Risk factors associated with adverse metabolic health in youth with perinatally acquired HIV living in the United Kingdom Authors: Caroline Foster, Alexandra Blenkinsop, Merle Henderson, Hermione Lyall, Sarah Fidler Published: 2026 ABSTRACT Objective:Non-AIDS-related morbidity and mortality in people with HIV are associated with adverse metabolic health, influenced by both traditional-related and HIV-related risk factors. There is a paucity of data for youth with perinatal HIV (PHIV). We explored the relationship between markers of metabolic health and antiretroviral therapy (ART) in youth with PHIV.Design:Longitudinal observational cohort study; 26 months.Methods:Eighty-five youth enrolled in the ‘Bone Density in Youth living with PHIV’ (BONDY) underwent assessment of metabolic health including fasting biochemistry, BMI, total body dual energy x-ray absorptiometry scan and hepatic transient elastography.Results:Of 85 participants with PHIV, mean age 21.7 years, 58% were female and 82% black African. Median ART exposure was 15 years, median CD4+ cell count 623 cells/μl at enrolment with 82% viral load less than 200 copies/ml. Median weight gain over 26 months was 3 kg, with BMI category overweight/obese increasing from 37 to 50% with 48% having dyslipidaemia. Metabolic syndrome criteria were fulfilled in 7%, with 13% having hypertension, 18% hepatic steatosis, and 7% fibrosis on transient elastography. Bayesian regression analyses demonstrated no association of integrase strand transfer inhibitor (INSTI) and/or tenofovir alafenamide (TAF) use, and sex at birth or prior CDC-C diagnoses or current HIV viraemia on metabolic outcomes.Conclusion:While adverse metabolic outcomes are common in this youth cohort with PHIV, no association was observed with INSTI and/or TAF use. Given the relatively young age of this cohort, preventative interventions targeting traditional metabolic risk factors are required to avoid comorbidities in later life. Introduction Metabolic syndrome (MetS), characterized by central obesity, hypertension and impaired lipid and glucose metabolism, is a predictor of future cardiovascular risk and type 2 diabetes (T2DM). Increased rates of MetS are reported in adults with horizontally acquired HIV (HHIV); associated with traditional factors including obesity, alcohol, smoking, sedentary lifestyle and genetic factors – and HIV-related risk factors including viraemia and some antiretroviral regimens [1]. People born with perinatally acquired HIV (PHIV) experience lifelong exposure to HIV-associated inflammation and increasingly antiretroviral therapy (ART), with the oldest entering their fifth decade of life; approaching the age clinical cardiovascular events and MetS-associated T2DM appear in the general population [2–5]. Emerging data suggests increased rates of MetS and cardiovascular (CVS) risk in PHIV. Cohorts from the United States report incidence rates of T2DM of 19%, dyslipidaemia 50% and hypertension 22% by 30 years of age [6]. In this setting, youth with PHIV (median age 26 years) had higher rates of MetS (13 vs. 4%), than aged/ethnically matched HIV-negative youth, and higher rates of truncal/android fat distribution; a pattern associated with increased CVS events [7]. While two-thirds were overweight/obese, rates were comparable to youth without HIV (47 vs. 51% overweight, 20 vs. 17% obese). In settings with lower rates of obesity, youth with PHIV remain at increased risk of MetS (France; 13.2% males, 10.4% females with PHIV vs. 10.6% and 1.7% youth without HIV and in 10.2% Thai PHIV youth) [8,9]. Conversely data from South Africa in youth, median age 18 