4 level. The baseline CD4 was not itself included in the set of CD4 outcomes, which began with the first CD4 test after cART initiation. A third analysis included a three-way interaction between time, age group, and baseline CD4; this interaction was not significant and these results are not presented. Patients with no CD4 tests recorded after cART initiation, or whose first CD4 test occurred after cART initiation, were excluded from analyses.
To help interpret the results of the second model, we calculated mean predicted outcomes for interactions between (1) time since cART initiation and age group and (2) age group and baseline CD4, and (3) time since cART initiation and baseline CD4.
Finally, to examine change in VL by age group, we conducted a sensitivity analysis that examined the proportion in each group who were suppressed (≤ 400 copies/mL) in each 24-week period. We also estimated a multilevel logistic model including time, age group, and their interaction. Analyses were performed using Stata Version 13 (StataCorp, TX), using a p-value of <0.05.
Results
Initially, 3,110 individuals were eligible for analysis. Of these, we excluded 246 (8%) whose baseline CD4 test occurred more than 16 weeks before the start of cART, 206 (6%) who had no CD4 test after cART, 58 (2%) whose first recorded CD4 test was after cART initiation, and 5 (<1%) missing all VL data.
For the remaining 2,595 patients, the median number of weeks in the observation period (between the cART initiation date and the last CD4 test) was 179 (IQR = 92–249 weeks). The median number of CD4 measures post-cART initiation was 9 (IQR = 4–14). Fifty-eight patients (2%) died during the observation period. The 2,595 patients contributed a total of 25,334 CD4 tests, of which 41 were excluded due to missing timely viral load data and another 17 due to missing CD4 test results, resulting in 25,276 CD4 measures in the analysis.
Of 2,595 patients, 435 (17%) were youth aged 13–24, but only 12 patients were aged < 18. Compared to adults, youth had significantly higher proportions of patients who were male, black, MSM, with baseline CD4 above 200 cells/mm3, and receiving an initial NNRTI regimen. (Table 1) The mean number of weeks in the observation period was similar across age groups: 171, 169, and 177, from the youngest to the oldest, respectively. CD4 analysis.
Baseline CD4
For 2,173 patients (84%), the baseline CD4 occurred prior to the date of cART initiation; and for 422 (16%) the baseline CD4 occurred in the same week at cART initiation. Overall, the median number of weeks between baseline CD4 and ART initiation was 3 weeks prior (IQR 6 to 2 weeks prior). Baseline CD4 medians were 320 (IQR: 232–410) for ages 13–24, 293 (IQR: 158–403) for ages 25–34 and 258 (IQR: 106–384) cells/mm3 for ages 35–44, respectively. Proportions with baseline CD4 ≤ 200 were 0.17, 0.30, and 0.39 in respective age groups.
Longitudinal CD4 trends
Fig 1 shows observed (unadjusted) median CD4 levels by 24-week periods by age group (see S1 Table for estimates). Median CD4 levels in the first 24-week period post-cART initiation were highest among youth and lowest among older adults, consistent with differences in baseline CD4. In each age group, median CD4 tended to rise (with 4 exceptions) from one 24-week period to the next, up to 215 weeks from cART start. Youth attained a higher maximum than older groups. For youth, median CD4 levels declined from their maximum after week 215; decline in CD4 levels was less pronounced for adults.
Table 2 displays results of two mixed-model multivariate regression analyses of longitudinal changes in CD4. Random effects estimates appear in S2 Table. The set of CD4 outcomes for each patient began with the first CD4 test after cART initiation. Without adjusting for baseline CD4 (Model 1), CD4 levels (i.e., the linear slope) after cART start increased an average of 36.73 cells/mm3 per 24-week period for the 13–24 age group. The significant negative coefficient for squared time period (-2.10) indicates that the CD4 trajectory tended to diminish (i.e., become flatter) in later time periods for the youngest group. Interactions between age group and time (linear and quadratic) were not significant, with the exception of a significant positive quadratic term for the oldest age group, implying that the CD4 trajectory for the oldest group showed less of a downturn than that for younger persons.
