Thanyawee Puthanakit

Federated learning (FL) is a method used for training artificial intelligence models with data from multiple sources while maintaining data anonymity, thus removing many barriers to data sharing. Here we used data from 20 institutes across the globe to train a FL model, called EXAM (electronic medical record (EMR) chest X-ray AI model), that predicts the future oxygen requirements of symptomatic patients with COVID-19 using inputs of vital signs, laboratory data and chest X-rays. EXAM achieved an average area under the curve (AUC) >0.92 for predicting outcomes at 24 and 72 h from the time of initial presentation to the emergency room, and it provided 16% improvement in average AUC measured across all participating sites and an average increase in generalizability of 38% when compared with models trained at a single site using that site's data. For prediction of mechanical ventilation treatment or death at 24 h at the largest independent test site, EXAM achieved a sensitivity of 0.950 and specificity of 0.882. In this study, FL facilitated rapid data science collaboration without data exchange and generated a model that generalized across heterogeneous, unharmonized datasets for prediction of clinical outcomes in patients with COVID-19, setting the stage for the broader use of FL in healthcare.

BACKGROUND: We previously reported similar AIDS-free survival at 3 years in children who were >1 year old initiating antiretroviral therapy (ART) and randomized to early versus deferred ART in the Pediatric Randomized to Early versus Deferred Initiation in Cambodia and Thailand (PREDICT) study. We now report neurodevelopmental outcomes. METHODS: Two hundred eighty-four HIV-infected Thai and Cambodian children aged 1-12 years with CD4 counts between 15% and 24% and no AIDS-defining illness were randomized to initiate ART at enrollment ("early," n = 139) or when CD4 count became <15% or a Centers for Disease Control (CDC) category C event developed ("deferred," n = 145). All underwent age-appropriate neurodevelopment testing including Beery Visual Motor Integration, Purdue Pegboard, Color Trails and Child Behavioral Checklist. Thai children (n = 170) also completed Wechsler Intelligence Scale (intelligence quotient) and Stanford Binet Memory test. We compared week 144 measures by randomized group and to HIV-uninfected children (n = 319). RESULTS: At week 144, the median age was 9 years and 69 (48%) of the deferred arm children had initiated ART. The early arm had a higher CD4 (33% versus 24%, P < 0.001) and a greater percentage of children with viral suppression (91% versus 40%, P < 0.001). Neurodevelopmental scores did not differ by arm, and there were no differences in changes between arms across repeated assessments in time-varying multivariate models. HIV-infected children performed worse than uninfected children on intelligence quotient, Beery Visual Motor Integration, Binet memory and Child Behavioral Checklist. CONCLUSIONS: In HIV-infected children surviving beyond 1 year of age without ART, neurodevelopmental outcomes were similar with ART initiation at CD4 15%-24% versus <15%, but both groups performed worse than HIV-uninfected children. The window of opportunity for a positive effect of ART initiation on neurodevelopment may remain in infancy.

BACKGROUND: Children with human immunodeficiency virus type 1 (HIV-1) infection have limited options for effective antiretroviral treatment (ART). METHODS: We conducted an open-label, randomized, noninferiority trial comparing three-drug ART based on the HIV integrase inhibitor dolutegravir with standard care (non-dolutegravir-based ART) in children and adolescents starting first- or second-line ART. The primary end point was the proportion of participants with virologic or clinical treatment failure by 96 weeks, as estimated by the Kaplan-Meier method. Safety was assessed. RESULTS: From September 2016 through June 2018, a total of 707 children and adolescents who weighed at least 14 kg were randomly assigned to receive dolutegravir-based ART (350 participants) or standard care (357). The median age was 12.2 years (range, 2.9 to 18.0), the median weight was 30.7 kg (range, 14.0 to 85.0), and 49% of the participants were girls. By design, 311 participants (44%) started first-line ART (with 92% of those in the standard-care group receiving efavirenz-based ART), and 396 (56%) started second-line ART (with 98% of those in the standard-care group receiving boosted protease inhibitor-based ART). The median follow-up was 142 weeks. By 96 weeks, 47 participants in the dolutegravir group and 75 in the standard-care group had treatment failure (estimated probability, 0.14 vs. 0.22; difference, -0.08; 95% confidence interval, -0.14 to -0.03; P = 0.004). Treatment effects were similar with first- and second-line therapies (P = 0.16 for heterogeneity). A total of 35 participants in the dolutegravir group and 40 in the standard-care group had at least one serious adverse event (P = 0.53), and 73 and 86, respectively, had at least one adverse event of grade 3 or higher (P = 0.24). At least one ART-modifying adverse event occurred in 5 participants in the dolutegravir group and in 17 in the standard-care group (P = 0.01). CONCLUSIONS: In this trial involving children and adolescents with HIV-1 infection who were starting first- or second-line treatment, dolutegravir-based ART was superior to standard care. (Funded by ViiV Healthcare; ODYSSEY ClinicalTrials.gov number, NCT02259127; EUDRACT number, 2014-002632-14; and ISRCTN number, ISRCTN91737921.).

