CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 RESEARCH ARTICLE Growth and CD4 patterns of adolescents living with perinatally acquired HIV worldwide, a CIPHER cohort collaboration analysis The Collaborative Initiative for Paediatric HIV Education and Research (CIPHER) Global Cohort Collaboration, Julie Jesson1,§ , Siobhan Crichton2 , Matteo Quartagno2 , Marcel Yotebieng3 , Elaine J. Abrams4 , Kulkanya Chokephaibulkit5 , Sophie Le Coeur6,7 , Marie-Hélène Aké-Assi8 , Kunjal Patel9 , Jorge Pinto10 , Mary Paul11 , Rachel Vreeman12 , Mary-Ann Davies13 , Jihane Ben-Farhat14 , Russell Van Dyke15 , Ali Judd2 , Lynne Mofenson16 , Marissa Vicari17 , George Seage III8 , Linda-Gail Bekker18 , Shaffiq Essajee19 , Diana Gibb2 , Martina Penazzato20 , Intira Jeannie Collins2 , Kara Wools-Kaloustian21 , Amy Slogrove22 , Kate Powis8 , Paige Williams8 , Mogomotsi Matshaba23 , Lineo Thahane24 , Phoebe Nyasulu25 , Bhekumusa Lukhele26 , Lumumba Mwita27 , Adeodata Kekitiinwa-Rukyalekere28 , Sebastian Wanless10 , Tessa Goetghebuer29 , Claire Thorne30 , Josiane Warszawski31 , Luisa Galli32,33 , Annemarie M.C van Rossum34 , Carlo Giaquinto35 , Magdalena Marczynska36 , Laura Marques37 , Filipa Prata38 , Luminita Ene39 , Lyuba Okhonskaya40 , Marisa Navarro41 , Antoinette Frick42 , Lars Naver43 , Christian Kahlert44 , Alla Volokha45 , Elizabeth Chappell2 , Jean William Pape46 , Vanessa Rouzier46 , Adias Marcelin46 , Regina Succi47 , Annette H. Sohn48 , Azar Kariminia49 , Andrew Edmonds50 , Patricia Lelo51 , Rita Lyamuya52 , Edith Apondi Ogalo53 , Francesca Akoth Odhiambo54 , Andreas D. Haas55 , Carolyn Bolton56 , Josephine Muhairwe57 , Hannock Tweya58 , Mariam Sylla59 , Marceline D’Almeida60 , Lorna Renner61 , Mark J. Abzug62 , James Oleske63 , Murli Purswani64 , Chloe Teasdale4 , Harriet Nuwagaba-Biribonwoha4 , Ruth Goodall2,* and Valériane Leroy1,* § Corresponding author: Julie Jesson, CERPOP, Inserm, Université de Toulouse, Université Paul Sabatier Toulouse 3, Faculté de Médecine Purpan, 37 Allées Jules Guesde, 31073 Toulouse Cedex 7, France. Tel: +33 5 61 14 59 57. (julie_jesson@sfu.ca) * Co-last authors. Abstract Introduction: Adolescents living with HIV are subject to multiple co-morbidities, including growth retardation and immunode- ficiency. We describe growth and CD4 evolution during adolescence using data from the Collaborative Initiative for Paediatric HIV Education and Research (CIPHER) global project. Methods: Data were collected between 1994 and 2015 from 11 CIPHER networks worldwide. Adolescents with perinatally acquired HIV infection (APH) who initiated antiretroviral therapy (ART) before age 10 years, with at least one height or CD4 count measurement while aged 10–17 years, were included. Growth was measured using height-for-age Z-scores (HAZ, stunt- ing if <-2 SD, WHO growth charts). Linear mixed-effects models were used to study the evolution of each outcome between ages 10 and 17. For growth, sex-specific models with fractional polynomials were used to model non-linear relationships for age at ART initiation, HAZ at age 10 and time, defined as current age from 10 to 17 years of age. Results: A total of 20,939 and 19,557 APH were included for the growth and CD4 analyses, respectively. Half were females, two-thirds lived in East and Southern Africa, and median age at ART initiation ranged from <3 years in North America and Europe to >7 years in sub-Saharan African regions. At age 10, stunting ranged from 6% in North America and Europe to 39% in the Asia-Pacific; 19% overall had CD4 counts <500 cells/mm3 . Across adolescence, higher HAZ was observed in females and among those in high-income countries. APH with stunting at age 10 and those with late ART initiation (after age 5) had the largest HAZ gains during adolescence, but these gains were insufficient to catch-up with non-stunted, early ART-treated adolescents. From age 10 to 16 years, mean CD4 counts declined from 768 to 607 cells/mm3 . This decline was observed across all regions, in males and females. Conclusions: Growth patterns during adolescence differed substantially by sex and region, while CD4 patterns were similar, with an observed CD4 decline that needs further investigation. Early diagnosis and timely initiation of treatment in early childhood to prevent growth retardation and immunodeficiency are critical to improving APH growth and CD4 outcomes by the time they reach adulthood. Keywords: adolescent; CD4; cohort studies; growth; HIV; perinatally acquired 1 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Additional information may be found under the Supporting Information tab of this article. Received 5 July 2021; Accepted 20 December 2021 Copyright © 2022 The Authors. Journal of the International AIDS Society published by John Wiley & Sons Ltd on behalf of the International AIDS Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 1 INTRODUCTION diatric Infections Cohort Collaboration (EPPICC); the Inter- national Epidemiology Databases to Evaluate AIDS (IeDEA) In 2019, an estimated 1.7 million adolescents aged 10–19 Consortium (IeDEA Asia-Pacific, IeDEA Central Africa, IeDEA years were living with HIV worldwide, with 90% of them in East Africa, IeDEA Southern Africa, IeDEA West Africa, the sub-Saharan Africa, and 8% in Asia and the Pacific [1]. They Caribbean, Central and South America network (CCASAnet)); either acquired HIV during adolescence or are adolescents International Maternal Pediatric Adolescent AIDS Clinical Tri- living with perinatally acquired HIV (APH). APH experience als (IMPAACT) 219C and P1074; Optimal Models (ICAP at specific challenges due to their life-long infection and long- Columbia University); and Pediatric HIV/AIDS Cohort Study term follow-up in HIV care [2–6]. They are at risk of inad- (PHACS). Pooled individual participant-level data were thus equate adherence to antiretroviral therapy (ART) [7,8] and collected from high-, middle- and low-income countries and poor retention in care during adolescence [9,10]. APH may in drawn from a range of care settings, including dedicated turn experience poor health outcomes [11], with a high mor- research cohorts, routine care cohorts and programmatic ser- tality rate [12,13], virologic failure [14,15] and HIV drug resis- vices. tance [16], poor growth [17] and immune status [18]. The inclusion criteria were: adolescents aged 10–19 years, Previous work has demonstrated that linear growth documented perinatally acquired HIV or unknown mode of improves after ART initiation, with better outcomes associ- infection and aged<10 years at first presentation to HIV care ated with earlier ART [19,20], but long-term growth evolution (proxy perinatal HIV transmission), age <10 years at ART initi- on ART has been less described [21,22], especially in APH ation and ≥1 height or CD4 count measurement between age [17,23]. Some studies have described APH growth velocity 10 and 19 years to contribute to the growth and CD4 anal- in resource-limited settings [17] and in Europe and Thailand yses, respectively. The analysis was subsequently truncated to [23], showing lower height gains and a delayed growth spurt include only measurements from age 10 to 17 years due to for APH compared to HIV-uninfected adolescents. small sample size at 18 and 19 years of age. Similarly, the immune response following ART initiation has Each participating cohort in CIPHER obtained ethics been extensively described in children [24,25] and adults approval from their respective institutional review boards for [26,27] living with HIV, but long-term CD4 evolution on ART data collection and transfer. Consent or assent requirements for APH has not been sufficiently explored. In previous stud- for study participation were deferred to local institutional ies looking at patterns of long-term trends in CD4 count in review boards. All analyses were pre-specified and agreed by children living with HIV in sub-Saharan Africa [28] and Europe the CIPHER Adolescent Study Group. [29], CD4 counts at adolescence were hypothesized to remain constant after the initial CD4 response following ART initi- 2.2 Outcomes and exploratory variables ation, likely reaching a similar level to that of the non-HIV- infected population of the same age [28]. Two main outcomes of interest were investigated: (1) growth, Our primary objective was to describe linear growth and defined by height-for-age Z-score (HAZ), based on the immune evolution across adolescence by sex and geographical WHO Child Growth standards [32,33] and (2) CD4 counts region among ART-treated APH enrolled in the Collaborative (cells/mm3 ). CD4 counts were also expressed as an age- Initiative for Paediatric HIV Education and Research (CIPHER) adjusted CD4 ratio, obtained by dividing observed CD4 global project. A secondary objective was to explore the cor- counts by CD4 reference values in HIV-uninfected adoles- relation between growth and CD4 evolution. cents of the same age [34]. The prevalence of malnutrition was assessed by stunting (i.e. HAZ <-2 Standard Deviations [SD]) and wasting (i.e. weight-for-height Z-score [WHZ] <-2 2 METHODS SD if <5 years old, BMI-for-age Z-score [BAZ] <-2 SD if older), severe malnutrition was defined as Z-scores <-3 SD. 2.1 Study population Immunodeficiency at ART initiation was determined according This multiregional study was based on pooled data from the to the 2006 WHO guidelines [35], immunodeficiency at age CIPHER collaboration. This global network of observational 10 was defined as severe (CD4 counts <250/mm3 ), moder- paediatric HIV cohorts previously described the epidemiology ate (250–500/mm3 ) and not immunodeficient (>500/mm3 ). of adolescents living with HIV worldwide [30], with a focus Regions were combined as follows: North America and in sub-Saharan Africa [31]. For this study, the following 11 Europe (NA&E); Central, South America and the Caribbean cohort networks contributed data: Baylor International Pedi- (CSA&C); Asia-Pacific; West and Central Africa (West&CA); atric AIDS Initiative (BIPAI); European Pregnancy and Pae- Botswana and South Africa (Botsw.