When baseline CD4 levels were also controlled (Model 2), the linear CD4 slope was 38.98 cells/mm3 per 24-week period for the 13–24 age group, with a quadratic component of -2.09. These values are close to those obtained in Model 1. Similarly, age group did not interact significantly with the linear time variable in Model 2. S3 Table shows mean CD4 levels predicted by Model 2 by age group and specific time point after cART initiation. Overall, CD4 trajectories over time were similar for youth and adults after adjusting for baseline CD4 level.
In addition, CD4 trajectories over time differed significantly as a function of baseline CD4. Compared to those with baseline CD4 ≤200, the linear slope was significantly lower by 2.70 units for patients with baseline CD4 201–500, and was 7.62 units lower for patients with baseline CD4>500. Thus, the overall rate of CD4 improvement declined as the baseline CD4 increased. S4 Table shows predicted CD4 levels by baseline CD4 category and specific time point.
Further, when patients had a gap in care, reflected in the time between successive CD4 tests of 365 days or longer, CD4 levels declined by 26 cells/mm3 in Model 2. In addition, compared to VL≤400, VL values between 401–1000 copies/mL were associated with a decrease of 35, and VL values above 1000 were associated with a decrease of 94 CD4 cells/mm3.
Differences at time of cART initiation
In the regression models, coefficients for variables that are not interacted with time represent predicted differences immediately after cART initiation (i.e., time = 0). Without adjusting for baseline CD4 category (Model 1), patients aged 25–34 and 35–44 had significantly lower CD4 values than youth just after starting cART. In Model 2, due to including an interaction between age and baseline CD4, the main effect for age group pertains to age differences among those with baseline CD4 ≤200; in this group, age differences were not significant. The coefficients of the baseline CD4 by age interaction show that age differences were not significant among those with baseline CD4 201–500, but were significant among those with baseline CD4 >500. Each adult age group had significantly higher CD4 levels right after cART initiation than youth when baseline CD4 was > 500. To help visualize this pattern, S5 Table shows median CD4 levels by age group and baseline CD4 just after cART initiation predicted from Model 2 and also at 24 weeks.
In Model 1, there were significant gender, racial/ethnic, cART regimen, and HIV risk factor differences in CD4 levels at cART initiation. However, when baseline CD4 levels were controlled (Model 2) the magnitude of these differences diminished, with gender, race, and risk factor becoming nonsignificant. In both models, however, CD4 after cART initiation was higher among those who initiated cART after 2008.
Trends in virologic suppression
Table 3 shows the proportion of patients in each age group who were virologically suppressed (≤400 copies/mL) in each 24-week period following cART initiation. The proportion increased between the first and second time periods for each age group, but, unlike CD4 levels, the proportion suppressed did not show a consistent trend subsequently and instead fluctuated within a few percentage points in each age group. The youngest age group began with a greater proportion suppressed than the older age groups, but by the ninth and tenth periods the older age groups had somewhat higher proportions with suppressed viral load.
To examine age trends further, we used a mixed logistic regression model. (Table 4) The model was similar to that used for CD4, but baseline CD4 variables were removed. In each 24-week period, the odds of being virologically suppressed increased by 30%. The significant squared term indicates that the rate of increase diminished over time. Age groups did not differ significantly in the proportion suppressed at the initiation of cART, but the rate of increase was significantly greater for the oldest group, compared to youth.
Discussion
Analyses of a large cohort of young PLHIV receiving cART reveals three major findings: youth who start cART do so at higher CD4 counts compared to older adults; youth have better initial CD4 responses than older adults; and, possibly due to decreased adherence, youth’s CD4 response plateaus, allowing adults to reach similar CD4 levels after sufficient time has passed.