INTRODUCTION: Strategies are needed to curb the increasing HIV incidence in young men who have sex with men (YMSM) and transgender women (YTGW) worldwide. We assessed the impact of youth-friendly services (YFS) and a mobile phone application (app) on adherence to pre-exposure prophylaxis (PrEP) in YMSM and YTGW in Thailand. METHODS: A randomized control trial was conducted in YMSM and YTGW aged 15 to 19 years. Participants were provided daily oral tenofovir disoproxil fumerate/emtricitabine (TDF/FTC), condoms and randomized to receive either YFS or YFS plus a PrEP app (YFS + APP), whose features included self-assessment of HIV acquisition risk, point rewards and reminders for PrEP and clinic appointments. Clinic visits occurred at zero, one, three and six months and telephone contact at two, four and five months. HIV testing was performed at every clinic visit. PrEP adherence was evaluated with intracellular tenofovir diphosphate (TFV-DP) concentrations in dried blood spot (DBS) samples at months 3 and 6. The primary endpoint assessed was "PrEP adherence" defined as TFV-DP DBS concentrations ≥700 fmol/punch (equivalent to ≥4 doses of TDF/week). RESULTS: Between March 2018 and June 2019, 489 adolescents were screened at three centres in Bangkok. Twenty-seven (6%) adolescents tested positive for HIV and 200 (41%) adolescents participated in the study. Of these, 147 were YMSM (74%) and 53 YTGW (26%). At baseline, median age was 18 years (IQR 17 to 19), 66% reported inconsistent condom use in the past month. Sexually transmitted infection prevalence was 23%. Retention at six months was 73%. In the YFS + APP arm, median app use duration was three months (IQR 1 to 5). PrEP adherence at month 3 was 51% in YFS and 54% in YFS + APP (p-value 0.64) and at month 6 was 44% in YFS and 49% in YFS + APP (p-value 0.54). No HIV seroconversions occurred during 75 person years of follow-up. CONCLUSIONS: Youth-friendly PrEP services enabled good adherence among half of adolescent PrEP users. However, the mobile phone application tested did not provide additional PrEP adherence benefit in this randomized trial. Adolescent risk behaviours are dynamic and require adaptive programmes that focus on "prevention-effective adherence."