&SA); East and Southern 2 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Africa (East&SA) (excluding Botswana and South Africa). Botswana and South Africa, both upper middle-income coun- tries, were classified as a separate group within sub-Saharan Africa. Adolescents were defined as alive and retained in care, transferred out, dead or lost-to-follow-up if no visit were observed for >365 days after their last observed visit. 2.3 Statistical analyses For both study populations contributing to the growth and CD4 analyses, characteristics were described at ART initiation and at age 10, and compared by sex and geographical region, using Chi-square tests for categorical variables and Kruskal– Wallis tests for continuous variables. We compared charac- teristics of the study population and adolescents who were excluded, due to lack of data on height or CD4 measurement and because they did not initiate ART by age 10 years. Growth and CD4 evolution were described between age 10 Figure 1. Selection of the adolescents living with HIV for the and 17 years by plotting mean HAZ and CD4 count over time, growth and CD4 analyses. CIPHER global cohort collaboration, by sex and regions. Growth and CD4 modelling included the 1994–2015. following parameters: age at ART initiation, immunodeficiency and stunting at ART initiation and at age 10, wasting at age 10 (growth modelling only) and year of birth (CD4 modelling measurement for height and CD4 during adolescence. After only), regions and sex. For the growth model, non-linearity of restricting to those initiated on ART before age 10 years, some parameters and the heterogeneity of results by sex and 20,939 (76%) APH were included in the growth analyses and regions had to be taken into account. Growth analyses were 19,557 (75%) in the CD4 analyses (Figure 1). thus stratified by sex and included fractional polynomials [36] Characteristics of both study populations are shown in for the following continuous variables: age at ART initiation, Table 1 by sex and in Tables S1 and S2 by regions. If not HAZ at age 10 years and time, defined as current age from specified, results detailed in the text refer to the growth 10 to 17 years of age. The fractional polynomial terms were analysis study population, the CD4 analysis study population inserted into a linear mixed model, with an exponential spa- was broadly similar. Overall, half were females, and two-thirds tial correlation structure, and interaction between time and lived in East&SA. Median age at ART initiation varied widely the explanatory variables. For the CD4 evolution modelling, a by region, ranging from 2.4 years (IQR 0.6;5.7) in NA&E to linear mixed model with an unstructured variance-covariance 7 years and more in sub-Saharan African regions. At ART ini- matrix and interaction with time was fitted. Variables were tiation, the prevalence of stunting ranged from 8% in NA&E selected in univariate models if p<0.25 and kept in final mul- to 40% in Asia-Pacific, severe immunodeficiency ranged from tivariate models if p<0.05 using a backward selection method. 20% in NA&E to 50% in the Asia-Pacific. Half of APH had A secondary objective was to explore the possible corre- missing data regarding anthropometric parameters or CD4 lation between HAZ and CD4 evolution among APH, imple- measurements at ART initiation. At age 10, the prevalence of menting multivariate multilevel models [37] and a multi- stunting ranged from 6% in NA&E to 39% in the Asia-Pacific. trajectory analysis [38]. The first strategy consisted of fit- In the CD4 analysis study population, 19% were moderately ting a bivariate normal joint model estimating the correlation or severely immunodeficient at age 10 (Table 1). between the two outcomes at a specific time and the correla- Regions differed by follow-up length, year of birth and data tion between the two outcomes evolution over time. The sec- contribution over time. Median age at last visit was 15.2 ond strategy’s aim was to identify profiles of co-evolution of years (IQR 12.8;17.4) in NA&E versus 12.4 (IQR 11.1;14.1) in CD4 and HAZ. West&CA. Mortality rate ranged from 1% to 4% (CSA&C). At Analyses were conducted using SAS version 9.3 (SAS Insti- last visit, from 9% to 13% were considered lost-to-follow-up tute Inc., Cary, NC, USA), fractional polynomials were initially in all regions except NA&E where this rate was 31%. Trans- determined using Stata version 13.0 (StataCorp, College Sta- fer to adult care differed greatly by regions as well, from 2% tion, TX, USA) and the “fp” package. Figures were plotted reported in West&CA to 19% (NA&E) and 21% in Botsw.&SA. using the ggplot2 package in R version 3.2.2 (R Foundation for Between age 10 and 14 years, East&SA contributed the most Statistical Computing, Vienna, Austria). growth and CD4 data (72% and 48%, respectively). Between age 15 and 17 years, East&SA data contribution decreased due to shorter follow-up, with NA&E contributing the majority 3 R E S U LT S of CD4 data (60%) (Tables S1 and S2). Compared to the study population, APH who were excluded 3.1 Selection and characteristics of the study because of lack of height or CD4 measurement were more population often from sub-Saharan Africa (87% vs. 73%) and more likely Between 1994 and 2015, data on 35,315 APH across 46 to have missing HAZ or CD4 (70% vs. 47%) at ART initiation. countries were collected. Respectively, 78% and 74% had ≥ 1 Those excluded because they had initiated ART after age 10 3 Table 1. Characteristics of study participants for the growth and CD4 analyses, stratified by sex Growth analyses CD4 analyses Total Males Females Total Males Females Characteristics (N = 20,939) (n = 10,512) (n = 10,427) p-value* (N = 19,557) (n = 9850) (n = 9707) p-value* At ART initiation Age, median (IQR) 6.9 (4.4; 8.5) 6.8 (4.4; 8.5) 6.9 (4.4; 8.5) 0.089 6.9 (4.4; 8.5) 6.8 (4.4; 8.5) 6.9 (4.3; 8.5) 0.227 Age groups (years), n % 0.017 0.014 0–2 2391 11 1162 11 1229 12 2379 12 1141 12 1238 13 2–5 3795 18 1977 19 1818 17 3477 18 1802 18 1675 17 5–10 14,753 70 7373 70 7380 71 13,701 70 6907 70 6794 70 HAZ, median (IQR) –1.94 –1.96 –1.92 0.052 –1.92 –1.94 –1.90 0.038 (–2.87; –1.02) (–2.91; –1.06) (–2.83; –1.00) (–2.84; –1.01) (–2.88; –1.05) (–2.80; –0.97) Severity of stunting, n % 0.100/ 0.225/ 0.147 0.152 None 5785 28 2836 27 2949 28 4755 24 2345 24 2410 25 Moderate (]–3; –2] SD) 2856 14 1417 13 1439 14 2338 12 1170 12 1168 12 Severe (<-3 SD) 2478 12 1273 12 1205 12 1947 10 1011 10 936 10 Missing 9820 47 4986 47 4834 46 10,517 54 5324 54 5193 53 WHZ/BAZ, median (IQR) –0.51 –0.53 –0.50 0.475 –0.51 –0.54 –0.48 0.110 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 (–1.36; 0.27) (–1.44; 0.32) (–1.29; 0.22) (–1.36; 0.27) (–1.44; 0.30) (–1.28; 0.24) Severity of wasting, n % <0.001 <0.001 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 None 9492 45 4653 44 4839 46 7717 39 3802 39 3915 40 Moderate (]–3; –2] SD) 959 5 512 5 447 4 775 4 424 4 351 4 Severe (<-3 SD) 564 3 319 3 245 2 462 2 267 3 195 2 Missing 9924 47 5028 48 4896 47 10,603 54 5357 54 5246 54 Immunodeficiency by agea , n % 0.002/ 0.020/ 0.001 0.014 None 2940 14 1384 13 1556 15 3093 16 1479 15 1614 17 Moderate 2090 10 1040 10 1050 10 2209 11 1111 11 1098 11 Severe 6011 29 3078 29 2933 28 6395 33 3262 33 3133 32 Missing 9898 47 5010 48 4888 47 7860 40 3998 41 3862 40 At 10 years of age HAZ, median (IQR) –1.48 –1.41 –1.55 <0.001 –1.38 –1.30 –1.46 <0.001 (–2.32; –0.67) (–2.21; –0.65) (–2.40; –0.68) (–2.22; –0.53) (–2.14; –0.49) (–2.32; –0.57) (Continued) 4 Table 1. Continued Growth analyses CD4 analyses Total Males Females Total Males Females Characteristics (N = 20,939) (n = 10,512) (n = 10,427) p-value* (N = 19,557) (n = 9850) (n = 9707) p-value* Severity of stunting, n % <0.001 <0.001 No 11,299 54 5885 56 5414 52 7287 37 3782 38 3505 36 Moderate (]–3; –2] SD) 4061 19 1992 19 2069 20 2374 12 1164 12 1210 12 Severe (<-3 SD) 1755 8 726 7 1029 10 975 5 406 4 569 6 Missing 3824 18 1909 18 1915 18 8921 46 4498 46 4423 46 BAZ, median (IQR) –0.50 –0.49 –0.51 0.810 –0.44 –0.44 –0.44 0.566 (–1.20; 0.18) (–1.22; 0.20) (–1.18; 0.17) (–1.15; 0.26) (–1.18; 0.27) (–1.11; 0.24) Severity of wasting, n % 0.012/ 0.029/ 0.004 0.011 None 15,455 74 7721 73 7734 74 9708 50 4852 49 4856 50 Moderate (]–3; –2] SD) 1092 5 563 5 529 5 617 3 326 3 291 3 Severe (<-3 SD) 341 2 200 2 141 1 174 1 105 1 69 1 Missing 4051 19 2028 19 2023 19 9058 46 4567 46 4491 46 CD4 count, median (IQR) 729 705 753 <0.001 731 710 751 <0.001 (478; 1003) (466; 978) (490; 1036) (491; 998) (477; 507) (507; 1025) Immunodeficiency, n % 0.016/ <0.001 0.009 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 None 8919 43 4428 42 4491 43 10,788 55 5305 54 5483 56 Moderate 1492 7 781 7 711 7 1810 9 964 10 846 9 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 Severe 1817 9 964 9 853 8 1921 10 1016 10 905 9 Missing 8711 42 4339 41 4372 42 5038 26 2565 26 2473 25 At last visit Age, median (IQR) 12.8 12.8 12.8 0.437 13.0 13.0 13.0 0.397 (11.3; 14.9) (11.3; 14.9) (11.3; 14.9) (11.6; 15.0) (11.6; 14.9) (11.6; 15.0) Follow-up status 0.678 0.857 Alive, in follow-up 13,803 66 6946 66 6857 66 12,387 63 6246 63 6141 63 Lost-to-follow-upb 3783 18 1878 18 1905 18 4309 22 2154 22 2154 22 Transferred 3022 14 1513 14 1509 14 2538 13 1281 13 1281 13 Dead 331 2 178 2 156 2 323 2 169 2 169 2 Note: Missing categories for stunting, wasting and immunodeficiency correspond to missing data for HAZ, WAZ or CD4 count, respectively. Abbreviations: ART, antiretroviral therapy; BAZ, BMI-for-age Z-score; ; HAZ, height-for-age Z-score; IQR, interquartile range; WHZ, weight-for-height Z-score. a By WHO 2006 guidelines. b Defined as no contact for more than 365 days or documented loss-to-follow-up. *Chi-square or Kruskal–Wallis tests, second p-value corresponding to the test excluding the missing category, if different from the initial test. CIPHER global cohort collaboration, 1994–2015. 