Specifically, the median CD4 levels of young PLHIV tended to rise consistently up to 215 weeks after cART initiation, however, the (adjusted) linear rate of increase did not differ significantly between youth (aged 13–24) and adults (aged 25–44). Those with higher baseline CD4 had higher subsequent CD4 at any point in time [37–39]. Higher viral load and gaps between CD4 tests of a year or longer were each associated with lower CD4 levels.
Studies that have compared the immune recovery of youth to that of adults in non-HIV settings have suggested that youth have a greater capacity to regenerate their immune system, likely as a result of their residual thymic capacity. [19,22,23] The relative CD4 recovery of nPHIV youth living with HIV is unclear; studies that have attempted to evaluate the question have tended to lump youth into older age groups or mixed nPHIV youth with perinatally HIV-infected youth, impairing assessment of the specific responses of nPHIV youth to cART. [21,24] [40] [41,42]
It is well established that the lower the CD4 at initiation of therapy, the lower CD4 threshold that one will be able to reach, as evidenced by the longitudinal data in this study. [38,43] Individuals with the lowest baseline CD4 values (≤200 cells/mm3), compared to those in the higher CD4 categories, had more rapid rises in CD4 counts (i.e., higher linear slope), but ultimately had lower CD4 levels at the end of follow-up. This finding likely relates to the reduced regenerative capacity in lower CD4 categories and the better preserved capacity to repopulate CD4 in higher CD4 categories, further underscoring the importance of early cART initiation. [12,44] Although youth may have an initial short-term advantage in reaching higher CD4 levels after cART initiation, the long-term difference (compared to older adults) is minimal, suggesting that treatment initiation for youth should not be delayed until CD4 counts decline. Additionally, the consequences of unchecked virologic replication, including the risk of secondary transmission and systemic sequelae of immune activation, likely outweigh any advantage of delaying therapy in this population. [45–47]
The virologic suppression rates in this study were higher than those previously reported in clinical studies of youth initiating cART. [7,8] Prior studies have shown lower virologic suppression rates for youth than adults. [8] These studies include a mix of observational and clinical trial data, and many examine older regimens (e.g., higher dose PI regimens) or less tolerable agents (e.g., efavirenz). [5,8] For the more contemporary regimens initiated during the study period between 2006 and 2012, we show that youth had similar virologic suppression rates as their adult counterparts. Given that regimens are becoming increasingly more streamlined, with combination tablets, tolerable agents, with higher barriers to resistance, this finding is promising and may improve confidence among providers to initiate cART earlier for youth.[13]
Ultimately, youth did not attain higher CD4 values than their adult counterparts. It is possible that the lack of continued improvement was related to increased non-compliance over time. In PACTG 381, Flynn et al reported decreased adherence, as measured by virologic suppression, from 59% to 49% to 24% over 24, 60, and 152 weeks on cART. [8,9] In contrast, the virologic suppression rates of youth remained similar to adults over the course of this study.
Some additional demographic characteristics were associated with higher CD4 after ART initiation in Model 1, specifically, female gender, white race, and initiating cART in the latter years of study (e.g., 2011 vs. 2008). However, these differences (except for year of cART initiation) were reduced by adjusting for baseline CD4, suggesting that baseline CD4 differences underlie some observed demographic associations.