years, suggested lower rates of MetS in those with PHIV (12%) when compared to youth with HHIV (34%) and HIV negative youth (23%) but with higher rates of dyslipidaemia in youth with PHIV without obesity and highlights the considerable geographic heterogeneity in obesity and metabolic health [10]. Obesity-driven metabolic dysfunction-associated steatotic liver disease (MASLD) is the main driver of liver fibrosis in adults with HHIV [11]. Transient elastography (FibroScan)-controlled attenuation parameter (CAP) scores have good correlation with steatosis in adults with HHIV and in youth without HIV [12,13]. Early data in adolescents with PHIV suggests increased rates of steatosis compared to HIV-negative youth, only partially explained by metabolic factors [14,15]. The potential metabolic impact of ART includes dyslipidaemia with boosted protease inhibitors (bPI), and weight gain with integrase inhibitors (INSTI) and tenofovir alafenamide (TAF) [4,16–18]. Data for PHIV is emerging; 70% of Thai adolescents receiving bPIs had dyslipidaemia, with tenofovir disoproxil fumarate (TDF) a protective factor [19]. The bPI use remains high for treatment-experienced youth with extensive drug resistance [20,21]. Studies of INSTI-related weight change in PHIV vary from no association to weight gain in young women with increased truncal fat deposition [4,22–25]. In a meta-analysis of four paediatric TAF studies, weight-for-age z scores increased by 0.25 at 48 weeks; however, data on the impact of INSTIs with TAF in youth is lacking [26]. The Reprieve Study demonstrated a 35% reduction in major CVS events following statin use in adults with HHIV (median 50 years) and a low/moderate CVS risk, acknowledging that cardiovascular risk scores underestimate risk in those with HIV and in younger populations [27,28]. Interventions to reduce risk factors for non-communicable diseases are required from early adult life, prior to established pathology [29]. In this study, we examined the relationship between markers of metabolic health, hepatic steatosis, HIV clinical status and ART exposure in PHIV youth in the United Kingdom, exploring the hypothesis that weight gain associated with INSTIs in conjunction with TAF would impact metabolic health. Methods Study population The ‘Bone Density in Youth living with perinatally acquired HIV’ (BONDY) study has been described in full previously [30]. Participants aged 15–24 years were eligible for the longitudinal study arm with a second assessment after 24 months. At this visit, they were invited to enrol in the sub-study (BONDY-Met) assessing metabolic health. Study procedures BONDY-Met Demographics, ART history, past CDC-C diagnoses, length of viral suppression, latest CD4+ cell count, HIV viral load, a family history of diabetes, cardiovascular and liver disease, concomitant medication and substance use were recorded [30]. During the study period, the NHS commissioned TAF for those younger than 25 years, with a TDF to TAF switch offered to participants and all ART changes documented [31]. Height, weight, Tanner staging, blood pressure (BP) and waist circumference were assessed. BMI was calculated as weight/height (kg/m2) and as an age-adjusted z score if 15-19 years and classified as overweight BMI greater than than 25–29 or obese BMI greater than 30. Metabolic biochemistry Participants fasted for 12 hours, and glucose, lipid profile, glycosylated haemoglobin (HbA1c), C-peptide, insulin and liver enzymes measured. Homeostatic model for assessment of insulin resistance (HOMA-IR) was calculated using insulin (μU/ml) × glucose (mmol/l)/22.5 [32]. Total body dual energy X-ray absorptiometry (DXA) scans measured total body fat, android (AFM) and gynoid fat mass (GFM), using Lunar Prodigy or iDXA (GE Healthcare) scanners [30]. The android : gynoid ratio (AGR) was calculated using the fat percentage in regions A and G. Transient elastography (FibroScan) Assessed CAP (dB/m) with correlates of steatosis graded as [33].1.238–260 dB/m S1 = 11 33% liver fatty change (LFC)2.260–290 dB/m S2 = 34–66% LFC3.