5 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Figure 2. Mean height-for-age Z-score evolution between 10 and 17 years of age, stratified by regions (sub-Saharan Africa at the bottom and other regions at the top) and sex, with sample size by year. CIPHER global cohort collaboration, 1994–2015. were almost all living in sub-Saharan Africa (97%). Those who Despite significant HAZ improvements over time, particularly were not part of the growth model because of missing HAZ in females, those moderately or severely stunted at age 10 did at age 10 had similar characteristics to those in the analysis. not reach similar values to those non-stunted at the same age (Figure 3). 3.2 Growth evolution during adolescence 3.3 CD4 evolution during adolescence APH from NA&E had the highest HAZ values throughout ado- lescence, followed by the Asia-Pacific and CSA&C. Growth All regions followed the same pattern, with a CD4 decline patterns in sub-Saharan African regions greatly differed by overtime (Figure 4). The same pattern was observed in sen- sex, with females experiencing HAZ gains over time, while sitivity analyses using age-adjusted CD4 ratio instead of CD4 males experienced a decline in HAZ in early adolescence cells/mm3 (Figure S2). From age 10 to age 16 years, mean before a return to HAZ values that were similar by age 17 to CD4 counts dropped from 768 to 607 cells/mm3 (i.e. 0.78– those at age 10 years (Figure 2). 0.65 in age-adjusted CD4 ratio). Overall, severe immunodeficiency and severe wasting at Adjusted estimates for CD4 counts at age 10 were higher age 10 was associated with smaller HAZ improvements over in females, those initiated early on ART and born after time compared to those with no immunodeficiency or wast- year 2000 (Table 3). A CD4 decline was observed what- ing. Female adolescents severely stunted at ART initiation ever the CD4 level at age 10. Those moderately or severely had smaller HAZ improvements over time compared to non- immunodeficient at age 10 had CD4 improvements in the stunted females (Table 2). Across regions, those who initiated first years of adolescence, followed by a plateau, those non- ART between age 5 and 10 years had lower HAZ values at immunodeficient at age 10 experienced a CD4 decline, even age 10 and over time during adolescence compared to those if overall they remain at higher CD4 count level than the two who initiated ART between age 0 and 5 years (Figure S1). other groups (Figure S3). APH born before 1996 had lower 6 Table 2. Associated factors to height-for-age Z-score evolution between age 10 and 17 years for males and females: multivariate linear mixed model with fractional polynomials For males For females Between 10 and 19 years of Between 10 and 19 years of At 10 years of age age At 10 years of age age Mean Mean Mean Mean difference difference by difference difference by Variables (coef (sd)) p-value year (coef (sd)) p-value (coef (sd)) p-value year (coef (sd)) p-value Immunodeficiency at age 10 0.896 <0.001 0.002 <0.001 Moderate versus no −0.003 (0.015) 0.853 −0.002 (0.001) 0.037 −0.030 (0.017) 0.084 −0.008 (0.004) 0.070 Severe versus no −0.009 (0.014) 0.502 −0.006 (0.001) <0.001 −0.047 (0.016) 0.004 −0.012 (0.004) 0.003 Missing versus no 0.002 (0.011) 0.850 −0.001 (0.001) 0.283 0.015 (0.012) 0.212 0.001 (0.004) 0.691 Stunting at ART initiation <0.001 <0.001 0.084 <0.001 Moderate versus no −0.032 (0.014) 0.021 0.001 (0.001) 0.191 −0.037 (0.015) 0.013 0.001 (0.004) 0.842 Severe versus no −0.062 (0.015) <0.001 0.001 (0.001) 0.485 −0.030 (0.018) 0.094 −0.014 (0.005) 0.003 Missing versus no −0.021 (0.011) 0.056 0.000 (0.001) 0.891 −0.018 (0.012) 0.140 −0.008 (0.003) 0.012 Wasting at age 10 <0.001 <0.001 <0.001 <0.001 Moderate versus no −0.083 (0.017) <0.001 −0.007 (0.001) <0.001 −0.131 (0.020) <0.001 −0.014 (0.006) 0.012 Severe versus no −0.160 (0.028) <0.001 −0.011 (0.002) <0.001 −0.304 (0.039) <0.001 −0.052 (0.012) <0.001 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Missing versus no −0.217 (0.040) <0.001 −0.025 (0.004) <0.001 −0.257 (0.046) <0.001 −0.064 (0.013) <0.001 Region (vs. North America 0.012 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 <0.001 <0.001 <0.001 and Europe) Central, South America and −0.034 (0.026) 0.192 0.000 (0.002) 0.995 −0.093 (0.026) <0.001 0.001 (0.005) 0.833 the Caribbean Asia-Pacific −0.042 (0.020) 0.032 0.007 (0.002) <0.001 −0.057 (0.021) 0.008 −0.003 (0.005) 0.542 West and Central Africa −0.027 (0.023) 0.232 −0.011 (0.002) <0.001 −0.039 (0.025) 0.128 −0.013 (0.007) 0.064 Botswana and South Africa −0.059 (0.017) <0.001 −0.009 (0.001) <0.001 −0.098 (0.019) <0.001 −0.013 (0.005) 0.003 East and Southern Africa −0.054 (0.016) <0.001 −0.010 (0.001) <0.001 −0.105 (0.018) <0.001 −0.035 (0.004) <0.001 (except Botsw. and SA) Note: Adjusted on fractional polynomials for time since age 10 (mfp 2,3), age at ART initiation (mfp 2,2) and height-for-age Z-score at age 10 (mfp –1,0). See Figures S1 and S3 for the estimated growth curves by age at ART initiation and stunting at age 10. Moderate stunting or wasting: HAZ or BAZ = [–3; –2[SD; Severe stunting or wasting: HAZ or BAZ<-3 SD. Moderate immunodeficiency: CD4 cells/mm3 = [250;500[; Severe immunode- ficiency: CD4 cells/ mm3 <250. CIPHER global cohort collaboration, 1994–2015. Bold values referred to statistical significance for each variable at a threshold of <0.05. 7 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Figure 3. Adjusted estimated mean height-for-age Z-scores for adolescents living with perinatally acquired HIV, stratified by stunting at age 10 years, by sex and regions. CIPHER global cohort collaboration, 1994–2015. CD4 counts than those born after, especially with those born 10 were significant markers of poor growth and poor immune after 2000 who had not reached late adolescence by the time response during adolescence. the database was closed. Thus, the CD4 decline observed Lower HAZ among children and adolescents living with HIV may be partly explained by different data contributions, with compared to their HIV-uninfected peers have been exten- higher CD4 count observed before age 15 due to data from sively described in both high and low-and-middle income those born after 2000, and lower CD4 counts after age 15 countries, with demonstrated benefits of early ART treat- because of those born before 2000. However, even adjusting ment on catch-up growth [17,19,22,23]. Growth differences on this parameter, the CD4 decline remained. observed by region may highlight differences in access to HIV care, with HAZ during adolescence proportionately increasing with the rate of APH receiving ART before the age of 2. The 3.4 Joint evolution between CD4 and burden of malnutrition in the general population also varies height-for-age greatly by country-income groups, which can be an underlying Neither of the two statistical methods implemented provided condition increasing the risk of non-reversible stunting dur- evidence of HAZ and CD4 co-evolution, even when tak- ing adolescence. Indeed, repeated exposures to malnutrition ing into account the heterogeneity of growth evolution by during infancy result in growth retardation, and may lead to regions and sex and conducting sub-group analyses. Correla- puberty delay, that can further slowdown height gains [39,40]. tions between the two outcomes were very low for every sub- Furthermore, maternal determinants, such as young age and group considered, and no clear patterns of a joint evolution poor nutrition status, are also strong predictors of long-term emerged (Supplementary results). stunting in children, and are most commonly found in low- income groups [40]. In the Asia-Pacific, the use of WHO growth standards instead of national growth chart references 4 DISCUSSION may have resulted in relatively lower HAZ values. APH experience puberty delays compared to HIV- This global analysis conducted in a large sample size repre- uninfected adolescents [41–43], which may explain the senting high-, middle- and low-income countries allowed an high prevalence of stunting observed in APH. Most studies extended description of growth and CD4 patterns of APH have been conducted in youth with late access to ART, at enrolled in care between 1994 and 2015. Adolescents living similar ages to our participants from sub-Saharan African with HIV worldwide experienced different linear growth pat- regions. Few studies have investigated puberty delays in terns by region and sex, while their CD4 patterns were sim- early treated APH, or explored its association with growth ilar, showing a decline from the age of 10. Late ART initia- for APH. In a study in the United States, APH had delayed tion, along with severe stunting and low CD4 cell count at age sexual maturity compared to HIV-exposed but uninfected 8 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Figure 4. Mean CD4 count evolution between 10 and 19 years of age, stratified by regions and sex, with sample size by year. CIPHER global cohort collaboration, 1994–2015. adolescents, with