Individuals with gaps in care had significantly lower CD4 levels compared to those who were consistently in care. In fact, each additional one year gap in care resulted in a 40 cells/mm3 reduction in CD4 levels, suggesting the importance of continued engagement in care and maintenance of cART, once initiated. This finding is consistent with other studies. [38,48,49] Short term gains while on therapy are not maintained when therapy is (probably) interrupted. [50,51] Additionally, the risk of virologic nonsuppression and the potential for rebound in immune activation and their potential sequelae provide additional evidence for measures supporting patient continuity of care. [45,46]
Our results should be interpreted with consideration of several important limitations. First, the number of youth under 18 was negligible, and our findings for youth may not generalize to those aged 13–17. Second, although our HIVRN demographics closely resemble those of PLHIV in the U.S. in general, our sample is not nationally representative. Third, mental health, substance abuse, socioeconomic status, and other psychosocial characteristics, which may be important factors that impact virologic suppression and by extension CD4 response, are not measured in the HIVRN and therefore could not be examined. Fourth, this analysis focuses on the response to cART initiation and does not examine regimen changes. Fifth, differences between CD4 outcomes of youth and those of adults ≥45 years old are likely pronounced, given the impact of ageing on the immune system. This was not the focus of this study, and we cannot comment on possible differences between youth and this older age group.
Finally, there was substantial attrition from the study over time, as is common in longitudinal investigations. Such attrition could potentially bias results if those who drop out are systematically different from those who remain (e.g., have lower or higher rates of CD4 improvement). The mixed-model analysis assumes that data are missing at random (and not the more restrictive missing completely at random assumption), and the regression model included several demographic factors that could be related to propensity to drop out, which may mitigate potential attrition biases.
In conclusion, youth followed in non-clinical-trials settings of HIVRN clinics achieved CD4 outcomes similar to adults. Despite youth’s residual thymic tissue and capacity of immune reconstitution, initiation of cART should not be delayed for youth with the expectation that residual capacity for immune reconstitution will result in greater CD4 responses that will be sustained. Initiating cART soon after diagnosis and maintaining virologic suppression are the most critical factors to enhancing CD4 outcomes, especially for youth who traditionally have higher nonadherence rates.[8,17,52–54]. Our data suggest that, even with their residual thymic capacity, youth and adults have similar immune recovery after initiating cART, adjusting for baseline CD4, and that emphasis should be placed on initiating cART as close to diagnosis as possible when CD4 counts are higher. Studies of mechanisms to support youth’s continued adherence to therapy and their impact on resultant virologic suppression are critical.
Supporting information
S1 Table. Unadjusted Median CD4 Levels, by Age Group and Weeks from cART Initiation.
Cell entries are median CD4, number of observations, and number of unique patients. Number of observations exceeds number of patients in each time interval because one person can contribute multiple observations. Observations in different time periods are not independent; the same person can have data in multiple time periods.
https://doi.org/10.1371/journal.pone.0171125.s001
(DOCX)
S2 Table. Random Effects Parameters from Second Regression Model.
Note: SD- standard deviation; corr- correlation.
https://doi.org/10.1371/journal.pone.0171125.s002
(DOCX)
S3 Table. Predicted (Adjusted) Mean CD4 Levels, by Age Group and 24-Week Periods from Baseline (Model 2).
Note: Entries are mean CD4 levels predicted by regression model 2, averaging over other covariates.
https://doi.org/10.1371/journal.pone.0171125.s003
(DOCX)
S4 Table. Predicted (Adjusted) Mean CD4 Levels, by Baseline CD4 and 24-Week Periods from Baseline (Model 2).
Note: Entries are mean CD4 levels predicted by regression model 2, averaging over other covariates.
https://doi.org/10.1371/journal.pone.0171125.s004
(DOCX)
S5 Table. Predicted mean CD4 levels by age group and baseline CD4 at Time of cART initiation and 24 Weeks after cART Initiation.
