CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

827 Contrasting CTL Impact in Distinct Phenotypes of Nonprogressing Pediatric HIV Ellen M. Leitman 1 ; Astrid Gall 2 ; Paul Kellam 2 ; Jacqui Brener 1 ; Azim Ansari 1 ; Christina F.Thobakgale 3 ; Philippa C. Matthews 1 ; Philip Goulder 1 1 Univ of Oxford, Oxford, UK; 2 Wellcome Trust Sanger Inst, Hinxton, UK; 3 HIV Pathogenesis Prog, Doris Duke Med Rsr Inst, Univ of KwaZulu-Natal, Durban, South Africa Background: ART-naïve non-progressing adult HIV infection is characterized by low viremia, strongly linked with expression of protective HLA I. In contrast, in the natural hosts of SIV infection such as sooty mangabeys non-progressing infection is associated with high viremia and a minimal role of MHC I. Non-progressing pediatric HIV infection is poorly characterized. Here we present two perinatally infected pediatric non-progressors (PNP) who adopt, respectively, an adult elite controller-like and a sooty mangabey-like phenotype in achieving long-termmaintenance of normal health and CD4 T-cell counts. Methods: Viral loads, CD4 counts and CTL responses in two South African vertically infected ART-naive children were monitored over the first decade of life. Full HIV genomes from children and their mothers were ultra-deep sequenced and interrogated for evidence of HLA-mediated CTL selection pressure. Results: ART-naïve child 133C maintained healthy CD4% (mean 32%) and absolute count (1,028c/mm 3 ) while gradually suppressing viremia to 520 copies/ml by 9yo. She expresses HLA-A*74 and B*81, protective alleles in adult C-clade infection. ART-naïve child 517C maintained normal CD4 counts (43%, 1,232c/mm 3 at 10yo), but in the setting of high viremia (median 35,500copies/ml) and without protective HLA. Both mothers expressed HLA-B*42:01/B*58:02. Deep sequencing showed that 133C selected escape mutations in five epitopes: HLA-B*81-Gag-TL9 (T186M), A*74-Gag-KR9 (K12N), B*81-Pol-SL10 (P159S), A*74-Pol-SR10 (R432K) and B*81-Nef-RM9 (L76V/I/T). These were mediated by child’s HLA- A*74/81 that were not shared with the mother. Minor variants arose as early as 3m. Variant-specific TL9-T186M responses came to replace wildtype TL9-specific CTL in the child. In contrast, 517C did not show evidence of escape in HLA-restricted epitopes. However, analysis of changes in genome diversity in non-epitopic regions showed more sequence evolution in 517C, particularly between 2.5 to 5yo. Conclusions: These data suggest that distinct immune strategies may be successful in non-progressing pediatric infection: adult elite controller-like CTL-mediated suppression of viremia via protective HLA alleles such as HLA-A*74 and B*81 as in child 133C; and a sooty mangabey-like approach in which high viremia is associated with low immune activation and a minimal role for CTL as in child 517C. Further studies are needed to define the mechanisms of low immune activation despite persistent high viremia in non-progressing children such as 517C. 828 Impact of HLA-B*81-Driven Escape Mutation L188F on VRC in Pediatric Slow Progression Ming-Han Tsai 1 ; Maximilian Muenchhoff 1 ; Emily Adland 1 ; Anna Carlqvist 1 ; David Cole 2 ; Andrew K. Sewell 2 ; Jonathan Carlson 3 ;Thumbi P. Ndungu 4 ; Philip Goulder 1 ; for the Professor Philip Goulder’s Group 1 Univ of Oxford, Oxford, UK; 2 Cardiff Univ, Cardiff, UK; 3 Microsoft Rsr, Redmond, WA, USA; 4 Univ of KwaZulu-Natal, Durban, South Africa Background: Two ART-naïve HIV-infected children were identified within the same South African family, who were infected by the same donor, via mother-to-child transmission and by grandmother-to-grandchild (breast milk) transmission, respectively (Fig. A). We aimed to study this family trio to define host genetic and virologic factors contributing to slow progression in the two children. All three family members expressed the protective HLA-B*81:01. In each subject autologous virus encoded the same rare escape mutation L188F within the immunodominant HLA-B*81:01-restricted TL9 epitope (Gag 180-188). Methods: L188F, the well-described HLA-B*81:01-associated TL9 escape mutants, most commonly T186S, and selected putative compensatory variants, were introduced by mutagenesis into a modified patient-derived Gag-protease sequence corresponding to Consensus subtype C. These TL9/Gag-Pro variants and the Gag-Pro sequence of the grand-daughter were inserted in the backbone of pNL4-3. These mutant viruses were then generated via electroporation and the viral replication capacity of each determined in comparison with that of NL4-3. Results: L188F or T186S were incapable of replicating at a sufficient level in vitro to measure viral replicative capacity (VRC). Although GFP+ expression in target cells remained negative up to >100 days in culture, viral RNA was successfully extracted from the supernatant and the respective L188F and T186S mutations were confirmed by sequencing. T190I compensated neither L188F nor T186S but Q182S and T190A in combination were able to rescue T186S. The VRC of the grand-daughter’s Gag-Pro chimeric virus, however, was similar to that of WT despite of the presence of L188F. This indicates that one or more compensatory mutants may be present in the grand-daughter virus to correct for the VRC cost of L188F (Fig. B). Conclusions: HLA-B*81:01-restricted L188F escape mutation alone substantially reduced VRC to the point where it could not be quantified. However since the VRC cost of L188F was corrected in the grand-daughter, it is likely that this was the virus transmitted and that low VRC did not contribute to slow progression in these two children. This conclusion is consistent with decent data indicating that HLA class I alleles such as HLA-B*57/58:01/81:01 that are protective in adult infection contribute little to pediatric slow progression.

Poster Abstracts

829 Increasing Adolescent HIV TestingWith a Hybrid Mobile Strategy in Uganda and Kenya Kevin Kadede 1 ;Theodore D. Ruel 2 ; Elizabeth Bukusi 3 ;Tamara Clark 2 ; Edwin Charlebois 2 ; Maya Petersen 4 ; Moses R. Kamya 5 ; DianeV. Havlir 2 ; Gabriel Chamie 2 ; for the Sustainable East Africa Research in Community Health 1 Kenya Med Rsr Inst, Kisumu, Kenya; 2 Univ of California San Francisco, San Francisco, CA, USA; 3 Cntr for Microbiology Rsr, Kenya Med Rsr Inst, Kisumu, Kenya; 4 Univ of California Berkeley, Berkeley, CA, USA; 5 Makerere Univ Coll of Hlth Scis, Kampala, Uganda Background: Ensuring adequate HIV testing coverage among adolescents in sub-Saharan Africa is crucial given the increased HIV risk faced during transition into adulthood. National survey data estimate that <20% of African adolescents know their HIV status. We sought to increase adolescent HIV testing across rural communities in Uganda and Kenya using a hybrid mobile testing approach, and to identify predictors of undiagnosed HIV. Methods: In 2013-14, we enumerated 116,326 adolescent (10-24 years) residents of 32 communities in Uganda (20), and Kenya (12) with a door-to-door census in an HIV “test and treat” trial (SEARCH:NCT01864603). 98,694 adolescents (85%) reported stable residence (living in community ≥6 months). In each community we performed a hybrid testing

346

CROI 2016