HAZ mediating the effect of HIV on age First, this decline may be related to low levels of viral sup- at sexual maturity for both males and females [42]. Puberty pression or ART adherence across adolescence [24,28,44], delays were also observed in APH living in Uganda and variables we were not able to measure. Other studies have Zimbabwe, with greater delays for those stunted [43]. Across shown that despite sustained viral suppression, people living regions, growth patterns varied by sex, with more pronounced with HIV may experience low CD4 count levels and chronic differences in sub-Saharan African regions, while males and immune activation [45–47], a potential factor explaining CD4 females had similar characteristics at ART initiation and age count depletion [48]. For example, a high proportion of adults 10. We previously highlighted these differences within the living with HIV in sub-Saharan Africa had suboptimal immune IeDEA and the EPPICC networks, from which data are also recovery after 6 years of ART in a multicountry prospective included in this analysis [17,23]. In Uganda and Zimbabwe, cohort [49]; the CD4/CD8 ratio, an indicator of immune acti- males had more height disadvantage compared to female APH vation and inflammation, has been shown to remain low in throughout adolescence [43]. Additional data are needed to long-term ART-treated adults [50], especially if initiated late further explain these sex differences in growth development on ART [50,51]. Also, based on this CD4/CD8 ratio, a third of for APH, describing the timing of puberty onset, and hor- children and adolescents living with HIV who maintained viral monal and metabolic changes that may be associated with load suppression did not achieve immune recovery in a Span- growth. ish cohort [52] as well as in a UK cross-sectional study [53]. While growth patterns may be dependent on nutrition and Less than half had reached CD4/CD8 ratio normalization in HIV care history during childhood, similar CD4 evolution was Thailand [54]. A Kenyan study showed high immune activation observed across settings, with comparable CD4 values at age levels and depleted CD4 percentages [55]. If APH experience 10 followed by no CD4 count increase, or even a CD4 decline suboptimal CD4 counts while being virologically suppressed, over time. While this observed decline rarely reached immun- that could increase their risk of experiencing opportunistic odeficiency levels, this result has to be further explored. infections and advanced stage of HIV disease, especially in 9 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Table 3. Associated factors to CD4 count evolution (cells/mm3 ), mutlivariate linear mixed model At 10 years of age Between 10 and 19 years of age Mean Mean difference difference by year Variables (coef (sd)) p-value (coef (sd)) p-value Sex (girls vs. boys) 41 (5) <0.001 −8 (2) <0.001 Age groups at ART initiation (years) <0.001 <0.001 0–2 versus 5–10 79 (8) <0.001 −7 (3) <0.001 2–5 versus 5–10 86 (7) <0.001 −13 (2) <0.001 Immunodeficiency for age at ART initiation <0.001 0.005 Moderate versus no −99 (9) <0.001 10 (3) <0.001 Severe versus no −126 (7) <0.001 7 (2) 0.004 Missing versus no −96 (7) <0.001 5 (2) 0.034 Immunodeficiency for age at 10 years of age <0.001 <0.001 Moderate versus no −443 (8) <0.001 53 (3) <0.001 Severe versus no −613 (8) <0.001 82 (2) <0.001 Missing versus no −116 (8) <0.001 4 (3) 0.105 Severity of stunting at 10 years of age <0.001 <0.001 Moderately versus no 38 (7) <0.001 1 (3) 0.765 Severely versus no 19 (11) 0.080 17 (4) <0.001 Missing versus no −16 (6) 0.012 9 (2) <0.001 Year of birth <0.001 0.011 <1996 versus [1966–2000[ −41 (8) <0.001 1 (2) 0.703 ≥2000 versus [1996–2000[ 25 (6) <0.001 8 (3) 0.003 Region (vs. North America and Europe) <0.001 0.676 Central, South America and the Caribbean −28 (13) 0.025 0 (4) 0.996 Asia-Pacific 18 (11) 0.084 1 (3) 0.816 West and Central Africa 12 (12) 0.299 −5 (4) 0.227 Botswana and South Africa 58 (9) <0.001 −3 (3) 0.247 East and Southern Africa (except Bostw. and 22 (9) 0.015 −1 (3) 0.651 SA) Note: Moderate stunting or wasting: HAZ or BAZ = [–3; –2[SD; Severe stunting or wasting: HAZ or BAZ<-3 SD. Moderate immunodeficiency: CD4 cells/mm3 = [250;500[; Severe immunodeficiency: CD4 cells/mm3 <250. CIPHER global cohort collaboration, 1994–2015. Bold values referred to statistical significance for each variable at a threshold of <0.05. sub-Saharan African settings where other infectious diseases possible differences in nutritional care that could mitigate the are highly prevalent. The phenomenon of chronic immune results. Potential height gains due to a recent immune recov- activation and persistent inflammation is of particular impor- ery can take time to be observed and then may not be trans- tance as it is associated with higher risk of morbidity and mor- lated easily with joint modelling methods. Future work might tality among people living with HIV [56]. The CD4 decline we investigate how the two outcomes interact by building and observed may be due to this phenomenon, but more specific testing a causal model. We showed that severe immunodefi- biological measurements are needed to verify this hypothesis. ciency at age 10 was associated with lower height gains dur- Another explanation for this CD4 decline would be infor- ing adolescence. Previous studies on the role of CD4 level mation bias, with adolescents born before year 2000, mostly in APH final height showed contradictory results, no associ- from North America and Europe, contributing more CD4 data ations between CD4 percentage and final height were seen in after 15 years of age compared to those born after 2000. a birth cohort in Thailand [57], while it was positively corre- This population may have initiated ART on an older regimen, lated in a small cohort in the United States [58]. which was less efficient regarding to long-term immune recov- The limitations to this study include that key variables, ery. However, even if less pronounced, a CD4 decline was such as viral load, ART adherence or ART regimen, were observed while adjusting by year of birth. not included because of their sporadic documentation across Our secondary objective on determining a joint evolu- settings. Regarding ART regimen, previous studies did not tion between CD4 and growth did not show any conclusive highlight significant associations between ART regimen at ART results, even when stratifying by regions to take into account initiation and catch-up growth [40] or final height [57], while 10 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 its role on long-term CD4 evolution has not been assessed in raj Hospital, Mahidol University, Thailand), Sophie Le Coeur APH to our knowledge. A conservative definition of perinatal (Institut National d’Etude Demographique (INED), Mortal- HIV acquisition was used, with a threshold of 10 years of ity, Health and Epidemiology Unit, Paris, France, Institut de age at inclusion if not documented. We may have excluded a Recherche pour le Developpement (IRD), UMI-174/PHPT, subset of APH accessing HIV care late, who could have higher Faculty of Associated Medical Sciences, Chiang Mai Univer- rates of stunting and different CD4 profiles due to either sity, Chiang Mai, Thailand), Marie-Hélène Aké-Assi (Univer- slow HIV progression [59] or advanced HIV disease. Children sity Hospital Yopougon, Abidjan, Côte d’Ivoire), Kunjal Patel who did not reach adolescence due to loss-to-follow-up or (Harvard T. H. Chan School of Public Health, Boston, Mas- death did not contribute to the data, leading to a survival sachusetts, United States of America), Jorge Pinto (School of bias. Data contribution varied across settings between early Medicine, Federal University of Minas Gerais, Belo Horizonte, and late adolescence, median age at last visit was especially Brazil), Mary Paul (Baylor International Pediatric AIDS Initia- low in sub-Saharan Africa; therefore, the growth and CD4 tive, Texas Children’s Hospital-USA, Houston, Texas, United data observed during late adolescence have to be interpreted States of America), Rachel Vreeman (Department of Global with caution as they may come from a selected popula- Health, Icahn School of Medicine at Mount Sinai, New York, tion. Despite those limitations, this study provides critical New York, USA), Mary-Ann Davies (School of Public Health evidence on growth and CD4 data for APH, thanks to a and Family Medicine, Faculty of Health Sciences, University of large, multi-regional dataset, developed by a unique research Cape Town, Cape Town, South Africa), Ruth Goodall † (MRC collaboration on paediatric and adolescent HIV. Clinical Trials Unit, University College London, London, United Kingdom), Valériane Leroy † (CERPOP, Inserm, Université de Toulouse, Université Paul Sabatier Toulouse 3, Toulouse, 5 CONCLUSIONS France). * Project Statisticians/Epidemiologists; † Project co- chairs, should be considered joint senior author. These original results support the importance of closely mon- Data Coordinating Team: Nicky Maxwell (University of itoring growth and CD4 patterns during adolescence, to fur- Cape Town, Cape Town, South Africa), Charlotte Duff (MRC ther develop appropriate and timely interventions allowing Clinical Trials Unit at University College London, London, UK). APH to reach adulthood as healthy as possible. Our find- Writing Team: CIPHER Steering Committee: Elaine Abrams ings on CD4 progression are especially relevant in a time (ICAP at Columbia University, Mailman School of Public where viral load monitoring is becoming the reference to mea- Health, Columbia University, New York, NY, United States of sure