Publications

Authors: Ruggiero A, Pascucci G.R, Cotugno N, Domínguez-Rodríguez S, Rinaldi S, Tagarro A, Rojo P, Foster C, Bamford A, De Rossi A, Nastouli E, Klein N, Morrocchi E, Fatou B, Smolen K.K, Ozonoff A, Di Pastena M, Luzuriaga K, Steen H, Giaquinto C, Goulder P, Rossi P, Levy O, Pahwa S, Palma P, the EPIICAL Consortium

Published in: Frontiers in Immunology 

Authors: S Domínguez-Rodríguez, A Tagarro, C Foster, P Palma, N Cotugno, AD Rossi, A Dalzini, SG. Pahwa, E Nastouli, K Gärtner, AG Marcelin, P Rossi, C Giaquinto, P Rojo

Presented at: CROI 2022

Abstract

Subphenotypes have been identified in several heterogeneous diseases. Having a specific subphenotype often has therapeutic implications or impacts disease progression. In this study, we aimed to assess if children with HIV may show subphenotypes according to clinical, virological and immunological features.

We collected data from 40 HIV+ children included in a cross-sectional multicentric study (CARMA Study, EPIICAL Consortium). All children commenced ART <2 years, suppressed (viral load (VL) <50 copies/ml) within 12 months and remained suppressed for >5 years. Immunological and virological assays were performed at a median of 12 years after ART initiation. We collected clinical and sociodemographic data, baseline VL, CD4 and CD8 data, age at ART, HIV DNA reservoir size, cell-associated RNA (CA-RNA), ultrasensitive VL, CD4 subsets (T effector CD25+, activated memory cells, Treg cells ), humoral-specific HIV response (T-bet B cells), innate response (CD56dim NK cells, NKp46+, perforin), exhaustion markers (PD-1, PD-L1, DNAM), CD8 senesence, and biomarkers for T-lymphocyte thymic output (TREC) and endothelial activation (VCAM). To build the subphenotypes, the most informative variables were selected using an unsupervised penalty selection. Hierarchical clustering was performed using Pearson correlation as distance metric and Ward.D2 as clustering method. Internal validation was applied to select the best number of clusters.

Three subphenotypes were revealed (Cluster 1 n=18, 45%; Cluster 2 n=11, 27.5%; Cluster 3 n=11, 27.5%). Cluster 1 (best controllers) consisted of early ART-treated patients with high baseline %CD4, low reservoir size, low WB score, high TREC values, and low VCAM values. In contrast, Cluster 3 (worst controller) consisted of later ART-treated patients with low baseline %CD4, high reservoir size, low TREC values, high innate response, immunosenescence markers and VCAM. Cluster 2 (low-level viremia, altered immune response) consisted of early-treated patients with low-level (10 to 50 c/mL) VL, high reservoir size but low CA-RNA, higher Treg CD4 than in the other clusters, low TREC, weak innate response and lower levels of T-bet expression.

Three subphenotypes with decreasing levels of viral control and increasing levels of immune well-being were discovered. Response to different therapies may be different across the different clusters.

View poster

Authors: A Tagarro, S Domínguez-Rodríguez, L Kuhn, T Nhampossa, K Otwombe, AJv Rensburg, N Klein, MG Lain, AI. Maiga, C Brehin, C Giaquinto, P Rossi, P Rojo

Presented at: CROI 2022

Abstract

The real-world evolution of very early treated children born with HIV in high-prevalence settings during first years of life is unclear. We assessed the probability of death, progression to AIDS, viral load (VL) suppression, immunosuppression, and continuation in care of a cohort of early treated children born with HIV.

EARTH-EPIICAL Cohort is underway at 2 rural and 4 urban sites in Mozambique, Mali, and South Africa (SA). Infants with HIV who started ART in the first 3 months of life, are followed at 2, 6 and 12 weeks, and then 6-monthly for 4 years. Hereby, we provide the probability of death, progression, suppression and continuation in care during the first 2 years of life with multivariable cox regression models were used. Backward stepwise elimination was applied to reach the final multivariable model.

212 participants were enrolled and followed during a median time of 17 [6.8;27.5] months; 84 reached 2 years of follow-up. ART started at 34 [26;74] days of life, mostly 3TC+ABC+LPV/r (65%). Adherence was suboptimal (<90%) in 56% of visits. 23 patients (10.8%) died, at a median of 2.5 [0.6;6.8] months of age. At 2 years, probability (P) of death was 12% (CI95%,7 to 17), (P) of progression was 11% (CI95%,6 to 16%), (P) of continuation in care, 80% (CI95%,74 to 86%), (P) of VL suppression was 46% (CI95%, 0.34-0.49), and (P) of severe IS, 54% (CI95%, 44 to 62). Death occurred predominantly in the first 6 months (74%); mostly due to pneumonia (43%), malnutrition (13%) or diarrhea (8.7%). (P) of death was associated with baseline VL 2.19 (1.4-3.3), and suboptimal adherence 2.88 (1.18-7.2). (P) of progression was associated with baseline VL 1.71 (CI95%,1.14-2.58), and weight for age 0.53 (CI95%,0.39-0.71). (P) of lost to follow up also was associated with baseline VL (HR, 1.60 (CI95%,1.06-2.40) and weight for age (HR, 1.67 (CI95%,1.11-2.50). (P) of suppression associated with baseline VL (HR, 0.60 (CI95%,0.51-0.71) and CD4% (HR, 1.04, (CI95%,1.01-1.06), p=0.01. (P) of severe IS was associated with baseline VL, (HR, 1.4 (CI95% 1.15-1.69), baseline CD4% (HR, 0.91 (0.88-0.93), baseline weight for age (HR, 0.86 (CI95%, 0.76-0.99), and Mozambique (HR 1.88 (CI95%, 1.16-3.1).

Despite early treatment and a close follow-up, death, progression, and severe IS remain high in children born with HIV in Africa, especially in the 6 first months. Additional strategies are required to improve the care of this population.

View poster