>290 dB/m S3 = 67+% LFC 1.238–260 dB/m S1 = 11 33% liver fatty change (LFC) 238–260 dB/m S1 = 11 33% liver fatty change (LFC) 2.260–290 dB/m S2 = 34–66% LFC 260–290 dB/m S2 = 34–66% LFC 3.>290 dB/m S3 = 67+% LFC >290 dB/m S3 = 67+% LFC Fibrosis scoring (F) assessed by E score (kPa) and graded as:1.E2-7 – F0-1 no/mild fibrosis2.E7-10 – F2 moderate3.E10-14 – F3 severe4.E14+ – F4 cirrhosis 1.E2-7 – F0-1 no/mild fibrosis E2-7 – F0-1 no/mild fibrosis 2.E7-10 – F2 moderate E7-10 – F2 moderate 3.E10-14 – F3 severe E10-14 – F3 severe 4.E14+ – F4 cirrhosis E14+ – F4 cirrhosis Metabolic syndrome Rates were defined using a consensus definition adjusted for age and ethnicity [34]. Three of five criteria led to a diagnosis of MetS.1.Waist circumference >102 cm men, >88 cm women (16+ years)2.Triglycerides ≥1.7 mmol/l3.HDL <1.04 mmol/l men, <1.29 mmol/l women4.Blood Pressure ≥130/85 mmHg5.Fasting glucose ≥5.6 mmol/l 1.Waist circumference >102 cm men, >88 cm women (16+ years) Waist circumference >102 cm men, >88 cm women (16+ years) 2.Triglycerides ≥1.7 mmol/l Triglycerides ≥1.7 mmol/l 3.HDL <1.04 mmol/l men, <1.29 mmol/l women HDL <1.04 mmol/l men, <1.29 mmol/l women 4.Blood Pressure ≥130/85 mmHg Blood Pressure ≥130/85 mmHg 5.Fasting glucose ≥5.6 mmol/l Fasting glucose ≥5.6 mmol/l Statistical analysis Participant characteristics were summarized using counts and proportions for categorical variables, means and standard deviations (SD) for normally distributed continuous variables, and medians with interquartile ranges (IQR) for skewed distributions. We examined six primary outcomes: change in BMI baseline to follow-up, and five metabolic outcomes at visit 2 (FibroScan CAP, HbA1c, HOMA-IR, MetS, and meeting two criteria for MetS). Associations of INSTI-based vs. other ART, TAF-containing vs. non-TAF-containing regimens, INSTI + TAF vs. other regimens and HIV parameters (CDC-C, HIV viraemia and CD4+ : CD8+ ratio) were estimated using Bayesian generalized linear models with Gaussian link functions for continuous outcomes and logit link functions for binary outcomes. All models were adjusted for age and sex, and models for BMI change also adjusted for baseline BMI. For each outcome, we quantified the evidence for an association by estimating the posterior probability that the corresponding regression coefficient exceeded 0 for continuous outcomes or 1 (on the odds ratio scale) for binary outcomes. The threshold for significance was set at 1% (one-sided) due to multiple hypothesis testing increasing the probability of a spurious finding, with probabilities greater than 99% interpreted as significant. Potential interactions between treatment and sex were assessed. Ethical considerations All participants provided written consent with ethical approval from the London-Stanmore Research Ethics Committee (REC) (IRAS 18/LO/1474). Results Of 100 youth eligible for the BONDY study longitudinal arm (perinatal HIV age 15–24 at study entry, on ART), 85 enrolled in BONDY-Met: 49 (58%) were female, 70 (82%) of black ethnicity with a mean age of 21.7 years (SD 2.7). Reasons for not enrolling included (n = 15): transfer of care 5, personal choice/did not attend 5, unable to consent: pregnant 1, psychosis 