ART effectiveness [60], at the expense of CD4 monitor- America), Mary Paul (Baylor International Pediatric AIDS Ini- ing that may become less available in low-and-middle income tiative at Texas Children’s Hospital, USA), Jihane Ben-Farhat countries. Our results advocate for maintaining CD4 count (Epicentre, Médecins Sans Frontières, France), Russell Van data collection as it provides helpful information on HIV- Dyke (Tulane University Health Sciences Center, USA), Ali related clinical complications [49,61]. CD4 depletion and the Judd (MRC Clinical Trials Unit at UCL, UK), Rachel Vreeman potential role of chronic inflammation on this mechanism [62] (Department of Global Health, Icahn School of Medicine at is a critical feature to describe and may be a threat for Mount Sinai, New York, New York, USA), Lynne Mofenson long-term ART-treated patients. Even though our results may (Elizabeth Glaser Pediatric AIDS Foundation, USA), Marissa appear slightly outdated compared to the 2020 recommended Vicari (International AIDS Society, Switzerland), George Seage paediatric HIV care standards, optimal access to high standard III (Harvard T. H. Chan School of Public Health, USA), Linda- of care remains challenging for some resource-limited regions Gail Bekker (Desmond Tutu HIV Centre, University of Cape that are still facing difficulties to access the most recent Town, South Africa), Shaffiq Essajee (UNICEF), Diana Gibb and efficient treatment strategies [63]. This multi-regional, (MRC Clinical Trials Unit at University College London, Lon- extended description of growth and CD4 patterns, in a large don, UK); and diverse population of APH, should help further support CIPHER Executive Committee: Lynne Mofenson (Elizabeth HIV care for APH worldwide. Glaser Pediatric AIDS Foundation, USA), Linda-Gail Bekker (Desmond Tutu HIV Centre, University of Cape Town, South T H E C I P H E R G LO B A L C O H O RT Africa), Marissa Vicari (International AIDS Society, Switzer- C O L L A B O R AT I O N land), Shaffiq Essajee (UNICEF), Martina Penazzato (World Health Organization, Switzerland); Project Team: Julie Jesson* (CERPOP, Inserm, Université CIPHER Project Oversight Group: Intira Jeannie Collins (MRC de Toulouse, Université Paul Sabatier Toulouse 3, Toulouse, Clinical Trials Unit at UCL, UK), Kara Wools-Kaloustian (Indi- France), Siobhan Crichton* (MRC Clinical Trials Unit, Uni- ana University School of Medicine, USA), Mary-Ann Davies versity College London, London, United Kingdom), Matteo (University of Cape Town, South Africa), Valériane Leroy Quartagno (MRC Clinical Trials Unit, University College Lon- (CERPOP, Inserm, Université de Toulouse, Université Paul don, London, United Kingdom), Marcel Yotebieng (Division of Sabatier Toulouse 3, Toulouse, France), Ruth Goodall (MRC General Internal Medicine, Department of Medicine, Albert Clinical Trials Unit at UCL, UK), Kunjal Patel (Harvard T. H. Einstein College of Medicine, Bronx, NY, United States of Chan School of Public Health, USA), Ali Judd (MRC Clini- America), Elaine. J. Abrams (ICAP at Columbia University, cal Trials Unit at UCL, UK), Rachel Vreeman (Department of Mailman School of Public Health, Columbia University, New Global Health, Icahn School of Medicine at Mount Sinai, New York, NY, United States of America), Kulkanya Chokephaibulkit York, New York, USA), Amy Slogrove (Department of Paedi- (Siriraj Institute of Clinical Research, Faculty of Medicine, Siri- atrics & Child Health, Faculty of Medicine & Health Sciences, 11 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Stellenbosch University, Worcester, South Africa), Kate Powis Jean William Pape (GHESKIO Center, Port-au-Prince, Haiti), (Harvard T. H. Chan School of Public Health, USA), Paige Vanessa Rouzier (GHESKIO Center, Port-au-Prince, Haiti), Williams (Harvard T. H. Chan School of Public Health, USA), Adias Marcelin (GHESKIO Center, Port-au-Prince, Haiti), Siobhan Crichton (MRC Clinical Trials Unit at UCL, UK), Julie Regina Succi (Universidade Federal de Sao Paulo, Brazil); Jesson (CERPOP, Inserm, Université de Toulouse, Université IeDEA Asia-Pacific: Kulkanya Chokephaibulkit (Faculty of Paul Sabatier Toulouse 3, Toulouse, France), George Seage III Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thai- (Harvard T. H. Chan School of Public Health, USA); land), Annette H Sohn (TREAT Asia/amfAR, Bangkok, Thai- BIPAI: Mogomotsi Matshaba (Baylor College of Medicine land), Azar Kariminia (Kirby Institute, University of New South Children’s Foundation-Botswana), Lineo Thahane (Baylor Col- Wales, Sydney, Australia); lege of Medicine Children’s Foundation-Lesotho), Phoebe IeDEA Central Africa: Marcel Yotebieng (Division of Gen- Nyasulu (Baylor College of Medicine Children’s Foundation- eral Internal Medicine, Department of Medicine, Albert Ein- Malawi), Bhekumusa Lukhele (Baylor College of Medicine stein College of Medicine, Bronx, NY, United States of Amer- Children’s Foundation-eSwatini), Lumumba Mwita (Baylor Col- ica), Andrew Edmonds (Gillings School of Global Public Health, lege of Medicine Children’s Foundation-Tanzania), Adeodata University of North Carolina at Chapel Hill, USA), Patricia Kekitiinwa-Rukyalekere (Baylor College of Medicine Children’s Lelo (Pediatric Hospital Kalembe Lembe, Lingwala, Kinshasa, Foundation – Uganda), Sebastian Wanless (Baylor Interna- Demogratic Republic of Congo); tional Pediatric AIDS Initiative at Texas Children’s Hospital, IeDEA East Africa: Rita Lyamuya (Morogoro Regional Hos- Data Manager); pital, Morogoro, Tanzania), Edith Apondi Ogalo (Moi Teach- EPPICC: Siobhan Crichton (MRC Clinical Trials Unit at ing and Referral Hospital, Kenya), Francesca Akoth Odhiambo University College London, London, UK), Matteo Quartagno (Center for Microbiology Research, Kenya Medical Research (MRC Clinical Trials Unit at University College London, Lon- Institute, Nairobi, Kenya), Kara Wools-Kaloustian (Indiana Uni- don, UK), Intira Jeannie Collins (MRC Clinical Trials Unit versity School of Medicine, Department of Medicine, Division at University College London, London, UK), Ruth Goodall of Infectious Diseases, Indianapolis, Indiana), Rachel Vreeman (MRC Clinical Trials Unit at University College London, Lon- (Department of Global Health, Icahn School of Medicine at don, UK), Tessa Goetghebuer (Hospital St Pierre, Brussels, Mount Sinai, New York, New York, USA); Belgium), Claire Thorne (UCL Great Ormond Street Insti- IeDEA Southern Africa: Andreas D Haas (Institute of Social tute of Child Health, University College London, UK), Josiane and Preventive Medicine, University of Bern, Switzerland), Warszawski (Inserm U1018, Centre de recherche en Epidémi- Carolyn Bolton (Centre for Infectious Disease Research in ologie et Santé des Populations, France), Luisa Galli (Infec- Zambia, Lusaka, Zambia), Josephine Muhairwe (SolidarMed, tion Disease Unit, Meyer Children’s University Hospital, Flo- Lesotho, Mozambique, Zimbabwe), Hannock Tweya (Light- rence, Italy, Department of Health Sciences, University of Flo- house Trust Clinic, Lilongwe, Malawi), Mary-Ann Davies (Cen- rence, Italy), Annemarie van Rossum (Erasmus MC University ter for Infectious Diseases Epidemiology and Research, Uni- Medical Center Rotterdam-Sophia Children’s Hospital, Rot- versity of Cape Town, Cape Town, South Africa); terdam, The Netherlands), Diana M Gibb (MRC Clinical Tri- IeDEA West Africa: Mariam Sylla (CHU Gabriel Toure, als Unit at University College London, London, UK), Carlo Bamako, Mali), Marceline D’Almeida, (Centre National Hospi- Giaquinto (Padova University/ PENTA Foundation, Italy), Mag- talier Universitaire Hubert K. Maga, Cotonou, Bénin), Lorna dalena Marczynska (Medical University of Warsaw, Hospital of Renner (Korle Bu Teaching Hospital, Accra, Ghana), Marie- Infectious Diseases in Warsaw, Poland), Laura Marques (Cen- Hélène Aké-Assi (University Hospital Yopougon, Abidjan, Côte tro Hospitalar do Porto, Portugal), Filipa Prata (Hospital de d’Ivoire), Valeriane Leroy (CERPOP, Inserm, Université de Santa Maria, Lisboa, Portugal), Luminita Ene (Victor Babes Toulouse, Université Paul Sabatier Toulouse 3, Toulouse, Hospital, Bucharest, Romania), Lyuba Okhonskaya (Republi- France); can Hospital of Infectious Diseases, St Petersburg, Russian IMPAACT/PHACS: Mark J Abzug (University of Colorado Federation), Marisa Navarro (Hospital General Universitario School of Medicine and Children’s Hospital Colorado, USA), “Gregorio Marañón”, Madrid, Spain), Antoinette Frick (Hospital Paige Williams (Harvard T. H. Chan School of Public Health, Universitari Vall d’ Hebron, Vall d’ Hebron Research Institute, USA), James Oleske (Rutgers - New Jersey Medical School, Universitat Autònoma de Barcelona; Barcelona, Spain), Lars USA), Murli Purswani (Bronx-Lebanon Hospital Center, USA), Naver (Karolinska University Hospital and Karolinska Insti- George Seage III (Harvard T. H. Chan School of Public Health, tutet, Stockholm, Sweden), Christian Kahlert (Children’s Hos- USA), Russell Van Dyke (Tulane University, USA), Kunjal Patel pital of Eastern Switzerland, Saint Gallen, Switzerland), Sophie (Harvard T. H. Chan School of Public Health, USA); Le Coeur (Institut National d’Etude Demographique (INED), ICAP-Optimal Models: Elaine Abrams (ICAP-Columbia Mortality, Health and Epidemiology Unit, Paris, France, Institut University, Mailman School of Public Health, USA), Chloe de Recherche pour le Developpement (IRD), UMI-174/PHPT, Teasdale (ICAP-Columbia University, Mailman School of Faculty of Associated Medical Sciences, Chiang Mai Univer- Public Health, USA), Harriet Nuwagaba-Biribonwoha (ICAP- sity, Chiang Mai, Thailand), Ali Judd (MRC Clinical Trials Columbia