Forty years ago, the first adult HIV cases were published, with infant cases following within a year [1]. As a few of these then-babies approach their 40th birthdays, both their growth and science’s growth tell dramatic stories. Antiretroviral therapy (ART) transformed HIV from a deadly infection into a chronic disease. Just as miraculous, an AIDS-free generation became imaginable, using ART to prevent >95% of perinatal transmission. While these advances in HIV prevention and treatment deserve celebration, attention should be devoted to remaining hurdles – such as behavioural, social viral suppression and drug resistance challenges – that must still be overcome to ensure successful life-long outcomes for the global population of children and adolescents who have grown up with HIV (CAWH). The global HIV impact for CAWH continues to be enormous: in 2020, 1.8 million children under 14 live with HIV, and every day 400 still acquire HIV and 270 die from it. Although ART access has expanded, only 53% of CAWH were receiving ART in 2019. Many countries do not screen mothers or infants for HIV, which leads to perinatal transmission, late childhood diagnosis, and deaths. Two million adolescents live with HIV globally, approximately 80% in sub-Saharan Africa, for whom HIV remains the top cause of death. Older adolescents (15 to 19 years) are the only age group in which HIV-related deaths are not decreasing. In the face of these young deaths, current care models clearly are not working; of adolescents 10 to 19 years with HIV, only 43% engage in care, 31% are retained in care and a dismal 30% are virally suppressed [2]. Added challenges of COVID-19 pandemic-related disruptions on HIV testing and care remain to be fully quantified and understood for CAWH. Behavioural and social challenges evolve throughout growth and development, from infants who spit out poorly palatable liquid medications like lopinavir/ritonavir; to school-aged CAWH who question why they take medicines; to adolescents living in stigmatizing environments and struggling to establish autonomy and fit in. Such challenges, on top of the difficulty accepting an HIV diagnosis, impact ART adherence and clinic engagement, with risks for worsening morbidity and mortality. Both adherence support and psychosocial counselling are less accessible to families in low-and-middle-income-countries (LMIC) [3]. Parents and caregivers struggle with disclosure to their children, often waiting until adolescence to explain how and why they have HIV out of concerns for their reaction and fears that they may disclose the family HIV status to others. Disclosure delays can hurt adherence, as can wrestling with diagnosis acceptance and avoiding sensitive disclosures to friends and sexual partners [4]. CAWH may have behavioural and mental health challenges, associated with more risky sexual behaviours, substance use and non-adherence [5]. CAWH are also at substantial risk of stigma and its deleterious effects. Challenges can be distinct from adults, like bullying, discrimination at or exclusion from schools, growing up with caregivers’ fears of stigma, internalizing stigma, violence and prosecution tied to sexual/gender orientation [6], and exacerbations of poverty, malnutrition and inequitable healthcare access [7]. While CAWH-focused HIV services may offer individualized behavioural and social support, transitioning into adult care often negatively affects older youth as they “age out” of more child-friendly paediatric health systems into larger clinics that lack such support services. This transition period can be a time of increased risks of poor adherence and viral outcomes [8]. Multiple factors contribute to challenges with viral suppression in paediatric HIV care, and CAWH do not achieve similar suppression levels as adults [9]. First, overall global roll-out of routine viral load testing has been slow, complicated by reliance on plasma-based specimens, and, particularly for CAWH, limitations such as phlebotomy expertise and insufficient evidence for optimal scheduling of routine monitoring. Second, the limited number of ART formulations appropriate for and available to children make regimen modifications difficult, especially in LMIC. Only a quarter of approved formulations have dosing guidelines and are approved for children <2 years old, and may still be unpalatable or complex to administer [10]. Third, CAWH are less likely to participate in new therapeutic clinical trials, and regulatory approvals and guidelines are typically slow to extend to paediatric populations. The amalgamation of limited VL monitoring, fewer ART options and poor adherence support contributes to undetected treatment failures, delayed ART switches and drug resistance accumulation [11]. Evaluating drug resistance across the lifespans of CAWH is, therefore, a particular concern. However, resistance testing as part of clinical care remains limited in LMIC, and its true burden is largely unknown [12]. The few available studies demonstrate extensive resistance to ART regimens commonly used in CAWH, whereas longitudinal outcomes data are limited to non-existent [13, 14]. Even as newer medications with high barriers to resistance are rolled out for CAWH (e.g. second-generation integrase strand transfer inhibitors), scale-up challenges in the context of implementation, prior relevant ART exposures, and limited availability of formulations may still negatively affect the trajectory and impact of resistance [15]. Largely unknown aspects of resistance (e.g. archived mutations, minority resistance variants, co-occurrence of mutations within genomes, alternative resistance mechanisms) may accompany CAWH into adulthood and become difficult to overcome in the future. An AIDS-free generation should be possible, but we are not there yet. HIV remains a burden, not only in LMIC, but also in resource-rich settings like the United States, where systemic impacts of racism on healthcare access, poverty, incarceration, stigma, mental health, and substance use are exacerbated by disparities in perinatal and adolescent HIV transmission and treatment outcomes. As we honour our gains in HIV care delivery over the past 40 years, we must address the specific, nuanced and long-term challenges of living with HIV for CAWH, and realize that they cannot be treated as “typical adults with HIV” as they mature and grow older. Until we transform care systems, treatment regimens, adherence and mental health support, viral and drug resistance monitoring and global policy, this generation will remain vulnerable to high HIV transmission, morbidity and mortality. Setting priorities to support CAWH requires greater attention to the long-term implications of the unique challenges they have faced over the first decades of the HIV pandemic so that they can successfully reach the next chapters of their lives. The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. All five authors (RV, NR, WN, TP and RK) meet the criteria for authorship and have made substantial contributions to various facets of the manuscript. This manuscript was initially conceptualized and written by RV and RK. All authors reviewed, edited and contributed significantly to writing subsequent versions of the manuscript. All authors read and approved the final manuscript. We acknowledge the incredible privilege of caring for families living with HIV, and we are thankful for the tremendous contributions of our patients, clinical colleagues, community members, activists and researchers. We appreciate the daily gift of being guided by both data and insights from children, adolescents and young adults living with HIV. This work was supported by the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health with support to RV and RK through NIH R01 AI147333 and R01 AI120792, with RK’s effort also supported through NIH K24 AI134359 and P30 AI042853 awards. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Allergy and Infectious Diseases or the National Institutes of Health.