1, central nervous system infection 1, travel 1, incarceration 1). The median follow-up was 26 months (IQR 25, 28) from study entry with a median weight gain of 3 kg [IQR −1, 7]. At latest follow-up, median CD4+ cell count was 623 cells/μl (IQR 467, 770), with 68 (80%) having a viral load less than 200 copies/ml. Mean ART duration was 15 years with 39 (46%) currently on INSTI-based ART, 32 (38%) on bPI [darunavir 27, atazanavir 5 (1 unboosted)] with 43 (51%) on TAF and 14 (16%) on INSTI +TAF (Table 1). Half had a prior CDC-C diagnosis, 22% were current or former smokers, 18% had a family history of cardiovascular disease, 26% a family history of diabetes mellitus and 32% drank alcohol. All were postpubertal and none had diabetes mellitus or were receiving lipid-lowering or antihypertensive agents. Median weight gain over 26 months was 3 kg, with BMI category overweight/obese increasing from 37 to 50% with 28 (33%) having an A/G ratio within the third tertile. Forty-one (48%) had MetS defined dyslipidaemia and 11 (13%) had hypertension [39]. Six (7%) fulfilled MetS criteria with a further 13/85 (15%) having two criteria. Insulin resistance, (HOMA-IR ≥3) was evident in 19% with none having an HbA1c above 48 mmol/l (Table 1). On transient elastography, 15 (18%) had evidence of steatosis (≥S2), 7 (8%) had evidence of fibrosis (≥F2), and 8% had liver transaminases above the upper limit of normal. Mean BMI was 24.7 (5.1) with no difference in change from baseline by TAF (n = 43) vs. non-TAF/TDF (n = 40). Table 1Participant characteristics.CharacteristicTotal (n = 85)Female (n = 49)Male (n = 36) Age (years)Mean (SD)21.7 (2.7)21.6 (2.8)21.9 (2.7) EthnicityBlack/African/Caribbean/Black British70 (82%)39 (80%)31 (86%)Other15 (18%)10 (20%)5 (14%) ART duration (years)Mean (SD)15.0 (6.1)14.3 (6.8)15.8 (5.2) ART anchorINSTI-based39 (46%)28 (57%)11 (31%)PI-based32 (38%)14 (29%)18 (50%)NNRTI-based14 (16%)7 (14%)7 (19%) TAF/TDFTAF43 (51%)28 (57%)15 (42%)TDF2 (2%)0 (0%)2 (6%)Non-TAF/TDF40 (47%)21 (43%)19 (53%) CD4+CD4+ (cells/μl) (median [IQR])623 [467, 770]650 [524, 791]583 [436, 736] CD4+: CD8+ ratioMedian [IQR]0.9 [0.6–1.3]0.9 [0.6–1.4]0.9 [0.6–1.1] Viral load (copies/ml)<2057 (67%)30 (61%)27 (75%)20–19911 (13%)8 (16%)3 (8%)≥20016 (19%)10 (20%)6 (17%) Previous CDC-CYes44 (52%)26 (53%)18 (50%) Smoking historySmoking never66 (78%)39 (80%)27 (75%)Smoking current/former19 (22%)10 (20%)9 (25%) Any alcohol27 (32%)15 (31%)12 (33%) Family history of CVD15 (18%)9 (18%)6 (17%) Family history of DM22 (26%)11 (22%)11 (31%)Metabolic parameters BMI (baseline)aUnderweight1 (1%)0 (0%)1 (3%)Normal weight53 (62%)27 (55%)26 (72%)Overweight22 (26%)17 (35%)5 (14%)Obese9 (11%)5 (10%)4 (11%) BMI (follow-up)aUnderweight2 (2%)0 (0%)2 (6%)Normal weight40 (47%)21 (43%)19 (53%)Overweight24 (28%)16 (33%)8 (22%)Obese19 (22%)12 (24%)7 (19%) Prevalence of metabolic syndrome6 (7%)5 (10%)1 (3%) Prevalence of dyslipidaemia41 (48%)28 (57%)13 (36%) Prevalence of hypertension11 (13%)5 (10%)6 (17%) Waist circumference (cm)Mean (SD)80.6 (14.4)80.4 (13.4)80.8 (16.0) SBPMean (SD)120.3 (12.5)116.5 (11.0)125.6 (12.6) DBPMean (SD)75.8 (10.5)75.5 (9.5)76.1 (11.8) TriglyceridesMean (SD)0.8 (0.5)0.8 (0.4)0.9 (0.6)≥1.7 mmol/l6 (7%)5 (10%)1 (3%) Total cholesterolMean (SD)4.2 (0.9)4.1 (0.9)4.3 (1.0)≥5.18 mmol/l14 (16%)7 (14%)7 (19%) HDL-cholesterolMean (SD)1.4 (0.4)1.4 (0.4)1.3 (0.4)<1.29 (M), <1.04 (F)31 (36%)24 (49%)7 (19%) LDL-cholesterolMean (SD)2.6 (0.7)2.5 (0.6)2.7 (0.7)≥3.4 mmol/l11 (13%)5 (10%)6 (17%) GlucoseMean (SD)4.5 (0.5)4.5 (0.4)4.6 (0.5)>=5.6 mmol/l1 (1%)1 (2%)0 (0%) HOMA-IRMean (SD)2.4 (2.0)2.5 (2.2)2.2 (1.6)≥3.016 (19%)9 (19%)7 (19%) HbA1cMean (SD)33.0 (4.5)32.2 (4.1)34.0 (4.8)>48 mmol/l0 (0%)0 (0%)0 (0%) C-peptideMean (SD)589.6 (333.8)588.7 (328.5)590.8 (346.6)>1500 pmol/l2 (3%)1 (2%)1 (3%) InsulinMean (SD)11.7 (8.7)12.1 (9.4)11.3 (7.6)>15 μU/ml14 (18%)7 (16%)7 (22%) ALTMean (SD)25.6 (21.7)18.5 (6.6)35.3 (30.1)>45 (M), >34 (F)7 (8%)2 (4%)5 (14%)Total body dual energy X-ray absorptiometry Total body fat (%)Mean (SD)33.9 (10.8)39.0 (7.5)26.5 (10.8) Android (A) fat (%)Mean (SD)34.0 (14.4)38.7 (11.9)27.4 (15.1) Gynoid (G) fat (%)Mean (SD)37.1 (12.6)43.7 (7.7)27.9 (12.4) A/G ratioaMean (SD)0.9 (0.2)0.9 (0.2)0.9 (0.2)Tertile 129 (34.5%)18 (36.7%)11 (31.4%)Tertile 227 (32.1%)16 (32.7%)11 (31.4%)Tertile 328 (33.3%)15 (30.6%)13 (37.1%)Transient elastography (FibroScan) CAP (dBM)Mean (SD)198.9 (52.9)196.8 (49.9)201.8 (57.4) SteatosisS1 (238–260)4 (4.7%)3 (6.1%)1 (2.8%)S2 (260–290)5 (5.9%)3 (6.1%)2 (5.6%)S3 (290–400)6 (7.1%)2 (4.1%)4 (11.1%) E-score (kPA)Mean (SD)5.3 (3.6)4.9 (2.6)5.9 (4.7)2–778 (91.8%)45 (91.8%)33 (91.7%)7–104 (4.7%)3 (6.1%)1 (2.8%)10–140 (0.0%)0 (0.0%)0 (0.0%)≥143 (3.5%)1 (2.0%)2 (5.6%)ART, antiretroviral therapy; ALT, alanine aminotransferase; BP, blood pressure; BMI, body mass indexa z scores aged 15–19 years; CAP, controlled attenuation parameter; CDC-C, Centre for Disease Control category C AIDS diagnostic condition; CVD, cardiovascular disease; DM, diabetes mellitus; E-score, elastography score; F, female; HbA1c, glycosylated haemoglobin; HDL, high-density lipoprotein; INSTI, integrase strand transfer inhibitor; IQR, interquartile range; LDL, low-density lipoprotein; M, male; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; SD, standard deviation; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumerate.aCentral fat deposition assessed by DEXA as an android (abdominal)/gynoid (hip) ratio of ≥0.8 in women and ≥1.0 in men has been associated with an increased risk of adverse metabolic and cardiovascular outcomes in youth and adult populations [34]. Participant characteristics. ART, antiretroviral therapy; ALT, alanine aminotransferase; BP, blood pressure; BMI, body mass indexa z scores aged 15–19 years; CAP, controlled attenuation parameter; CDC-C, Centre for Disease Control category C AIDS diagnostic condition; CVD, cardiovascular disease; DM, diabetes mellitus; E-score, elastography score; F, female; HbA1c, glycosylated haemoglobin; HDL, high-density lipoprotein; INSTI, integrase strand transfer inhibitor; IQR, interquartile range; LDL, low-density lipoprotein; M, male; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; SD, standard deviation; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumerate. Central fat deposition assessed by DEXA as an android (abdominal)/gynoid (hip) ratio of ≥0.8 in women and ≥1.0 in men has been associated with an increased risk of adverse metabolic and cardiovascular outcomes in youth and adult populations [34]. Bayesian regression analyses demonstrated no association of INSTI and/or TAF containing ART on metabolic outcomes including, BMI change, MetS, hepatic steatosis, HbA1c and HOMA-IR (Table 2). Further analyses showed no evidence of interactions between sex (male vs. female at birth) and HIV parameters (CDC-C, current viraemia viral load ≥200 and CD4+ : CD8+ ratio <1) or INSTI and/or TAF-containing regimens on metabolic outcomes, adjusting for age group at study entry (Supplementary Table 1 and Figure 1). Table 2Estimated effect of INSTI-based antiretroviral thearpy regimens, tenofovir alafenamide-containing regimens and HIV clinical markers on metabolic outcomes, obtained from Bayesian generalized linear regression models (Gaussian link for continuous outcomes; logit link for binary outcomes).Outcome typeOutcomeRisk factorEstimate [95% credible interval]aPosterior probability of positive associationSignificantContinuous outcomesBMI changeINSTI0.16 [−1,1. 