University, Mailman School of Public Health, USA). Unit at University College London, London, UK), Alla Volokha (Shupyk National Medical Academy of Postgraduate Educa- A U T H O R S ’ A F F I L I AT I O N S tion, Kiev), Elizabeth Chappell (MRC Clinical Trials Unit at 1 CERPOP, Inserm, Université Paul Sabatier Toulouse 3, Toulouse, France; 2 MRC University College London, London, UK); Clinical Trials Unit, University College London, London, UK; 3 Division of General IeDEA-CCASAnet: Jorge Pinto (Department of Pediatrics, Internal Medicine, Department of Medicine, Albert Einstein College of Medicine, School of Medicine, Federal University of Minas Gerais, Brazil) Bronx, New York, USA; 4 ICAP at Columbia University, Mailman School of Public 12 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 Health, Columbia University, New York, USA; 5 Siriraj Institute of Clinical Research, lines for children, young adults and families all over the world. He is sorely missed Faculty of Medicine Siriraj Hospital, Mahidol University, Salaya, Thailand; 6 Institut by the CIPHER community. The full list of acknowledgements is available as sup- National d’Etude Demographique (INED), Mortality, Health and Epidemiology porting information (Data S1). Unit, Paris, France; 7 Institut de Recherche pour le Developpement (IRD), UMI- 174/PHPT, Faculty of Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand; 8 University Hospital Yopougon, Abidjan, Côte d’Ivoire; 9 Harvard COMPETING INTERESTS T. H. Chan School of Public Health, Boston, Massachusetts, USA; 10 Department MV’s work at CIPHER is funded through Unrestricted Educational grants received of Pediatrics, School of Medicine, Federal University of Minas Gerais, Belo Hor- from ViiV Healthcare and Janssen to the International AIDS Society. AS’s institu- izonte, Brazil; 11 Baylor International Pediatric AIDS Initiative, Texas Children’s tion receives research and travel funding from ViiV Healthcare. CT has received Hospital-USA, Houston, Texas, USA; 12 Department of Global Health, Icahn School grant funding from ViiV Healthcare via the Penta Foundation (to UCL), and person- of Medicine at Mount Sinai, New York, USA; 13 School of Public Health and Fam- ally from ViiV for participation on an advisory board. AJ reports grants from Abb- ily Medicine, Faculty of Health Sciences, University of Cape Town, Cape Town, vie, Bristol Myers Squibb, Gilead, Janssen Pharmaceuticals and ViiV Healthcare South Africa; 14 Epicentre, Médecins Sans Frontières, Paris, France; 15 Tulane Uni- through the PENTA Foundation, and from the European and Developing Countries versity Health Sciences Center, New Orleans, Louisiana, USA; 16 Elizabeth Glaser Clinical Trials Partnership, Gilead Sciences, the International AIDS Society, NHS Pediatric AIDS Foundation, Washington, DC, USA; 17 International AIDS Society, England, Medical Research Council and PENTA Foundation outside the submitted Geneva, Switzerland; 18 Desmond Tutu HIV Centre, University of Cape Town, Cape work. All monies were paid to her institution. IJC received grants from the follow- Town, South Africa; 19 UNICEF, New York, USA; 20 HIV Department, World Health ing companies (via her institution) in the past 3 years: ViiV, AbbVie and Gilead. Organization, Geneva, Switzerland; 21 Indiana University School of Medicine, Indi- All remaining authors declare no competing interests. anapolis, Indiana, USA; 22 Department of Paediatrics & Child Health, Faculty of Medicine & Health Sciences, Stellenbosch University, Worcester, South Africa; 23 Baylor College of Medicine Children’s Foundation, Lilongwe, Botswana; 24 Baylor AUTHORS’ CONTRIBUTIONS College of Medicine Children’s Foundation, Maseru, Lesotho; 25 Baylor College of Medicine Children’s Foundation, Lilongwe, Malawi; 26 Baylor College of Medicine JJ conducted the analysis and wrote the paper, with methodological support from Children’s Foundation, Mbabane, eSwatini; 27 Baylor College of Medicine Chil- SC and MQ and under the supervision of VL and RG. JJ, SC and RG contributed dren’s Foundation, Mwanza, Tanzania; 28 Baylor College of Medicine Children’s to the study design and protocol development. MY represents the IeDEA Central Foundation, Kampala, Uganda; 29 Hospital St Pierre, Brussels, Belgium; 30 UCL African cohort. EA and KP represent the IMPAACT and PHACS cohort and EA rep- Great Ormond Street Institute of Child Health, University College London, Lon- resents Optimal Models. KC represents the IeDEA Asia-Pacific cohort. MhAa, VL don, UK; 31 Inserm U1018, Centre de recherche en Epidémiologie et Santé des and JJ represent the IeDEA West African cohort. JP represents CCASAnet. SLC, Populations, Paris, France; 32 Infection Disease Unit, Meyer Children’s University RG and SC represent the EPPICC cohort. MP represents BIPAI. RV represents Hospital, Florence, Italy; 33 Department of Health Sciences, University of Florence, the IeDEA East Africa cohort. MaD represents the IeDEA Southern Africa cohort. Florence, Italy; 34 Erasmus MC University Medical Center Rotterdam-Sophia Chil- All co-authors participated to the interpretation of the results, and subse- dren’s Hospital, Rotterdam, The Netherlands; 35 Padova University/PENTA Foun- quently revised the manuscript. All authors listed as belonging to the Project dation, Padua, Italy; 36 Medical University of Warsaw, Hospital of Infectious Dis- Team, Data Coordinating Team and the Writing Team have contributed sufficiently eases in Warsaw, Warsaw, Poland; 37 Centro Hospitalar do Porto, Porto, Portugal; to the conception, design, data collection, analysis, writing and/or review of the 38 Hospital de Santa Maria, Lisboa, Portugal; 39 Victor Babes Hospital, Bucharest, manuscript to take public responsibility for it. All authors have read and approved Romania; 40 Republican Hospital of Infectious Diseases, St Petersburg, Russian the final manuscript. Federation; 41 Hospital General Universitario “Gregorio Marañón”, Madrid, Spain; 42 Hospital Universitari Vall d’ Hebron, Vall d’ Hebron Research Institute, Univer- sitat Autònoma de Barcelona, Barcelona, Spain; 43 Karolinska University Hospi- FUNDING tal and Karolinska Institutet, Stockholm, Sweden; 44 Children’s Hospital of East- This work was supported by the International AIDS Society – Collaborative Initia- ern Switzerland, Saint Gallen, Switzerland; 45 Shupyk National Medical Academy of tive for Paediatric HIV Education & Research (IAS-CIPHER, http://www.iasociety. Postgraduate Education, Kiev, Ukraine; 46 GHESKIO Center, Port-au-Prince, Haiti; org/CIPHER), which is made possible through funding from CIPHER Founding 47 Universidade Federal de Sao Paulo, Sao Paulo, Brazil; 48 TREAT Asia/amfAR, Sponsor ViiV Healthcare (https://www.viivhealthcare.com) and Janssen (http:// Bangkok, Thailand; 49 Kirby Institute, University of New South Wales, Sydney, www.janssen.com). New South Wales, Australia; 50 Gillings School of Global Public Health, University Individual networks contributing to the CIPHER Cohort Collaboration have of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA; 51 Pediatric received the following financial support: CCASAnet receives funding from Hospital Kalembe Lembe, Lingwala, Demogratic Republic of Congo; 52 Morogoro the United States (US) National Institutes of Health (NIH) (grant number Regional Hospital, Morogoro, Tanzania; 53 Moi Teaching and Referral Hospital, U01AI069923); IeDEA Asia Pacific receives funding from the US NIH (grant Eldoret, Kenya; 54 Center for Microbiology Research, Kenya Medical Research number U01AI069907); IeDEA Central Africa receives funding from the US Institute, Nairobi, Kenya; 55 Institute of Social and Preventive Medicine, Univer- NIH (grant number U01AI096299); IeDEA East Africa receives funding from sity of Bern, Bern, Switzerland; 56 Centre for Infectious Disease Research in Zam- the US NIH (grant number U01A1069911); IeDEA Southern Africa receives bia, Lusaka, Zambia; 57 SolidarMed, Lesotho, Zimbabwe; 58 Lighthouse Trust Clinic, funding from the US NIH (grant number U01AI069924); IeDEA West Africa Lilongwe, Malawi; 59 CHU Gabriel Toure, Bamako, Mali; 60 Centre National Hospi- receives funding from the US NIH (grant number U01AI069919); overall sup- talier Universitaire Hubert K. Maga, Cotonou, Benin; 61 Korle Bu Teaching Hospital, port for the International Maternal Pediatric Adolescent AIDS Clinical Trials Accra, Ghana; 62 University of Colorado School of Medicine and Children’s Hospital Network (IMPAACT) was provided by the National Institute of Allergy and Colorado, Aurora, Colorado, USA; 63 Rutgers - New Jersey Medical School, Newark, Infectious Diseases (NIAID) with co-funding from the Eunice Kennedy Shriver New Jersey, USA; 64 Bronx-Lebanon Hospital Center, Bronx, New York, USA National Institute of Child Health and Human Development (NICHD) and the National Institute of Mental Health (NIMH), all components of the NIH, under Award Numbers UM1AI068632 (IMPAACT LOC), UM1AI068616 (IMPAACT SDMC) and UM1AI106716 (IMPAACT LC), and by NICHD contract number ACKNOWLEDGEMENTS HHSN275201800001I; PHACS receives funding from the US NIH (grant numbers The authors thank all participating networks, clinics, clinic personnel and patients U01 HD052102 and U01 HD052104); the Optimal Models (ICAP at Columbia who contributed data and made the Collaborative Initiative for Paediatric HIV Edu- University) project was supported by the President’s Emergency Plan for AIDS cation and Research (CIPHER) Global Cohort Collaboration Adolescent Project Relief (PEPFAR) through the Centers for Disease Control and Prevention (coop- possible. We also thank the CIPHER Executive Committee, the CIPHER Steering erative agreement