https://doi.org/10.1371/journal.pone.0171125.s005
(DOCX)
Acknowledgments
HIVRN Participating Sites and Principal Investigators: Alameda County Medical Center, Oakland, California (Howard Edelstein, M.D.) Children's Hospital of Philadelphia, Philadelphia, Pennsylvania (Richard Rutstein, M.D.) Community Health Network, Rochester, New York (Roberto Corales, D.O.) Drexel University, Philadelphia, Pennsylvania (Jeffrey Jacobson, M.D., Sara Allen, C.R.N.P.) Johns Hopkins University, Baltimore, Maryland (Kelly Gebo, M.D., Richard Moore, M.D., Allison Agwu M.D.) Montefiore Medical Group, Bronx, New York (Robert Beil, M.D.) Montefiore Medical Center, Bronx, New York (Lawrence Hanau, M.D.) Oregon Health and Science University, Portland, Oregon (P. Todd Korthuis, M.D.) Parkland Health and Hospital System, Dallas, Texas (Ank Nijhawan, M.D., Muhammad Akbar, M.D.) St. Jude's Children's Hospital and University of Tennessee, Memphis, Tennessee (Aditya Gaur, M.D.) St. Luke's Roosevelt Hospital Center, New York, New York (Victoria Sharp, M.D., Stephen Arpadi, M.D.) Tampa General Health Care, Tampa, Florida (Charurut Somboonwit, M.D.) University of California, San Diego, California (W. Christopher Mathews, M.D.) Wayne State University, Detroit, Michigan (Jonathan Cohn, M.D.). Sponsoring Agencies: Agency for Healthcare Research and Quality, Rockville, Maryland (Fred Hellinger, Ph.D., John Fleishman, Ph.D., Irene Fraser, Ph.D.) Health Resources and Services Administration, Rockville, Maryland (Robert Mills, Ph.D., Faye Malitz, M.S.). Data Coordinating Center: Johns Hopkins University (Richard Moore, M.D., Jeanne Keruly, C.R.N.P., Kelly Gebo, M.D., Cindy Voss, M.A.)
Author Contributions
- Conceptualization: ALA KAG KA BY.
- Data curation: KAG.
- Formal analysis: GM BAN GM.
- Funding acquisition: KAG RM SAB.
- Investigation: KAG AA CM AN RR.
- Methodology: ALA JF GM BAN KA KAG.
- Project administration: AA KAG.
- Resources: GM BAN JF.
- Software: GM BAN JF.
- Supervision: KAG RM JK.
- Validation: BAN JF GM.
- Visualization: AA JF.
- Writing – original draft: AA JF GM BAN KAG.
- Writing – review & editing: AN RR AA RM KAG CM JK JF BAN GM BY JA.
References
- 1. Moore RD, Keruly JC, Bartlett JG (2012) Improvement in the health of HIV-infected persons in care: reducing disparities. Clin Infect Dis 55: 1242–1251. pmid:23019271
- 2. Samji H, Cescon A, Hogg RS, Modur SP, Althoff KN, et al. (2013) Closing the gap: increases in life expectancy among treated HIV-positive individuals in the United States and Canada. PLoS One 8: e81355. pmid:24367482
- 3.
Centers for Disease CNCfHA, Viral Hepatitis, STD, and TB Prevention (2012) More Than Half of Young HIV-Infected Americans Are Not Aware of Their Status. Vital Signs.
- 4. Agwu AL, Neptune A, Voss C, Yehia B, Rutstein R, et al. (2014) CD4 counts of nonperinatally HIV-infected youth and young adults presenting for HIV care between 2002 and 2010. JAMA Pediatr 168: 381–383. pmid:24493261
- 5. Gagliardo C, Brozovich A, Birnbaum J, Radix A, Foca M, et al. (2014) A multicenter study of initiation of antiretroviral therapy and transmitted drug resistance in antiretroviral-naive adolescents and young adults with HIV in New York City. Clin Infect Dis 58: 865–872. pmid:24429431
- 6. Farmer C YB, Fleishman JA, Rutstein R, Matthews CW, Nijhawan A, Moore RD, Gebo KA, and, Agwu AL for the HIV Research Network. (2014) Factors Associated with Retention among Non-Perinatally HIV-Infected Youth in the HIV Research Network. Journal of the Pediatric Infectious Diseases Society. 5(1): 39–46. pmid:26908490
- 7. Ryscavage P, Anderson EJ, Sutton SH, Reddy S, Taiwo B (2011) Clinical outcomes of adolescents and young adults in adult HIV care. J Acquir Immune Defic Syndr 58: 193–197. pmid:21826014
- 8. Flynn PM, Rudy BJ, Douglas SD, Lathey J, Spector SA, et al. (2004) Virologic and immunologic outcomes after 24 weeks in HIV type 1-infected adolescents receiving highly active antiretroviral therapy. JInfectDis 190: 271–279.