37]0.60NoTAF−0.72 [−1.94, 0.5]0.12NoINSTI+TAF−0.07 [−1.7, 1.58]0.46NoPrior CDC-C0.17 [−1.04, 1.35]0.61No**VL ≥2000.21 [−1.3, 1.7]0.60NoCD4+ : CD8+ ratio <1−0.47 [−1.71, 0.74]0.22NoFibroscan CAPINSTI−2.73 [−26.74, 21.02]0.41NoTAF−3.79 [−28.1, 19.73]0.38NoINSTI+TAF−17.46 [−47.65, 13.35]0.13NoPrior CDC-C−4.19 [−28.62, 19.19]0.36NoVL ≥200−0.95 [−30.35,29.45]0.46NoCD4+ : CD8+ ratio <11.06 [−23.05, 24.88]0.54NoHbA1cINSTI1.73 [−0.28, 3.76]0.96NoTAF−0.36 [−2.32, 1.65]0.35NoINSTI+TAF0.19 [−2.51, 2.77]0.56NoPrior CDC-C−0.34 [−2.28, 1.57]0.36NoVL ≥2001.4 [-1,3.83]0.87NoCD4+ : CD8+ ratio <10.97 [−0.88, 2.89]0.84NoHOMA-IRTAF0.3 [−0.63, 1.2]0.75NoINSTI0.41 [−0.45, 1.26]0.82NoINSTI+TAF0.47 [−0.71, 1.62]0.80NoPrior CDC-C0.29 [−0.54, 1.11]0.75NoVL ≥200−0.14 [−1.21, 0.88]0.40NoCD4+ : CD8+ ratio <10.19 [−0.72, 1.13]0.65NoBinary outcomesHepatic steatosisINSTI0.58 [0.13, 1.62]0.12NoTAF1.52 [0.37, 4.07]0.66NoINSTI+TAF0.42 [0.02, 1.68]0.09NoPrior CDC-C0.63 [0.17, 1.64]0.14NoVL ≥2003.39 [0.7, 9.73]0.92NoCD4+ : CD8+ ratio <11.07 [0.28, 2.92]0.42NoMetSINSTI3.97 [0.43, 16.26]0.85NoTAF2.81 [0.29, 12.23]0.71NoINSTI+TAF7.62 [0.83, 30.85]0.96NoPrior CDC-C15.02 [0.74, 82.02]0.95NoVL ≥2001.11 [0.03, 5.09]0.35NoCD4+ : CD8+ ratio <13.38 [0.4, 13.93]0.79No2 criteria of MetSINSTI1.14 [0.25, 3.23]0.44NoTAF1.35 [0.27, 4.06]0.53NoINSTI+TAF1.22 [0.11, 4.23]0.44NoPrior CDC-C3.86 [0.71, 13.9]0.93NoVL ≥2002.07 [0.29, 6.75]0.71NoCDC 4:8 ratio <13.91 [0.75, 13.6]0.93NoEach row corresponds to a distinct model, adjusting for age and sex, and models for change in BMI also adjusted for baseline BMI.aEstimated coefficients and 95% credible intervals on odds ratio scale for binary outcomes.ART, antiretroviral therapy; CAP, controlled attenuation parameter; CDC-C, Centre for Disease Control category C AIDS diagnostic condition; HbA1c, glycosylated haemoglobin; HOMA-IR, homeostatic model for assessment of insulin resistance; INSTI, integrase strand transfer inhibitor; MetS, metabolic syndrome; TAF, tenofovir alafenamide; **VL, most recent viral load. Estimated effect of INSTI-based antiretroviral thearpy regimens, tenofovir alafenamide-containing regimens and HIV clinical markers on metabolic outcomes, obtained from Bayesian generalized linear regression models (Gaussian link for continuous outcomes; logit link for binary outcomes). Each row corresponds to a distinct model, adjusting for age and sex, and models for change in BMI also adjusted for baseline BMI. Estimated coefficients and 95% credible intervals on odds ratio scale for binary outcomes. ART, antiretroviral therapy; CAP, controlled attenuation parameter; CDC-C, Centre for Disease Control category C AIDS diagnostic condition; HbA1c, glycosylated haemoglobin; HOMA-IR, homeostatic model for assessment of insulin resistance; INSTI, integrase strand transfer inhibitor; MetS, metabolic syndrome; TAF, tenofovir alafenamide; **VL, most recent viral load. Discussion In this cohort of youth with PHIV (mean age 21.7 years), rates of MetS were low (7%) although half were overweight or obese, with 48% having dyslipidaemia. Rates of MetS were comparable to PHIV cohorts in France and Thailand but lower than an older cohort in the Unites States (13% MetS; median age 30 years), with MetS prevalence increasing with age in the general population [6,8,9]. Rates of overweight/obesity in 16–24-year-olds in England are estimated at 37%, associated with female sex, black ethnicity and