numbers: 5U62PS223540 and 5U2GPS001537); EPPICC Committee and the Project Oversight Group. The study was sponsored by the receives funding from the PENTA Foundation (http://penta-id.org), and received International AIDS Society-CIPHER. support from the European Union’s Horizon 2020 research and innovation pro- We would like to acknowledge George R. Seage III (1957–2021) for all of his gramme under grant agreement no 825579 for the REACH study. The MRC contributions to the CIPHER Global Cohort Collaboration as the Principal Inves- Clinical Trials Unit at UCL is supported by the Medical Research Council (pro- tigator of the Harvard CIPHER data center and the Pediatric HIV/AIDS Cohort gramme number MC_UU_12023/26). Individual cohorts contributing to EPPICC Study Data and Operations Center. His leadership, collaborative spirit and con- receive the following support: the ATHENA database is maintained by Stichting tributions to understanding the long-term effects of perinatal HIV infection and HIV Monitoring and supported by a grant from the Dutch Ministry of Health, antiretroviral treatment have improved care and antiretroviral drug safety guide- Welfare and Sport through the Centre for Infectious Disease Control of the 13 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 National Institute for Public Health and the Environment. CoRISPE-cat receives 10. Munyayi FK, van Wyk B. The effects of Teen Clubs on retention in HIV care financial support from the Instituto de Salud Carlos III through the Red Temática among adolescents in Windhoek, Namibia. South Afr J HIV Med. 2020;21(1): de Investigación Cooperativa en Sida (grant numbers RED RIS RD06/0006/0035 1031. yRD06/0006/0021). Financial support for CoRISpeS and Madrid Cohort was pro- 11. Makadzange AT, Higgins-Biddle M, Chimukangara B, Birri R, Gordon M, vided by the Instituto de Salud Carlos III through the Red Tematica de Investiga- Mahlanza T, et al. Clinical, virologic, immunologic outcomes and emerging HIV drug cion Cooperativa en Sida (RED-RIS) project (grant number RD16/0025/0019) as resistance patterns in children and adolescents in public ART care in Zimbabwe. part of the Plan R+D+I and cofinanced by ISCIII-Subdireccion General de Eval- PLoS One. 2015;10(12):e0144057. uación and Fondo Europeo de Desarrollo Regional (FEDER). FIS PI19/01530. 12. Kariminia A, Law M, Davies MA, Vinikoor M, Wools-Kaloustian K, Leroy V, The Swiss HIV Cohort Study is supported by the Swiss National Science Foun- et al. Mortality and losses to follow-up among adolescents living with HIV in the dation (grant number: 177499) and by the SHCS Research Foundation. The Thai IeDEA Global Cohort Collaboration. J Int AIDS Soc. 2018;21(12):e25215. cohort study was funded by the Global Fund to fight AIDS, Tuberculosis and 13. Jerene D, Abebe W, Taye K, Ruff A, Hallstrom I. Adolescents living with HIV Malaria, Thailand (PR-A-N-008); Oxfam Great Britain, Thailand (THAA51); Min- are at higher risk of death and loss to follow up from care: analysis of cohort data istry of Public Health, Thailand; and Institut de Recherche pour le Développement from eight health facilities in Ethiopia. PLoS One. 2019;14(10):e0223655. (IRD), France. Ukraine Paediatric HV Cohort is supported by the PENTA Foun- 14. Bitwale NZ, Mnzava DP, Kimaro FD, Jacob T, Mpondo BCT, Jumanne S. dation. CHIPS is funded by the NHS (London Specialized Commissioning Group) Prevalence and factors associated with virological treatment failure among chil- and has received additional support from Bristol-Myers Squibb, Boehringer Ingel- dren and adolescents on antiretroviral therapy attending HIV/AIDS care and treat- heim, GlaxoSmithKline, Roche, Abbott and Gilead Sciences. The MRC Clinical Tri- ment clinics in Dodoma Municipality, Central Tanzania. J Pediatr Infect Dis Soc. als Unit at UCL is supported by the Medical Research Council (programme number 2021;10(2):131–40. MC_UU_12023/26). 15. Fairlie L, Karalius B, Patel K, van Dyke RB, Hazra R, Hernán MA, et al. CD4+ No funding bodies had any role in study design, data collection and analysis, and viral load outcomes of antiretroviral therapy switch strategies after virologic decision to publish or preparation of the manuscript. failure of combination antiretroviral therapy in perinatally HIV-infected youth in the United States. AIDS. 2015;29(16):2109–19. 16. Kouamou V, Varyani B, Shamu T, Mapangisana T, Chimbetete C, Mudzviti T, DISCLAIMER et al. Drug resistance among adolescents and young adults with virologic failure of This work is solely the responsibility of the authors and does not necessarily rep- first-line antiretroviral therapy and response to second-line treatment. AIDS Res resent the official views of any of the institutions mentioned above. Hum Retroviruses. 2020;36(7):566–73. 17. Jesson J, Schomaker M, Malasteste K, Wati DK, Kariminia A, Sylla M, et al. Stunting and growth velocity of adolescents with perinatally acquired HIV: differ- D ATA AVA I L A B I L I T Y S TAT E M E N T ential evolution for males and females. A multiregional analysis from the IeDEA Global Paediatric Collaboration. J Int AIDS Soc. 2019;22(11):e25412. Data are accessible in principle by applying to the Collaborative Initiative for Pae- 18. Chappell E, Lyall H, Riordan A, Thorne C, Foster C, Butler K, et al. The cascade diatric HIV Education and Research (CIPHER) Global Cohort Collaboration Data of care for children and adolescents with HIV in the UK and Ireland, 2010 to 2016. Centres. The CIPHER Global Cohort Collaboration is a multinetwork, multisite J Int AIDS Soc. 2019;22(9):e25379. collaboration and this study combined data from different sites. The data do not 19. McGrath CJ, Diener L, Richardson BA, Peacock-Chambers E, John-Stewart belong to the CIPHER Global Cohort Collaboration itself; data ownership remains GC. Growth reconstitution following antiretroviral therapy and nutritional sup- with the participating sites. Each site has approval from its own local Institutional plementation: systematic review and meta-analysis. AIDS. 2015;29(15):2009– Review Board to collect routine data on patients and to transfer those data anony- 23. mously to the CIPHER Global Cohort Collaboration Project University of Cape 20. Jesson J, Koumakpaï S, Diagne NR, Amorissani-Folquet M, Kouéta F, Aka A, Town Research Centre (Cape Town, South Africa). For some sites and networks, et al. Effect of age at antiretroviral therapy initiation on catch-up growth within the Institutional Review Board approval for use of this data is restricted to the specific first 24 months among HIV-infected children in the IeDEA West African Pediatric protocols approved to protect patient identities. Requests for access to data can Cohort. Pediatr Infect Dis J. 2015;34(7):e159–68. be directed to the corresponding author. 21. Jesson J, Ephoevi-Ga A, Desmonde S, Ake-Assi MH, D’Almeida M, Sy HS, et al. Growth in the first 5 years after antiretroviral therapy initiation among HIV- infected children in the IeDEA West African Pediatric Cohort. Trop Med Int Health. 2019;24(6):775–85. REFERENCES 22. Feucht UD, Van Bruwaene L, Becker PJ, Kruger M. Growth in HIV- 1. UNAIDS. UNAIDS 2019 estimates. Geneva; 2020. infected children on long-term antiretroviral therapy. Trop Med Int Health. 2. Frigati LJ, Ameyan W, Cotton MF, Gregson CL, Hoare J, Jao J, et al. Chronic 2016;21(5):619–29. comorbidities in children and adolescents with perinatally acquired HIV infection in 23. European Pregnancy and Paediatric HIV Cohort Collaboration (EPPICC) sub-Saharan Africa in the era of antiretroviral therapy. Lancet Child Adolesc Health. study group. Height and timing of growth spurt during puberty in young people liv- 2020;4(9):688–98. ing with vertically acquired HIV in Europe and Thailand. AIDS. 2019;33(12):1897– 3. Dahourou D, Raynaud JP, Leroy V. The challenges of timely and safe HIV disclo- 910. sure among perinatally HIV-infected adolescents in sub-Saharan Africa. Curr Opin 24. Desmonde S, Dicko F, Koueta F, Eboua T, Balestre E, Amani-Bosse C, et al. HIV AIDS. 2018;13(3):220–9. Association between age at antiretroviral therapy initiation and 24-month immune 4. Judd A, Davies MA. Adolescent transition among young people with peri- response in West-African HIV-infected children. AIDS. 2014;28(11):1645–55. natal HIV in high-income and low-income settings. Curr Opin HIV AIDS. 25. Simms V, Rylance S, Bandason T, Dauya E, McHugh G, Munyati S, et al. CD4+ 2018;13(3):236–48. cell count recovery following initiation of HIV antiretroviral therapy in older child- 5. Hudelson C, Cluver L. Factors associated with adherence to antiretroviral ther- hood and adolescence. AIDS. 2018;32(14):1977–82. apy among adolescents living with HIV/AIDS in low- and middle-income countries: 26. Stirrup OT, Copas AJ, Phillips AN, Gill MJ, Geskus RB, Touloumi G, et al. Pre- a systematic review. AIDS Care. 2015;27(7):805–16. dictors of CD4 cell recovery following initiation of antiretroviral therapy among 6. Kim SH, Gerver SM, Fidler S, Ward H. Adherence to antiretroviral ther- HIV-1 positive patients with well-estimated dates of seroconversion. HIV Med. apy in adolescents living with HIV: systematic review and meta-analysis. AIDS. 2018;19(3):184–94. 2014;28(13):1945–56. 27. Kufa T, Shubber Z, MacLeod W, Takuva S, Carmona S, Bor J, et al. CD4 count 7. Pantelic M, Casale M, Cluver L, Toska E, Moshabela M. Multiple forms of dis- recovery and associated factors among individuals enrolled in the South African crimination and internalized stigma compromise retention in HIV care among ado- antiretroviral therapy programme: an analysis of national laboratory based data. lescents: findings from a South African cohort. J Int AIDS Soc. 2020;23(5):e25488. PLoS One. 2019;14(5):e0217742. 