- 9. Flynn PM, Rudy BJ, Lindsey JC, Douglas SD, Lathey J, et al. (2007) Long-term observation of adolescents initiating HAART therapy: three-year follow-up. AIDS ResHumRetroviruses 23: 1208–1214.
- 10. Fleishman JA, Yehia BR, Moore RD, Korthuis PT, Gebo KA (2012) Establishment, retention, and loss to follow-up in outpatient HIV care. J Acquir Immune Defic Syndr 60: 249–259. pmid:22531758
- 11. El-Sadr WM, Grund B, Neuhaus J, Babiker A, Cohen CJ, et al. (2008) Risk for opportunistic disease and death after reinitiating continuous antiretroviral therapy in patients with HIV previously receiving episodic therapy: a randomized trial. AnnInternMed 149: 289–299.
- 12. Group ISS, Lundgren JD, Babiker AG, Gordin F, Emery S, et al. (2015) Initiation of Antiretroviral Therapy in Early Asymptomatic HIV Infection. N Engl J Med 373: 795–807. pmid:26192873
- 13.
(2014) Panel in Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in the HIV-1-infected adults and adolescents. In: Services. DoHaH, editor.
- 14.
World Health O (2015) Guideline on when to start antiretroviral therapy and on pre-exposure prophylaxis for HIV.
- 15. Agwu AL, Fleishman JA, Korthuis PT, Siberry GK, Ellen JM, et al. (2011) Disparities in antiretroviral treatment: a comparison of behaviorally HIV-infected youth and adults in the HIV Research Network. J Acquir Immune Defic Syndr 58: 100–107. pmid:21637114
- 16. Blum RW, McNeely C, Nonnemaker J (2002) Vulnerability, risk, and protection. JAdolescHealth 31: 28–39.
- 17. Murphy DA, Sarr M, Durako SJ, Moscicki AB, Wilson CM, et al. (2003) Barriers to HAART adherence among human immunodeficiency virus-infected adolescents. ArchPediatrAdolescMed 157: 249–255.
- 18. Naar-King S, Kolmodin K, Parsons JT, Murphy D (2010) Psychosocial factors and substance use in high-risk youth living with HIV: a multi-site study. AIDS Care 22: 475–482. pmid:20146112
- 19. Kalayjian RC, Landay A, Pollard RB, Taub DD, Gross BH, et al. (2003) Age-related immune dysfunction in health and in human immunodeficiency virus (HIV) disease: association of age and HIV infection with naive CD8+ cell depletion, reduced expression of CD28 on CD8+ cells, and reduced thymic volumes. JInfectDis 187: 1924–1933.
- 20. Collaboration of Observational HIVERESG, Sabin CA, Smith CJ, d'Arminio Monforte A, Battegay M, et al. (2008) Response to combination antiretroviral therapy: variation by age. AIDS 22: 1463–1473. pmid:18614870
- 21. Althoff KN, Justice AC, Gange SJ, Deeks SG, Saag MS, et al. (2010) Virologic and immunologic response to HAART, by age and regimen class 3. AIDS 24: 2469–2479. pmid:20829678
- 22. Mackall CL, Fleisher TA, Brown MR, Andrich MP, Chen CC, et al. (1995) Age, thymopoiesis, and CD4+ T-lymphocyte regeneration after intensive chemotherapy. N Engl J Med 332: 143–149. pmid:7800006
- 23. Hakim FT, Cepeda R, Kaimei S, Mackall CL, McAtee N, et al. (1997) Constraints on CD4 recovery postchemotherapy in adults: thymic insufficiency and apoptotic decline of expanded peripheral CD4 cells. Blood 90: 3789–3798. pmid:9345067
- 24. Althoff KN, Gebo KA, Gange SJ, Klein MB, Brooks JT, et al. (2010) CD4 count at presentation for HIV care in the United States and Canada: are those over 50 years more likely to have a delayed presentation? 2. AIDS ResTher 7: 45.