deprivation; risk factors common in this study population [36]. The median weight gain over 26 months was 3 kg, with BMI category overweight/obese moving from 37 to 50%. Obesity rates increase with age; however, recent general population data in England suggest those currently aged 18–24 years have the highest risk of BMI gain compared to older age groups [37]. Increasing youth weight gain is associated with environmental factors (calorie-rich processed food, increasingly sedentary lifestyle) and genetic hereditability results in both metabolic and psychological consequences [38]. While weight gain has been reported in association with INSTIs and TAF, we found no association with adverse metabolic outcomes, including BMI change, suggesting that addressing traditional risk factors is more important for metabolic health than switching ART. Current viraemia and past CDC-C diagnosis were not associated with adverse metabolic outcomes; however, maintaining viral suppression remains a priority for youth where rates of viral suppression are lower than in adults, with uncontrolled viraemia associated with multimorbidity and potential onward transmission [5]. Rates of MetS defined dyslipidaemia approached 50% in this cohort. The Reprieve study led to the recommendation of empirical statin use from 40 years in adults with HHIV [27]. However, there are no recommendations for statins in PHIV despite high rates of dyslipidaemia reported in both high-income and low-income settings (34–70%) and in the context of lifelong inflammation and increased cardiovascular risk [7,10,19,22–25,39]. With the metabolic impact of lifestyle interventions for youth limited and pharmacotherapy currently targeted to older adults with established morbidity, the optimal timing and modality of intervention strategies to prevent cardiovascular disease remains a research priority for this emerging adult population with PHIV [5,10,27,39]. The limitations of this study include the relatively short duration of follow-up, potential selection bias of those consenting to the sub study and variable rates of viral suppression during the study period potentially influencing the impact of ART, although reflecting real-world adherence patterns in this population and the challenges of maintaining viral suppression during this period of life [5]. Major limitations include the low participant number resulting in likely underpowering to study ART interactions and lack of age and ethnically matched HIV-negative youth. However, the lack of effect on outcomes by ART regimen (INSTI vs. other, TAF vs. non-TAF/TDF and INSTI with TAF vs. other) adds to the evidence base underlining the importance of addressing traditional risk factors for adverse metabolic health from early childhood and through into adult life, while maintaining sustained viral suppression. Acknowledgements The writing team would like to thank all the participants who took part in the BONDY-Met Study and the members of the BONDY study group Sara Ayres, Alice Walley, Rebecca Hall, Natalie Kirkhope, Fiona Brown, and Wilbert Ayap. Authors’ contributions: C.F. conceived the study, recruited participants, collated the data, and wrote the draft article. A.B. performed the statistical analyses and reviewed the article. H.L. recruited participants and reviewed the draft article. S.F. and M.H. recruited patients, reviewed the data and draft article. and are supported by Imperial College NIHR BRC. Funding: this study was funded by an Investigator led grant from Gilead Sciences (C.F.). Conflicts of interest There are no conflicts of interest. Supplementary Material Supplemental Digital Content