8. Mugglin C, Haas AD, van Oosterhout JJ, Msukwa M, Tenthani L, Estill J, et al. 28. Picat MQ, Lewis J, Musiime V, Prendergast A, Nathoo K, Kekitiinwa A, et al. Long-term retention on antiretroviral therapy among infants, children, adolescents Predicting patterns of long-term CD4 reconstitution in HIV-infected children and adults in Malawi: a cohort study. PLoS One. 2019;14(11):e0224837. starting antiretroviral therapy in sub-Saharan Africa: a cohort-based modelling 9. Foster C, Ayers S, Mcdonald S, Frize G, Chhabra S, Pasvol TJ, et al. Clini- study. PLoS Med. 2013;10(10):e1001542. cal outcomes post transition to adult services in young adults with perinatally 29. Lewis J, Walker SA, Castro H, De Rossi A, Gibb DM, Giaquinto C, et al. Age and acquired HIV infection: mortality, retention in care and viral suppression. AIDS. CD4 count at initiation of antiretroviral therapy in HIV-infected children: effects on 2020;34(2):261–6. long-term T-cell reconstitution. J Infect Dis. 2020;205(4):548–56. 14 CIPHER Global Cohort Collaboration et al. Journal of the International AIDS Society 2022, 25:e25871 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25871/full | https://doi.org/10.1002/jia2.25871 30. Slogrove AL, Schomaker M, Davies MA, Williams P, Balkan S, Ben-Farhat J, 53. Seers T, Vassallo P, Pollock K, Thornhill JP, Fidler S, Foster C. CD4:CD8 ratio et al. The epidemiology of adolescents living with perinatally acquired HIV: a cross- in children with perinatally acquired HIV-1 infection. HIV Med. 2018;19(9):668– region global cohort analysis. PLoS Med. 2018;15(3):e1002514. 72. 31. CIPHER Global Cohort Collaboration. Inequality in outcomes for adolescents 54. Han WM, Apornpong T, Handoko R, Jantarabenjakul W, Gatechompol S, living with perinatally acquired HIV in sub-Saharan Africa: a Collaborative Initiative Ubolyam S, et al. CD4/CD8 ratio recovery of children and adolescents living with for Paediatric HIV Education and Research (CIPHER) Cohort Collaboration analy- HIV with virological suppression: a prospective cohort study. J Pediatr Infect Dis sis. J Int AIDS Soc. 2018;21(Suppl 1):e25044. Soc. 2021;10(2):88–96. 32. WHO Multicentre Growth Reference Study Group. WHO Child Growth Stan- 55. Alvarez P, Mwamzuka M, Marshed F, Kravietz A, Ilmet T, Ahmed A, et al. dards based on length/height, weight and age. Acta Paediatr Suppl. 2006;450:76– Immune activation despite preserved CD4 T cells in perinatally HIV-infected chil- 85. dren and adolescents. PLoS One. 2017;12(12):e0190332. 33. de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J. Develop- 56. Serrano-Villar S, Sainz T, Lee SA, Hunt PW, Sinclair E, Shacklett BL, et al. HIV- ment of a WHO growth reference for school-aged children and adolescents. Bull infected individuals with low CD4/CD8 ratio despite effective antiretroviral ther- World Health Organ. 2007;85(9):660–7. apy exhibit altered T cell subsets, heightened CD8+ T cell activation, and increased 34. Huenecke S, Behl M, Fadler C, Zimmermann SY, Bochennek K, Tramsen risk of non-AIDS morbidity and mortality. PLoS Pathog. 2014;10(5):e1004078. L, et al. Age-matched lymphocyte subpopulation reference values in childhood 57. Traisathit P, Urien S, Le Coeur S, Srirojana S, Akarathum N, Kanjanavanit S, and adolescence: application of exponential regression analysis. Eur J Haematol. et al. Impact of antiretroviral treatment on height evolution of HIV infected chil- 2008;80(6):532–9. dren. BMC Pediatr. 2019;19(1):287. 35. WHO. WHO case definitions of HIV for surveillance and revised clinical stag- 58. Kessler M, Kaul A, Santos-Malavé C, Borkowsky W, Kessler J, Shah B. Growth ing and immunological classification of HIV-related disease in adults and children. patterns in pubertal HIV-infected adolescents and their correlation with cytokines, Geneva: WHO; 2007. https://apps.who.int/iris/handle/10665/43699 IGF-1, IGFBP-1, and IGFBP-3. J Pediatr Endocrinol Metab. 2013;26(7-8):639–44. 36. Royston P, Altman DG. Regression using fractional polynomials of continuous 59. Warszawski J, Lechenadec J, Faye A, Dollfus C, Firtion G, Meyer L, et al. Long- covariates: parsimonious parametric modelling. Appl Statist. 1994;43(3):429–67. term nonprogression of HIV infection in children: evaluation of the ANRS prospec- 37. Baldwin SA, Imel ZE, Braithwaite SR, Atkins DC. Analyzing multiple outcomes tive French Pediatric Cohort. Clin Infect Dis. 2007;45(6):785–94. in clinical research using multivariate multilevel models. J Consult Clin Psychol. 60. WHO. Consolidated guidelines on the use of antiretroviral drugs for treat- 2014;82(5):920–30. ing and preventing HIV infection: recommendations for a public health approach. 38. Nagin DS, Jones BL, Passos VL, Tremblay RE. Group-based multi-trajectory Geneva: WHO; 2016. modeling. Stat Methods Med Res. 2018;27(7):2015–23. 61. Mark D, Geng E, Vorkoper S, Essajee S, Bloch K, Willis N, et al. Making imple- 39. Prentice AM, Ward KA, Goldberg GR, Jarjou LM, Moore SE, Fulford AJ, et al. mentation science work for children and adolescents living with HIV. J Acquir Critical windows for nutritional interventions against stunting. Am J Clin Nutr. Immune Defic Syndr. 2018;78(Suppl 1):S58–62. 2013;97(5):911–8. 62. Dalzini A, Petrara MR, Ballin G, Zanchetta M, Giaquinto C, De Rossi A. Bio- 40. Reinhardt K, Fanzo J. Addressing chronic malnutrition through multi-sectoral, logical aging and immune senescence in children with perinatally acquired HIV. J sustainable approaches: a review of the causes and consequences. Front Nutr. Immunol Res. 2020;2020:8041616. 2014;1:13. 63. UNAIDS. The Western and Central Africa catch-up plan: putting HIV treat- 41. Williams PL, Jesson J. Growth and pubertal development in HIV-infected ado- ment on the fast-track by 2018. Geneva: UNAIDS; 2017. https://www.unaids.org/ lescents. Curr Opin HIV AIDS. 2018;13(3):179–86. sites/default/files/media_asset/WCA-catch-up-plan_en.pdf 42. Bellavia A, Williams PL, DiMeglio LA, Hazra R, Abzug MJ, Patel K, et al. Delay in sexual maturation in perinatally HIV-infected youths is mediated by poor growth. AIDS. 2017;31(9):1333–41. 43. Szubert AJ, Musiime V, Bwakura-Dangarembizi M, Nahirya-Ntege P, Kekiti- S U P P O R T I N G I N F O R M AT I O N inwa A, Gibb DM, et al. Pubertal development in HIV-infected African children on first-line antiretroviral therapy. AIDS. 2015;29(5):609–18. Additional information may be found under the Supporting 44. Lewis J, Walker AS, Castro H, De Rossi A, Gibb DM, Giaquinto C, et al. Age and Information tab for this article: CD4 count at initiation of antiretroviral therapy in HIV-infected children: effects on Additional files: Additional information may be found under long-term T-cell reconstitution. J Infect Dis. 2020;205(4):548–56. 45. Guihot A, Dentone C, Assoumou L, Parizot C, Calin R, Seang S, et al. Residual the Supporting Information tab for this article. immune activation in combined antiretroviral therapy-treated patients with maxi- Table S1: Characteristics of study participants for the growth mally suppressed viremia. AIDS. 2016;30(2):327–30. and the CD4 analyses, by regions. 46. Dabee S, Mkhize NN, Jaspan HB, Lewis D, Gumbi PP, Passmore JS. Initiation Table S2: Characteristics of study participants for the growth of antiretroviral therapy differentially influences genital and systemic immune acti- and the CD4 analyses, by regions. vation in HIV-infected women. AIDS Res Hum Retroviruses. 2020;36(10):821–30. 47. European Pregnancy and Paediatric HIV Cohort Collaboration (EPPICC) Figure S1: Adjusted estimated mean height-for-age Z-scores study group. Prevalence and clinical outcomes of poor immune response despite for adolescents living with perinatally acquired HIV, stratified virologically suppressive antiretroviral therapy among children and adolescents by age at ART initiation, according to sex and regions. with human immunodeficiency virus in Europe and Thailand: cohort study. Clin Figure S2: Mean CD4 evolution during adolescence for ado- Infect Dis. 2020;70(3):404–15. 48. Paiardini M, Müller-Trutwin M. HIV-associated chronic immune activation. lescents living with perinatally acquired HIV, stratified by sex, Immunol Rev. 2013;254(1):78–101. expressed in cells/mm3 (left) and in age-adjusted ratio (right). 49. Kroeze S, Ondoa P, Kityo CM, Siwale M, Akanmu S, Wellington M, et al. Subop- Figure S3: Mean CD4 evolution during adolescence for ado- timal immune recovery during antiretroviral therapy with sustained HIV suppres- lescents living with perinatally acquired HIV, stratified by sion in sub-Saharan Africa. AIDS. 2018;32(8):1043–51. immunodeficiency at ART initiation and at age 10, as well as 50. Caby F, Guihot A, Lambert-Niclot S, Guiguet M, Boutolleau D, Agher R, et al. Determinants of a low CD4/CD8 ratio in HIV-1-infected individuals despite long- by age at ART initiation and year of birth. term viral suppression. Clin Infect Dis. 2016;62(10):1297–303. Supplementary results: Joint modelling of growth and 51. Davy-Mendez T, Napravnik S, Zakharova O, Kuruc J, Gay C, Hicks CB, et al. immunology, illustrations for males, for two regions (North Acute HIV infection and CD4/CD8 ratio normalization after antiretroviral therapy America and Europe, East and Southern Africa excluding initiation. J Acquir Immune Defic Syndr. 2018;79(4):510–8. 52. Guillén S, Prieto L, Jiménez de Ory S, González-Tomé MI, Rojo P, Navarro ML, Botswana and South Africa) et al. Prognostic factors of a lower CD4/CD8 ratio in long term viral suppression Data S1: Full list of acknowledgments for the CIPHER Global HIV infected children. PLoS One. 2019;14(8):e0220552. Cohort Collaboration 15