- 25. (2008) Response to combination antiretroviral therapy: variation by age. AIDS 22: 1463–1473. pmid:18614870
- 26. Rudy BJ, Kapogiannis BG, Worrell C, Squires K, Bethel J, et al. (2015) Immune Reconstitution but Persistent Activation After 48 Weeks of Antiretroviral Therapy in Youth With Pre-Therapy CD4 >350 in ATN 061. J Acquir Immune Defic Syndr 69: 52–60. pmid:25942459
- 27. Rudy BJ, Lindsey JC, Flynn PM, Bosch RJ, Wilson CM, et al. (2006) Immune reconstitution and predictors of virologic failure in adolescents infected through risk behaviors and initiating HAART: week 60 results from the PACTG 381 cohort. AIDS ResHumRetroviruses 22: 213–221.
- 28. Gebo KA (2008) Epidemiology of HIV and response to antiretroviral therapy in the middle aged and elderly. Aging health 4: 615–627. pmid:19915688
- 29. Silverberg MJ, Leyden W, Horberg MA, DeLorenze GN, Klein D, et al. (2007) Older age and the response to and tolerability of antiretroviral therapy. ArchInternMed 167: 684–691.
- 30. Mpondo BC; Gunda DKSM, E (2016) Immunological and Clinical Responses following the Use of Antiretroviral Therapy among Elderly HIV-Infected Individuals Attending Care and Treatment Clinic in Northwestern Tanzania: A Retrospective Cohort Study. Journal of Sexually Transmitted Diseases 2016.
- 31. Althoff KN, Justice AC, Gange SJ, Deeks SG, Saag MS, et al. (2010) Virologic and immunologic response to HAART, by age and regimen class. AIDS 24: 2469–2479. pmid:20829678
- 32. Viard JP, Mocroft A, Chiesi A, Kirk O, Roge B, et al. (2001) 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 183: 1290–1294. pmid:11262215
- 33. Tumbarello M, Rabagliati R, de Gaetano Donati K, Bertagnolio S, Montuori E, et al. (2004) Older age does not influence CD4 cell recovery in HIV-1 infected patients receiving highly active antiretroviral therapy. BMC Infect Dis 4: 46. pmid:15530169
- 34.
HIV Research Network (HIVRN). pp. Goals of the HIV Research Network. Available at https://cds.johnshopkins.edu/hivrn/. Accessed December 21, 2016.
- 35. Gebo KA, Moore RD, Fleishman JA (2003) The HIV Research Network: a unique opportunity for real time clinical utilization analysis in HIV. HopkinsHIVRep 15: 5–6.
- 36. Mocroft A, Phillips AN, Gatell J, Ledergerber B, Fisher M, et al. (2007) Normalisation of CD4 counts in patients with HIV-1 infection and maximum virological suppression who are taking combination antiretroviral therapy: an observational cohort study. Lancet 370: 407–413. pmid:17659333
- 37. Althoff KN, Gange SJ, Klein MB, Brooks JT, Hogg RS, et al. (2010) Late presentation for human immunodeficiency virus care in the United States and Canada 4. ClinInfectDis 50: 1512–1520.
- 38. Moore RD, Keruly JC (2007) CD4+ cell count 6 years after commencement of highly active antiretroviral therapy in persons with sustained virologic suppression 5. ClinInfectDis 44: 441–446.
- 39. Asfaw A, Ali D, Eticha T, Alemayehu A, Alemayehu M, et al. (2015) CD4 cell count trends after commencement of antiretroviral therapy among HIV-infected patients in Tigray, Northern Ethiopia: a retrospective cross-sectional study. PLoS One 10: e0122583. pmid:25816222
- 40. Nglazi MD, Kranzer K, Holele P, Kaplan R, Mark D, et al. (2012) Treatment outcomes in HIV-infected adolescents attending a community-based antiretroviral therapy clinic in South Africa. BMCInfectDis 12: 21.
- 41. Gutierrez F, Padilla S, Masia M, Iribarren JA, Moreno S, et al. (2006) Clinical outcome of HIV-infected patients with sustained virologic response to antiretroviral therapy: long-term follow-up of a multicenter cohort. PLoS One 1: e89. pmid:17183720
- 42. Luz PM, Grinsztejn B, Velasque L, Pacheco AG, Veloso VG, et al. (2014) Long-term CD4+ cell count in response to combination antiretroviral therapy. PLoS One 9: e93039. pmid:24695533
- 43. Lewis J, Walker AS, Castro H, De Rossi A, Gibb DM, et al. (2012) Age and CD4 count at initiation of antiretroviral therapy in HIV-infected children: effects on long-term T-cell reconstitution. J Infect Dis 205: 548–556. pmid:22205102
- 44. Chaisson RE, Keruly JC, Moore RD (2000) Association of initial CD4 cell count and viral load with response to highly active antiretroviral therapy 15. JAMA 284: 3128–3129. pmid:11135775
- 45. Kitahata MM, Gange SJ, Abraham AG, Merriman B, Saag MS, et al. (2009) Effect of early versus deferred antiretroviral therapy for HIV on survival. NEnglJMed 360: 1815–1826.
- 46. Deeks SG (2011) HIV infection, inflammation, immunosenescence, and aging. Annu Rev Med 62: 141–155. pmid:21090961
- 47. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, et al. (2011) Prevention of HIV-1 infection with early antiretroviral therapy 1. NEnglJMed 365: 493–505.
- 48. Lifson AR, Krantz EM, Eberly LE, Dolan MJ, Marconi VC, et al. (2011) Long-term CD4+ lymphocyte response following HAART initiation in a U.S. Military prospective cohort. AIDS Res Ther 8: 2. pmid:21244701
- 49. Hunt PW, Deeks SG, Rodriguez B, Valdez H, Shade SB, et al. (2003) Continued CD4 cell count increases in HIV-infected adults experiencing 4 years of viral suppression on antiretroviral therapy. AIDS 17: 1907–1915. pmid:12960823
- 50. Touloumi G, Pantazis N, Antoniou A, Stirnadel HA, Walker SA, et al. (2006) Highly active antiretroviral therapy interruption: predictors and virological and immunologic consequences. J Acquir Immune Defic Syndr 42: 554–561. pmid:16868497
- 51. Siberry GK, Patel K, Van Dyke RB, Hazra R, Burchett SK, et al. (2011) CD4+ lymphocyte-based immunologic outcomes of perinatally HIV-infected children during antiretroviral therapy interruption. JAcquirImmuneDeficSyndr 57: 223–229.
- 52. Rudy BJ, Murphy DA, Harris DR, Muenz L, Ellen J (2009) Patient-related risks for nonadherence to antiretroviral therapy among HIV-infected youth in the United States: a study of prevalence and interactions. AIDS PatientCare STDS 23: 185–194.
- 53. Agwu AL, Fairlie L (2013) Antiretroviral treatment, management challenges and outcomes in perinatally HIV-infected adolescents. J Int AIDS Soc 16: 18579. pmid:23782477
- 54. Murphy DA, Belzer M, Durako SJ, Sarr M, Wilson CM, et al. (2005) Longitudinal antiretroviral adherence among adolescents infected with human immunodeficiency virus. ArchPediatrAdolescMed 159: 764–770.